1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.560 2009/01/02 16:58:05 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.
95 TK/01 Add preliminary DKIM support. Currently requires a forked version of
96 ALT-N's libdkim that I have put here:
97 http://duncanthrax.net/exim-experimental/
99 Note to Michael Haardt: I had to rename some vars in sieve.c. They
100 were called 'true' and it seems that C99 defines that as a reserved
101 keyword to be used with 'bool' variable types. That means you could
102 not include C99-style headers which use bools without triggering
103 build errors in sieve.c.
105 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
106 as mailq or other aliases. Changed the --help handling significantly
107 to do whats expected. exim_usage() emits usage/help information.
109 SC/01 Added the -bylocaldomain option to eximstats.
111 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
113 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
115 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
121 PH/01 Another patch from the Sieve maintainer.
123 PH/02 When an IPv6 address is converted to a string for single-key lookup
124 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
125 dots are used instead of colons so that keys in lsearch files need not
126 contain colons. This was done some time before quoting was made available
127 in lsearch files. However, iplsearch files do require colons in IPv6 keys
128 (notated using the quote facility) so as to distinguish them from IPv4
129 keys. This meant that lookups for IP addresses in host lists did not work
130 for iplsearch lookups.
132 This has been fixed by arranging for IPv6 addresses to be expressed with
133 colons if the lookup type is iplsearch. This is not incompatible, because
134 previously such lookups could never work.
136 The situation is now rather anomolous, since one *can* have colons in
137 ordinary lsearch keys. However, making the change in all cases is
138 incompatible and would probably break a number of configurations.
140 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
143 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
144 conversion specification without a maximum field width, thereby enabling
145 a rogue spamd server to cause a buffer overflow. While nobody in their
146 right mind would setup Exim to query an untrusted spamd server, an
147 attacker that gains access to a server running spamd could potentially
148 exploit this vulnerability to run arbitrary code as the Exim user.
150 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
151 $primary_hostname instead of what libspf2 thinks the hosts name is.
153 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
154 a directory entry by the name of the lookup key. Previously, if a
155 symlink pointed to a non-existing file or a file in a directory that
156 Exim lacked permissions to read, a lookup for a key matching that
157 symlink would fail. Now it is enough that a matching directory entry
158 exists, symlink or not. (Bugzilla 503.)
160 PH/03 The body_linecount and body_zerocount variables are now exported in the
163 PH/04 Added the $dnslist_matched variable.
165 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
166 This means they are set thereafter only if the connection becomes
169 PH/06 Added the client_condition to authenticators so that some can be skipped
170 by clients under certain conditions.
172 PH/07 The error message for a badly-placed control=no_multiline_responses left
173 "_responses" off the end of the name.
175 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
177 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
178 (without spaces) instead of just copying the configuration text.
180 PH/10 Added the /noupdate option to the ratelimit ACL condition.
182 PH/11 Added $max_received_linelength.
184 PH/12 Added +ignore_defer and +include_defer to host lists.
186 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
187 way in which PCRE > 7.0 is built.
189 PH/14 Implemented queue_only_load_latch.
191 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
192 MAIL command. The effect was to mangle the value on 64-bit systems.
194 PH/16 Another patch from the Sieve maintainer.
196 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
198 PH/18 If a system quota error occurred while trying to create the file for
199 a maildir delivery, the message "Mailbox is full" was not appended to the
200 bounce if the delivery eventually timed out. Change 4.67/27 below applied
201 only to a quota excession during the actual writing of the file.
203 PH/19 It seems that peer DN values may contain newlines (and other non-printing
204 characters?) which causes problems in log lines. The DN values are now
205 passed through string_printing() before being added to log lines.
207 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
208 and InterBase are left for another time.)
210 PH/21 Added message_body_newlines option.
212 PH/22 Guard against possible overflow in moan_check_errorcopy().
214 PH/23 POSIX allows open() to be a macro; guard against that.
216 PH/24 If the recipient of an error message contained an @ in the local part
217 (suitably quoted, of course), incorrect values were put in $domain and
218 $local_part during the evaluation of errors_copy.
224 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
225 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
228 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
229 issue a MAIL command.
231 PH/02 In an ACL statement such as
233 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
235 if a client was not listed at all, or was listed with a value other than
236 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
237 the condition was not true (as it should be), so access was not denied.
238 The bug was that the ! inversion was incorrectly passed on to the second
239 item. This has been fixed.
241 PH/03 Added additional dnslists conditions == and =& which are different from
242 = and & when the dns lookup returns more than one IP address.
244 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
245 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
247 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
248 FSYNC, which compiles an option called disable_fsync that allows for
249 bypassing fsync(). The documentation is heavily laced with warnings.
251 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
253 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
254 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
255 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
256 including adding "make clean"; (3) Added -fPIC when compiling the test
257 dynamically loaded module, to get rid of a warning.
259 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
260 message fails, move_frozen_messages = true and ignore_bounce_errors_after
261 = 0s. The bug is otherwise harmless.
263 PH/07 There was a bug in the dovecot authenticator such that the value of
264 $auth1 could be overwritten, and so not correctly preserved, after a
265 successful authentication. This usually meant that the value preserved by
266 the server_setid option was incorrect.
268 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
270 PH/09 Installed PCRE release 7.0.
272 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
273 run for batched SMTP input. It is now run at the start of every message
274 in the batch. While fixing this I discovered that the process information
275 (output by running exiwhat) was not always getting set for -bs and -bS
276 input. This is fixed, and it now also says "batched" for BSMTP.
278 PH/11 Added control=no_pipelining.
280 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
281 patch, slightly modified), and move the expansion of helo_data till after
282 the connection is made in the smtp transport (so it can use these
285 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
287 PH/14 Added log_selector = +pid.
289 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
291 PH/16 Add ${if forany and ${if forall.
293 PH/17 Added dsn_from option to vary the From: line in DSNs.
295 PH/18 Flush SMTP output before performing a callout, unless control =
296 no_callout_flush is set.
298 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
299 was true (the default) a successful delivery failed to delete the retry
300 item, thus causing premature timeout of the address. The bug is now
303 PH/20 Added hosts_avoid_pipelining to the smtp transport.
305 PH/21 Long custom messages for fakedefer and fakereject are now split up
306 into multiline reponses in the same way that messages for "deny" and
307 other ACL rejections are.
309 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
310 with slight modification.
312 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
313 draft, changing the syntax and factoring some duplicate code".
315 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
316 for deliveries of the second and subsequent messages over the same SMTP
319 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
320 ${reduce, with only minor "tidies".
322 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
324 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
325 expansion side effects.
327 PH/27 When a message times out after an over-quota error from an Exim-imposed
328 quota, the bounce message says "mailbox is full". This message was not
329 being given when it was a system quota that was exceeded. It now should
332 MH/03 Made $recipients available in local_scan(). local_scan() already has
333 better access to the recipient list through recipients_list[], but
334 $recipients can be useful in postmaster-provided expansion strings.
336 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
337 in the case of a MAIL command with additional options following the
338 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
339 were accidentally chopped off.
341 PH/29 SMTP synchronization checks are implemented when a command is read -
342 there is a check that no more input is waiting when there shouldn't be
343 any. However, for some commands, a delay in an ACL can mean that it is
344 some time before the response is written. In this time, more input might
345 arrive, invalidly. So now there are extra checks after an ACL has run for
346 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
347 pipelining has not been advertised.
349 PH/30 MH's patch to allow iscntrl() characters to be list separators.
351 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
352 returned in the SMTP response when smtp_return_error_details was false.
355 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
356 instead of the C I/O that was originally supplied, because problems were
359 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
360 Exim which did not show up earlier: it was assuming that a call to
361 SSL_CTX_set_info_callback() might give an error value. In fact, there is
362 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
363 was a macro that became an assignment, so it seemed to work. This has
364 changed to a proper function call with a void return, hence the compile
365 error. Exim's code has been fixed.
367 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
370 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
372 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
374 PH/37 If a message is not accepted after it has had an id assigned (e.g.
375 because it turns out to be too big or there is a timeout) there is no
376 "Completed" line in the log. When some messages of this type were
377 selected by exigrep, they were listed as "not completed". Others were
378 picked up by some special patterns. I have improved the selection
379 criteria to be more general.
381 PH/38 The host_find_failed option in the manualroute router can now be set
382 to "ignore", to completely ignore a host whose IP address cannot be
383 found. If all hosts are ignored, the behaviour is controlled by the new
384 host_all_ignored option.
386 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
387 homing or because of multiple MX records with /mx) generated more than
388 one IP address, and the following item turned out to be the local host,
389 all the secondary addresses of the first item were incorrectly removed
390 from the list, along with the local host and any following hosts (which
391 is what is supposed to happen).
393 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
394 whoever called Exim into the -H file. In the case of the daemon it was
395 behaving confusingly. When first started, it used values for whoever
396 started the daemon, but after a SIGHUP it used the Exim user (because it
397 calls itself on a restart). I have changed the code so that it now always
400 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
401 message are rejected with the same error (e.g. no authentication or bad
402 sender address), and a DATA command is nevertheless sent (as can happen
403 with PIPELINING or a stupid MUA), the error message that was given to the
404 RCPT commands is included in the rejection of the DATA command. This is
405 intended to be helpful for MUAs that show only the final error to their
408 PH/42 Another patch from the Sieve maintainer.
410 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
411 Eximstats - Fixed some broken HTML links and added missing column headers
413 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
414 columns for Rejects, Temp Rejects, Ham, and Spam rows.
416 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
418 PH/43 Yet another patch from the Sieve maintainer.
420 PH/44 I found a way to check for a TCP/IP connection going away before sending
421 the response to the final '.' that terminates a message, but only in the
422 case where the client has not sent further data following the '.'
423 (unfortunately, this is allowed). However, in many cases there won't be
424 any further data because there won't be any more messages to send. A call
425 to select() can be used: if it shows that the input is "ready", there is
426 either input waiting, or the socket has been closed. An attempt to read
427 the next input character can distinguish the two cases. Previously, Exim
428 would have sent an OK response which the client would never have see.
429 This could lead to message repetition. This fix should cure that, at
430 least in a lot of common cases.
432 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
433 advertised in response to EHLO.
439 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
440 fixed by 4.65/MH/01 (is this a record?) are fixed:
442 (i) An empty string was always treated as zero by the numeric comparison
443 operators. This behaviour has been restored.
445 (ii) It is documented that the numeric comparison operators always treat
446 their arguments as decimal numbers. This was broken in that numbers
447 starting with 0 were being interpreted as octal.
449 While fixing these problems I realized that there was another issue that
450 hadn't been noticed. Values of message_size_limit (both the global option
451 and the transport option) were treated as octal if they started with 0.
452 The documentation was vague. These values are now always treated as
453 decimal, and I will make that clear in the documentation.
459 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
460 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
463 MH/01 Don't check that the operands of numeric comparison operators are
464 integers when their expansion is in "skipping" mode (fixes bug
465 introduced by 4.64-PH/07).
467 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
468 child addresses, Exim now panics and dies. Previously, because the count
469 is held in a short int, deliveries were likely to be lost. As such a
470 large number of recipients for a single message is ridiculous
471 (performance will be very, very poor), I have chosen to impose a limit
472 rather than extend the field.
478 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
479 leftover -K file (the existence of which was triggered by #402).
480 While we were at it, introduced process PID as part of the -K
481 filename. This should rule out race conditions when creating
484 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
485 processing considerably. Previous code took too long for large mails,
486 triggering a timeout which in turn triggers #401.
488 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
489 in the DK code in transports.c. sendfile() is not really portable,
490 hence the _LINUX specificness.
492 TF/01 In the add_headers option to the mail command in an Exim filter,
493 there was a bug that Exim would claim a syntax error in any
494 header after the first one which had an odd number of characters
497 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
498 callout verification, Exim cached a "reject" for the entire domain. This
499 is correct for most verifications, but it is not correct for a recipient
500 verification with use_sender or use_postmaster set, because in that case
501 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
502 case of MAIL FROM:<> rejection from other early rejections (e.g.
503 rejection of HELO). When verifying a recipient using a non-null MAIL
504 address, the cache is ignored if it shows MAIL FROM:<> rejection.
505 Whatever the result of the callout, the value of the domain cache is
506 left unchanged (for any other kind of callout, getting as far as trying
507 RCPT means that the domain itself is ok).
509 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
512 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
513 manifest itself as EPIPE rather than ECONNECT. When tidying away a
514 session, the daemon ignores ECONNECT errors and logs others; it now
515 ignores EPIPE as well.
517 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
518 (quoted-printable decoding).
520 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
521 later the small subsequent patch to fix an introduced bug.
523 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
525 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
527 PH/08 An error is now given if message_size_limit is specified negative.
529 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
530 to be given (somewhat) arbitrary names.
532 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
535 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
536 miscellaneous code fixes
538 PH/10 Added the log_reject_target ACL modifier to specify where to log
541 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
542 hostname. This is wrong, because it relates to the incoming message (and
543 probably the interface on which it is arriving) and not to the outgoing
544 callout (which could be using a different interface). This has been
545 changed to use the value of the helo_data option from the smtp transport
546 instead - this is what is used when a message is actually being sent. If
547 there is no remote transport (possible with a router that sets up host
548 addresses), $smtp_active_hostname is used.
550 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
551 tweaks were necessary in order to get it to work (see also 21 below):
552 (a) The code assumed that strncpy() returns a negative number on buffer
553 overflow, which isn't the case. Replaced with Exim's string_format()
555 (b) There were several signed/unsigned issues. I just did the minimum
556 hacking in of casts. There is scope for a larger refactoring.
557 (c) The code used strcasecmp() which is not a standard C function.
558 Replaced with Exim's strcmpic() function.
559 (d) The code set only $1; it now sets $auth1 as well.
560 (e) A simple test gave the error "authentication client didn't specify
561 service in request". It would seem that Dovecot has changed its
562 interface. Fortunately there's a specification; I followed it and
563 changed what the client sends and it appears to be working now.
565 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
568 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
569 address A is aliased to B and C, where B exists and C does not. Without
570 -v the output is "A verified" because verification stops after a
571 successful redirection if more than one address is generated. However,
572 with -v the child addresses are also verified. Exim was outputting "A
573 failed to verify" and then showing the successful verification for C,
574 with its parentage. It now outputs "B failed to verify", showing B's
575 parentage before showing the successful verification of C.
577 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
578 look up a TXT record in a specific list after matching in a combined
581 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
582 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
583 they consult the DNS. I had assumed they would set it the way they
584 wanted; and indeed my experiments on Linux seem to show that in some
585 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
586 To be on the safe side, however, I have now made the interface to
587 host_find_byname() similar to host_find_bydns(), with an argument
588 containing the DNS resolver options. The host_find_byname() function now
589 sets these options at its start, just as host_find_bydns() does. The smtp
590 transport options dns_qualify_single and dns_search_parents are passed to
591 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
592 of host_find_byname() use the default settings of RES_DEFNAMES
593 (qualify_single) but not RES_DNSRCH (search_parents).
595 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
596 spool_read_header() do less string testing, by means of a preliminary
597 switch on the second character of optional "-foo" lines. (This is
598 overdue, caused by the large number of possibilities that now exist.
599 Originally there were few.) While I was there, I also converted the
600 str(n)cmp tests so they don't re-test the leading "-" and the first
601 character, in the hope this might squeeze out yet more improvement.
603 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
604 flag allowing group syntax was set by the header_syntax check but not
605 turned off, possible causing trouble later; (2) The flag was not being
606 set at all for the header_verify test, causing "group"-style headers to
607 be rejected. I have now set it in this case, and also caused header_
608 verify to ignore an empty address taken from a group. While doing this, I
609 came across some other cases where the code for allowing group syntax
610 while scanning a header line wasn't quite right (mostly, not resetting
611 the flag correctly in the right place). These bugs could have caused
612 trouble for malformed header lines. I hope it is now all correct.
614 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
615 with the "reply" argument non-NULL. The code, however (which originally
616 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
617 but it didn't always do it. This confused somebody who was copying the
618 code for some other use. I have removed all the tests.
620 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
621 feature that was used to support insecure browsers during the U.S. crypto
622 embargo. It requires special client support, and Exim is probably the
623 only MTA that supported it -- and would never use it because real RSA is
624 always available. This code has been removed, because it had the bad
625 effect of slowing Exim down by computing (never used) parameters for the
626 RSA_EXPORT functionality.
628 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
629 authenticator to fail if there's a tab character in the incoming data
630 (there should never be unless someone is messing about, as it's supposed
631 to be base64-encoded). Also added, on Timo's advice, the "secured" option
632 if the connection is using TLS or if the remote IP is the same as the
633 local IP, and the "valid-client-cert option" if a client certificate has
636 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
637 authenticators. This can be used for authorization after authentication
638 succeeds. (In the case of plaintext, it servers for both authentication
641 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
642 if any retry times were supplied.
644 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
645 connection, where there is no client IP address to check. In this
646 situation, the verify now always succeeds.
648 PH/25 Applied John Jetmore's -Mset patch.
650 PH/26 Added -bem to be like -Mset, but loading a message from a file.
652 PH/27 In a string expansion for a processed (not raw) header when multiple
653 headers of the same name were present, leading whitespace was being
654 removed from all of them, but trailing whitespace was being removed only
655 from the last one. Now trailing whitespace is removed from each header
656 before concatenation. Completely empty headers in a concatenation (as
659 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
660 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
662 PH/29 [Removed. This was a change that I later backed out, and forgot to
663 correct the ChangeLog entry (that I had efficiently created) before
664 committing the later change.]
666 PH/30 Exim was sometimes attempting to deliver messages that had suffered
667 address errors (4xx response to RCPT) over the same connection as other
668 messages routed to the same hosts. Such deliveries are always "forced",
669 so retry times are not inspected. This resulted in far too many retries
670 for the affected addresses. The effect occurred only when there were more
671 hosts than the hosts_max_try setting in the smtp transport when it had
672 the 4xx errors. Those hosts that it had tried were not added to the list
673 of hosts for which the message was waiting, so if all were tried, there
674 was no problem. Two fixes have been applied:
676 (i) If there are any address or message errors in an SMTP delivery, none
677 of the hosts (tried or untried) are now added to the list of hosts
678 for which the message is waiting, so the message should not be a
679 candidate for sending over the same connection that was used for a
680 successful delivery of some other message. This seems entirely
681 reasonable: after all the message is NOT "waiting for some host".
682 This is so "obvious" that I'm not sure why it wasn't done
683 previously. Hope I haven't missed anything, but it can't do any
684 harm, as the worst effect is to miss an optimization.
686 (ii) If, despite (i), such a delivery is accidentally attempted, the
687 routing retry time is respected, so at least it doesn't keep
688 hammering the server.
690 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
691 in ${readsocket because some servers need this prod.
693 PH/32 Added some extra debug output when updating a wait-xxx database.
695 PH/33 The hint "could be header name not terminated by colon", which has been
696 given for certain expansion errors for a long time, was not being given
697 for the ${if def:h_colon_omitted{... case.
699 PH/34 The spec says: "With one important exception, whenever a domain list is
700 being scanned, $domain contains the subject domain." There was at least
701 one case where this was not true.
703 PH/35 The error "getsockname() failed: connection reset by peer" was being
704 written to the panic log as well as the main log, but it isn't really
705 panic-worthy as it just means the connection died rather early on. I have
706 removed the panic log writing for the ECONNRESET error when getsockname()
709 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
710 runs only) independently of the message's sender address. This meant
711 that, if the 4xx error was in fact related to the sender, a different
712 message to the same recipient with a different sender could confuse
713 things. In particualar, this can happen when sending to a greylisting
714 server, but other circumstances could also provoke similar problems.
715 I have changed the default so that the retry time for these errors is now
716 based a combination of the sender and recipient addresses. This change
717 can be overridden by setting address_retry_include_sender=false in the
720 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
721 remote server are returned as part of bounce messages. This was not
722 happening for LMTP over a pipe (the lmtp transport), but now it is the
723 same for both kinds of LMTP.
725 PH/38 Despite being documented as not happening, Exim was rewriting addresses
726 in header lines that were in fact CNAMEs. This is no longer the case.
728 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
729 and queue runs started by the daemon processed all messages. This has
730 been fixed so that -R and -S can now usefully be given with -q<time>.
732 PH/40 Import PCRE release 6.7 (fixes some bugs).
734 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
736 PH/42 Give an error if -q is specified more than once.
738 PH/43 Renamed the variables $interface_address and $interface_port as
739 $received_ip_address and $received_port, to make it clear that these
740 values apply to message reception, and not to the outgoing interface when
741 a message is delivered. (The old names remain recognized, of course.)
743 PH/44 There was no timeout on the connect() call when using a Unix domain
744 socket in the ${readsocket expansion. There now is.
746 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
747 be meaningful with "accept".
749 SC/01 Eximstats V1.43
750 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
752 SC/02 Eximstats V1.44
753 Use a glob alias rather than an array ref in the generated
754 parser. This improves both readability and performance.
756 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
757 Collect SpamAssassin and rejection statistics.
758 Don't display local sender or destination tables unless
759 there is data to show.
760 Added average volumes into the top table text output.
762 SC/04 Eximstats V1.46
763 Collect data on the number of addresses (recipients)
764 as well as the number of messages.
766 SC/05 Eximstats V1.47
767 Added 'Message too big' to the list of mail rejection
768 reasons (thanks to Marco Gaiarin).
770 SC/06 Eximstats V1.48
771 Mainlog lines which have GMT offsets and are too short to
772 have a flag are now skipped.
774 SC/07 Eximstats V1.49 (Alain Williams)
775 Added the -emptyok flag.
777 SC/08 Eximstats V1.50
778 Fixes for obtaining the IP address from reject messages.
780 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
781 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
782 whitesspace changes from 4.64-PH/27
784 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
787 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
788 are found, allow negative numbers in numeric criteria)
790 JJ/06 exipick.20061117.2, added new $message_body_missing variable
792 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
793 to match changes made in 4.64-PH/43
795 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
797 PH/47 Put in an explicit test for a DNS lookup of an address record where the
798 "domain" is actually an IP address, and force a failure. This locks out
799 those revolvers/nameservers that support "A-for-A" lookups, in
800 contravention of the specifications.
802 PH/48 When a host name was looked up from an IP address, and the subsequent
803 forward lookup of the name timed out, the host name was left in
804 $sender_host_name, contrary to the specification.
806 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
807 restricted to single-key lookups, Exim was not diagnosing an error if
808 * or *@ was used with a query-style lookup.
810 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
812 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
813 long ago, but noone interested enough thought of it. Let's just say that
814 the "1.1" means that there are some new functions that weren't there at
815 some point in the past.
817 PH/51 Error processing for expansion failure of helo_data from an smtp
818 transport during callout processing was broken.
820 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
821 tested/used via the -bh/-bhc/-bs options.
823 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
824 bug, fixed in subsequent PCRE releases).
826 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
827 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
829 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
835 SC/01 Use a glob alias rather than an array ref in eximstats generated
836 parser. This improves both readability and performance.
838 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
839 Don't display local sender or destination tables in eximstats unless
840 there is data to show.
841 Added average volumes into the eximstats top table text output.
843 SC/03 Collect data on the number of addresses (recipients) as well
844 as the number of messages in eximstats.
846 TF/01 Correct an error in the documentation for the redirect router. Exim
847 does (usually) call initgroups() when daemonizing.
849 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
850 with consistent privilege compared to when running as a daemon.
852 TF/03 Note in the spec that $authenticated_id is not set for local
853 submissions from trusted users.
855 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
856 Thanks to Dean Brooks <dean@iglou.com> for the patch.
858 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
859 by adding some example configuration directives to the default
860 configuration file. A little bit of work is required to uncomment the
861 directives and define how usernames and passwords are checked, but
862 there is now a framework to start from.
864 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
865 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
866 without this. I don't know how relevant this is to other LDAP libraries.
868 PH/02 Add the verb name to the "unknown ACL verb" error.
870 PH/03 Magnus Holmgren's patch for filter_prepend_home.
872 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
874 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
875 directory not expanded when it should be if an expanded home directory
876 was set for the address (which is overridden by the transport).
878 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
881 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
882 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
883 because it is too late at that time, and has no effect.
885 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
886 security issue with \' (bugzilla #107). I could not use the
887 PQescapeStringConn() function, because it needs a PGconn value as one of
890 PH/08 When testing addresses using -bt, indicate those final addresses that
891 are duplicates that would not cause an additional delivery. At least one
892 person was confused, thinking that -bt output corresponded to deliveries.
893 (Suppressing duplicates isn't a good idea as you lose the information
894 about possibly different redirections that led to the duplicates.)
896 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
897 systems where poll() doesn't work, in particular OS X.
899 PH/10 Added more information to debugging output for retry time not reached.
901 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
902 operations in malware.c.
904 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
907 PH/13 If write_rejectlog was set false when logging was sent to syslog with
908 syslog_duplication set false, log lines that would normally be written
909 both the the main log and to the reject log were not written to syslog at
912 PH/14 In the default configuration, change the use of "message" in ACL warn
913 statements to "add_header".
915 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
916 not followed by a command (e.g. "seen endif").
918 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
919 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
922 PH/17 Added extra conditions to the default value of delay_warning_condition
926 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
927 { match{$h_precedence:}{(?i)bulk|list|junk} } \
928 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
931 The Auto-Submitted: and various List- headers are standardised, whereas I
932 don't think Precedence: ever was.
934 PH/18 Refactored debugging code in route_finduser() to show more information,
935 in particular, the error code if getpwnam() issues one.
937 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
938 This is apparently needed in addition to the PH/07 change above to avoid
939 any possible encoding problems.
941 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
942 but not after initializing Perl.
944 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
945 output them only if debugging. By default they are written stderr,
946 apparently, which is not desirable.
948 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
951 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
954 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
956 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
957 authenticated or an ident call has been made. Suppress the default
958 values for $authenticated_id and $authenticated_sender (but permit -oMai
959 and -oMas) when testing with -bh.
961 PH/24 Re-jigged the order of the tests in the default configuration so that the
962 tests for valid domains and recipients precede the DNS black list and CSA
963 tests, on the grounds that those ones are more expensive.
965 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
966 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
967 This bug exists in every version of Exim that I still have, right back to
970 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
971 However, an attempt to turn on case-sensitivity in a regex key by
972 including (?-i) didn't work because the subject string was already
973 lowercased, and the effects were non-intuitive. It turns out that a
974 one-line patch can be used to allow (?-i) to work as expected.
980 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
981 other effects) broke the use of negated acl sub-conditions.
983 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
986 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
987 "Deny" causes Exim to reject the incoming connection with a 554 error.
988 Unfortunately, if there is a major crisis, such as a disk failure,
989 tcp-wrappers gives "deny", whereas what one would like would be some
990 kind of temporary error. A kludge has been added to help with this.
991 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
992 554 error is used if errno is still zero or contains ENOENT (which occurs
993 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
996 PH/03 Add -lutil to the default FreeBSD LIBS setting.
998 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
999 errors. Otherwise a message that provokes a temporary error (when other
1000 messages do not) can cause a whole host to time out.
1002 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1003 addresses were routed directly to files or pipes from a redirect router.
1004 File deliveries just didn't batch; pipe deliveries might have suffered
1007 PH/06 A failure to get a lock for a hints database would erroneously always say
1008 "Failed to get write lock", even when it was really a read lock.
1010 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1011 of 0600. This has been changed to use the value of the lockfile_mode
1012 option (which defaults to 0600).
1014 PH/08 Applied small patch from the Sieve maintainer.
1016 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1017 folder from quota calculations, a direct delivery into this folder messed
1018 up the contents of the maildirsize file. This was because the regex was
1019 used only to exclude .Trash (or whatever) when the size of the mailbox
1020 was calculated. There was no check that a delivery was happening into an
1021 excluded directory. This bug has been fixed by ignoring all quota
1022 processing for deliveries into excluded directories.
1024 PH/10 Added the maildirfolder_create_regex option to appendfile.
1030 PH/01 The code for finding all the local interface addresses on a FreeBSD
1031 system running IPv6 was broken. This may well have applied to all BSD
1032 systems, as well as to others that have similar system calls. The broken
1033 code found IPv4 interfaces correctly, but gave incorrect values for the
1034 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1035 that it would not match correctly against @[] and not recognize the IPv6
1038 PH/02 The ipliteral router was not recognizing addresses of the form user@
1039 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1041 PH/03 Added disable_ipv6.
1043 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1044 decoded form, because it is most often used to construct To: headers
1045 lines in autoreplies, and the decoded form may well be syntactically
1046 invalid. However, $reply_address has leading white space removed, and all
1047 newlines turned into spaces so that the autoreply transport does not
1050 PH/05 If group was specified without a user on a router, and no group or user
1051 was specified on a transport, the group from the router was ignored.
1053 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1054 for visible compile-time settings that can be used to change these
1055 numbers, for those that want even more. Backwards compatibility with old
1056 spool files has been maintained. However, going back to a previous Exim
1057 release will lost any variables that are in spool files.
1059 PH/07 Two small changes when running in the test harness: increase delay when
1060 passing a TCP/IP connection to a new process, in case the original
1061 process has to generate a bounce, and remove special handling of
1062 127.0.0.2 (sic), which is no longer necessary.
1064 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1065 be the same on different OS.
1067 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1070 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1071 whether --show-vars was specified or not
1073 JJ/02 exipick: Added support for new ACL variable spool format introduced
1076 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1077 syntactically invalid From: or Reply-to: line, and a filter used this to
1078 generate an autoreply, and therefore failed to obtain an address for the
1079 autoreply, Exim could try to deliver to a non-existent relative file
1080 name, causing unrelated and misleading errors. What now happens is that
1081 it logs this as a hard delivery error, but does not attempt to create a
1084 PH/11 The exinext utility has a -C option for testing purposes, but although
1085 the given file was scanned by exinext itself; it wasn't being passed on
1086 when Exim was called.
1088 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1089 an end-of-file indication when reading a command response.
1091 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1092 compiled. In many other places in Exim, IPv6 addresses are always
1093 recognized, so I have changed this. It also means that IPv4 domain
1094 literals of the form [IPV4:n.n.n.n] are now always recognized.
1096 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1097 used if the router is not running as root, for example, when verifying at
1098 ACL time, or when using -bh. The debugging output from this situation was
1099 non-existent - all you got was a failure to exec. I have made two
1102 (a) Failures to set uid/gid, the current directory, or a process leader
1103 in a subprocess such as that created by queryprogram now generate
1104 suitable debugging ouput when -d is set.
1106 (b) The queryprogram router detects when it is not running as root,
1107 outputs suitable debugging information if -d is set, and then runs
1108 the subprocess without attempting to change uid/gid.
1110 PH/15 Minor change to Makefile for building test_host (undocumented testing
1113 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1114 additional section of a DNS packet that returns MX or SRV records.
1115 Instead, it always explicitly searches for A/AAAA records. This avoids
1116 major problems that occur when a DNS server includes only records of one
1117 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1118 fixed another bug: if SRV records were looked up and the corresponding
1119 address records were *not* found in the additional section, the port
1120 values from the SRV records were lost.
1122 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1123 using the correct key (the original address) when searching the retry
1124 rules in order to find which one to use for generating the retry hint.
1126 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1127 adding the default one. Similarly, if it contains a Reply-To: header, the
1128 errors_reply_to option, if set, is not used.
1130 PH/19 When calculating a retry time, Exim used to measure the "time since
1131 failure" by looking at the "first failed" field in the retry record. Now
1132 it does not use this if it is later than than the arrival time of the
1133 message. Instead it uses the arrival time. This makes for better
1134 behaviour in cases where some deliveries succeed, thus re-setting the
1135 "first failed" field. An example is a quota failure for a huge message
1136 when small messages continue to be delivered. Without this change, the
1137 "time since failure" will always be short, possible causing more frequent
1138 delivery attempts for the huge message than are intended.
1139 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1141 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1142 $1, $2, $3) because the numerical variables can be reset during some
1143 expansion items (e.g. "match"), thereby losing the authentication data.
1145 PH/21 Make -bV show the size of off_t variables so that the test suite can
1146 decide whether to run tests for quotas > 2G.
1148 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1149 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1150 filecount value is greater than 2G or if a quota value is greater than 2G
1151 on a system where the size of off_t is not greater than 4, a panic error
1154 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1155 never match. The debug and -bh output now contains an explicit error
1156 message indicating a malformed IPv4 address or mask.
1158 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1159 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1160 PH/23 above applies.
1162 PH/25 Do not write to syslog when running in the test harness. The only
1163 occasion when this arises is a failure to open the main or panic logs
1164 (for which there is an explicit test).
1166 PH/26 Added the /no_tell option to "control=freeze".
1168 PH/27 If a host name lookup failed very early in a connection, for example, if
1169 the IP address matched host_lookup and the reverse lookup yielded a name
1170 that did not have a forward lookup, an error message of the form "no IP
1171 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1172 could be logged. Now it outputs the IP address instead of "NULL".
1174 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1175 allows the sender and the authenticated sender to be set when
1176 submitting a message from within Exim. Since child_open_exim() is
1177 documented for local_scan(), the new function should be too.
1179 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1180 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1181 results in an empty string is now treated as unset.
1183 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1185 PH/31 Added sender_verify_fail logging option.
1187 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1188 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1189 tidied the source and removed it altogether.
1191 PH/33 When a queue run was abandoned because the load average was too high, a
1192 log line was always written; now it is written only if the queue_run log
1193 selector is set. In addition, the log line for abandonment now contains
1194 information about the queue run such as the pid. This is always present
1195 in "start" and "stop" lines but was omitted from the "abandon" line.
1197 PH/34 Omit spaces between a header name and the colon in the error message that
1198 is given when verify = headers_syntax fails (if there are lots of them,
1199 the message gets confusing).
1201 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1202 names, as there are now some PTR records that contain slashes. This check
1203 is only to protect against broken name servers that fall over on strange
1204 characters, so the fact that it applies to all lookups doesn't matter.
1206 PH/36 Now that the new test suite is complete, we can remove some of the
1207 special code in Exim that was needed for the old test suite. For example,
1208 sorting DNS records because real resolvers return them in an arbitrary
1209 order. The new test suite's fake resolver always returns records in the
1212 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1213 bounces) except when queue_only is set, to avoid logging races between
1214 the different processes.
1216 PH/38 Panic-die if .include specifies a non-absolute path.
1218 PH/39 A tweak to the "H" retry rule from its user.
1220 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1221 a label. They prevented compilation on older perls.
1223 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1224 a warning to be raised on newish perls.
1226 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1227 on queue. Changes to match documented behaviour of showing count of
1228 messages matching specified criteria.
1230 PH/40 Changed the default ident timeout from 30s to 5s.
1232 PH/41 Added support for the use of login_cap features, on those BSD systems
1233 that have them, for controlling the resources used by pipe deliveries.
1235 PH/42 The content-scanning code uses fopen() to create files in which to put
1236 message data. Previously it was not paying any attention to the mode of
1237 the files. Exim runs with umask(0) because the rest of the code creates
1238 files with open(), and sets the required mode explicitly. Thus, these
1239 files were ending up world-writeable. This was not a big issue, because,
1240 being within the spool directory, they were not world-accessible. I have
1241 created a function called modefopen, which takes an additional mode
1242 argument. It sets umask(777), creates the file, chmods it to the required
1243 mode, then resets the umask. All the relevant calls to fopen() in the
1244 content scanning code have been changed to use this function.
1246 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1247 to 24 hours. This avoids potential overflow problems when processing G
1248 and H retry rules. I suspect nobody ever tinkers with this value.
1250 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1252 PH/45 When the plaintext authenticator is running as a client, the server's
1253 challenges are checked to ensure they are valid base64 strings. By
1254 default, the authentication attempt is cancelled if an invalid string is
1255 received. Setting client_ignore_invalid_base64 true ignores these errors.
1256 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1257 they are received. Thus, the responses can be made to depend on the
1258 challenges. If an invalid string is ignored, an empty string is placed in
1261 PH/46 Messages that are created by the autoreply transport now contains a
1262 References: header, in accordance with RFCs 2822 and 3834.
1264 PH/47 Added authenticated_sender_force to the smtp transport.
1266 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1268 PH/49 Installed latest patch from the Sieve maintainer.
1270 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1271 was also set, the appendfile transport was unnecessarily scanning a
1272 directory of message files (e.g. for maildir delivery) to find the count
1273 of files (along with the size), even though it did not need this
1274 information. It now does the scan only if it needs to find either the
1275 size of the count of files.
1277 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1279 PH/52 Two bugs concerned with error handling when the smtp transport is
1282 (i) Exim was not creating retry information for temporary errors given
1283 for individual recipients after the DATA command when the smtp transport
1284 was used in LMTP mode. This meant that they could be retried too
1285 frequently, and not timed out correctly.
1287 (ii) Exim was setting the flag that allows error details to be returned
1288 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1289 recipients that were returned after the DATA command.
1291 PH/53 This is related to PH/52, but is more general: for any failing address,
1292 when detailed error information was permitted to be returned to the
1293 sender, but the error was temporary, then after the final timeout, only
1294 "retry timeout exceeded" was returned. Now it returns the full error as
1295 well as "retry timeout exceeded".
1297 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1298 do this, and (what is worse) MTAs that accept it.
1300 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1301 will now be deprecated.
1303 PH/56 New os.c-cygwin from the Cygwin maintainer.
1305 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1306 formats (previously only available in exim formats via -bpr, -bpru,
1307 and -bpra. Now also available in native and exiqgrep formats)
1309 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1310 with very large, slow to parse queues
1312 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1314 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1316 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1317 responses to authentication challenges, though it was showing the
1318 challenges; (ii) I've removed the CR characters from the debug output for
1321 PH/58 Allow for the insertion of a newline as well as a space when a string
1322 is turned into more than one encoded-word during RFC 2047 encoding. The
1323 Sieve code now uses this.
1325 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1326 data_4xx, lost_connection, tls_required.
1328 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1329 message was being sent as an SMTP response.
1331 PH/61 Add -l and -k options to exicyclog.
1333 PH/62 When verifying, if an address was redirected to one new address, so that
1334 verification continued, and the new address failed or deferred after
1335 having set something in $address_data, the value of $address_data was not
1336 passed back to the ACL. This was different to the case when no
1337 redirection occurred. The value is now passed back in both cases.
1339 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1340 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1341 use login_cap.h, so on its own it isn't the distinguishing feature. The
1342 new name refers directly to the setclassresources() function.
1344 PH/65 Added configuration files for NetBSD3.
1346 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1348 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1349 is preferred over IPv4.
1351 PH/68 The bounce_return_message and bounce_return_body options were not being
1352 honoured for bounces generated during the reception of non-SMTP messages.
1353 In particular, this applied to messages rejected by the ACL. This bug has
1354 been fixed. However, if bounce_return_message is true and bounce_return_
1355 body is false, the headers that are returned for a non-SMTP message
1356 include only those that have been read before the error was detected.
1357 (In the case of an ACL rejection, they have all been read.)
1359 PH/69 The HTML version of the specification is now built in a directory called
1360 spec_html instead of spec.html, because the latter looks like a path with
1361 a MIME-type, and this confuses some software.
1363 PH/70 Catch two compiler warnings in sieve.c.
1365 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1366 function verify_get_ident() calls ip_connect() to connect a socket, but
1367 if the "connect()" function timed out, ip_connect() used to close the
1368 socket. However, verify_get_ident() also closes the socket later, and in
1369 between Exim writes to the log, which may get opened at this point. When
1370 the socket was closed in ip_connect(), the log could get the same file
1371 descriptor number as the socket. This naturally causes chaos. The fix is
1372 not to close the socket in ip_connect(); the socket should be closed by
1373 the function that creates it. There was only one place in the code where
1374 this was missing, in the iplookup router, which I don't think anybody now
1375 uses, but I've fixed it anyway.
1377 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1378 well as to direct DNS lookups. Otherwise the handling of names in host
1379 lists is inconsistent and therefore confusing.
1385 PH/01 Two changes to the default runtime configuration:
1387 (1) Move the checks for relay_from_hosts and authenticated clients from
1388 after to before the (commented out) DNS black list checks.
1390 (2) Add control=submission to the relay_from_hosts and authenticated
1391 clients checks, on the grounds that messages accepted by these
1392 statements are most likely to be submissions.
1394 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1396 (1) Generate an error if the third argument for the ${prvs expansion is
1399 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1402 (3) Reset the variables that are obtained from the first argument of
1403 ${prvscheck and used in the second argument before leaving the code,
1404 because their memory is reclaimed, so using them afterwards may do
1407 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1408 one (it's much easier than Tom thought :-).
1410 (5) Because of (4), we can now allow for the use of $prvscheck_result
1411 inside the third argument.
1413 PH/03 For some reason, the default setting of PATH when running a command from
1414 a pipe transport was just "/usr/bin". I have changed it to
1417 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1418 anything to be listed in the output from -bV.
1420 PH/05 When a filter generated an autoreply, the entire To: header line was
1421 quoted in the delivery log line, like this:
1423 => >A.N.Other <ano@some.domain> <original@ddress> ...
1425 This has been changed so that it extracts the operative address. There
1426 may be more than one such address. If so, they are comma-separated, like
1429 => >ano@some.domain,ona@other.domain <original@ddress> ...
1431 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1432 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1433 not being looked up in the rDNS to get a host name, Exim was showing the
1434 IP address twice in Received: lines, even though the IP addresses were
1435 identical. For example:
1437 Received: from [1.2.3.4] (helo=[1.2.3.4])
1439 However, if the real host name was known, it was omitting the HELO data
1440 if it matched the actual IP address. This has been tidied up so that it
1441 doesn't show the same IP address twice.
1443 PH/07 When both +timestamp and +memory debugging was on, the value given by
1444 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1445 called by the debug printing, thereby overwriting the timestamp buffer.
1446 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1448 PH/08 When the original message was included in an autoreply transport, it
1449 always said "this is a copy of the message, including all the headers",
1450 even if body_only or headers_only was set. It now gives an appropriate
1453 PH/09 Applied a patch from the Sieve maintainer which:
1455 o fixes some comments
1456 o adds the (disabled) notify extension core
1457 o adds some debug output for the result of if/elsif tests
1458 o points to the current vacation draft in the documentation
1459 and documents the missing references header update
1463 o fixes a bug in processing the envelope test (when testing
1464 multiple envelope elements, the last element determinted the
1467 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1468 Electronic Mail") by including:
1470 Auto-submitted: auto-generated
1472 in the messages that it generates (bounce messages and others, such as
1473 warnings). In the case of bounce messages for non-SMTP mesages, there was
1474 also a typo: it was using "Auto_submitted" (underscore instead of
1475 hyphen). Since every message generated by Exim is necessarily in response
1476 to another message, thes have all been changed to:
1478 Auto-Submitted: auto-replied
1480 in accordance with these statements in the RFC:
1482 The auto-replied keyword:
1484 - SHOULD be used on messages sent in direct response to another
1485 message by an automatic process,
1487 - MUST NOT be used on manually-generated messages,
1489 - MAY be used on Delivery Status Notifications (DSNs) and Message
1490 Disposition Notifications (MDNs),
1492 - MUST NOT be used on messages generated by automatic or periodic
1493 processes, except for messages which are automatic responses to
1496 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1497 to the default Received: header definition.
1499 PH/12 Added log selector acl_warn_skipped (default on).
1501 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1502 variables because (a) they are in the wrong storage pool and (b) even if
1503 they were copied, it wouldn't work properly because of the caching.
1505 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1506 checking when decoding. Apparently there are clients that generate
1507 overlong encoded strings. Why am I not surprised?
1509 PH/15 If the first argument of "${if match_address" was not empty, but did not
1510 contain an "@" character, Exim crashed. Now it writes a panic log message
1511 and treats the condition as false.
1513 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1515 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1516 extension "envelope-auth". The code is finished and in agreement with
1517 other implementations, but there is no documentation so far and in fact,
1518 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1519 not changing the active code.
1521 Print executed "if" and "elsif" statements when debugging is used. This
1522 helps a great deal to understand what a filter does.
1524 Document more things not specified clearly in RFC3028. I had all this
1525 sorted out, when out of a sudden new issues came to my mind. Oops."
1527 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1530 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1531 canonical form (as documented). However, after a host name lookup from
1532 the IP address, check_host() was doing a simple string comparison with
1533 addresses acquired from the DNS when checking that the found name did
1534 have the original IP as one of its addresses. Since any found IPv6
1535 addresses are likely to be in abbreviated form, the comparison could
1536 fail. Luckily, there already exists a function for doing the comparison
1537 by converting both addresses to binary, so now that is used instead of
1538 the text comparison.
1540 PH/20 There was another similar case to PH/19, when a complete host name was
1541 given in a host list; looking up its IP address could give an abbreviated
1542 form, whereas the current host's name might or might not be abbreviated.
1543 The same fix has been applied.
1549 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1550 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1553 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1554 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1556 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1558 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1559 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1560 Also, for "Auto-Submitted", treat anything other than "no" as
1561 non-personal, in accordance with RFC 3834. (Previously it treated
1562 anything starting "auto-" as non-personal.)
1564 TF/01 The control=submission/name=... option had a problem with syntax
1565 errors if the name included a slash character. The /name= option
1566 now slurps the rest of the string, so it can include any characters
1567 but it must come last in the list of options (after /sender_retain
1570 PH/05 Some modifications to the interface to the fake nameserver for the new
1578 TK/01 Added the "success_on_redirect" address verification option. See
1579 NewStuff for rationale and an example.
1581 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1583 PH/02 Patch to exigrep to allow it to work on syslog lines.
1585 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1586 fread() to skip over the body file's header line, because in Cygwin the
1587 header line is locked and is inaccessible.
1589 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1590 co-exist for some time) to make it clear that it is the Exim ID that is
1591 referenced, not the Message-ID: header line.
1593 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1594 string_format() function, because snprintf() does not exist on all
1597 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1598 ${stat: expansion item.
1600 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1601 protocol synchronization error", to keep the pedants happy.
1603 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1604 well as for IRIX systems, when gcc is being used. See the host.c source
1607 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1609 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1612 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1613 transport and to the smtp transport in LMTP mode.
1615 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1617 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1618 run a filter in a subprocess. This could lead to confusion in subsequent
1619 lookups in the parent process. There should also be a search_tidyup() at
1620 the end of the subprocess.
1622 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1623 only if the host matched helo_try_verify_hosts, which caused the
1624 verification to occur when the EHLO/HELO command was issued. The ACL just
1625 tested the remembered result. Now, if a previous verification attempt has
1626 not happened, "verify = helo" does it there and then.
1628 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1630 TK/03 Fix log output including CR from clamd.
1632 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1633 exist provoked a memory error which could cause a segfault.
1635 PH/15 Installed PCRE 6.2
1637 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1639 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1640 of the problem. Specifically, suggested +O2 rather than +O1 for the
1643 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1645 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1646 record, the port was not correctly logged when the outgoing_port log
1647 selector was set (it logged the transort's default port).
1649 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1650 fallback_hosts, and "hosts" in the smtp transport.
1652 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1653 host errors such as "Connection refused".
1655 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1656 authentication with radiusclient 0.4.9:
1658 - Error returned from rc_read_config was caught wrongly
1659 - Username/password not passed on to radius server due to wrong length.
1661 The presumption is that some radiusclient API changes for 4.51/PH/17
1662 were not taken care of correctly. The code is still untested by me (my
1663 Linux distribution still has 0.3.2 of radiusclient), but it was
1664 contributed by a Radius user.
1666 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1667 expanding the "port" option of the smtp transport.
1669 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1670 while reading a MIME header. Thanks to Tom Hughes for a patch.
1672 PH/24 Include config.h inside local_scan.h so that configuration settings are
1675 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1676 that in an ACL for RCPT (for example), you can examine exactly what was
1679 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1680 commands, but it was not correctly comparing the address with the actual
1681 client host address. Thus, it would show the EHLO address in Received:
1682 header lines when this was not necessary.
1684 PH/27 Added the % operator to ${eval:}.
1686 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1687 it should be ignoring failures (because with -C, for example, it has lost
1688 privilege). It wasn't ignoring creation failures other than "already
1691 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1694 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1695 because an input line was too long, either on its own, or by virtue of
1696 too many continuations, the temporary file was not being removed, and the
1697 return code was incorrect.
1699 PH/31 Missing "BOOL" in function definition in filtertest.c.
1701 PH/32 Applied Sieve patches from the maintainer.
1703 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1705 PH/33 Added "verify = not_blind".
1707 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1708 Local/Makefile (with some defaults set). These are used in built scripts
1709 such as exicyclog, but they have never been used in the exim_install
1710 script (though there are many overriding facilities there). I have
1711 arranged that the exim_install script now takes note of these two
1714 PH/35 Installed configuration files for Dragonfly.
1716 PH/36 When a locally submitted message by a trusted user did not contain a
1717 From: header, and the sender address was obtained from -f or from an SMTP
1718 MAIL command, and the trusted user did not use -F to supply a sender
1719 name, $originator_name was incorrectly used when constructing a From:
1720 header. Furthermore, $originator_name was used for submission mode
1721 messages from external hosts without From: headers in a similar way,
1722 which is clearly wrong.
1724 PH/37 Added control=suppress_local_fixups.
1726 PH/38 When log_selector = +received_sender was set, and the addition of the
1727 sender made the log line's construction buffer exactly full, or one byte
1728 less than full, an overflow happened when the terminating "\n" was
1731 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1732 when the result of a list match is failure because a DNS lookup failed.
1734 PH/40 RM_COMMAND is now used in the building process.
1736 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1737 the "build-* directories that it finds.
1739 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1740 address in a domain literal was a prefix of an interface address.
1742 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1743 when verifying a sender address, unless rewrite_headers is false.
1745 PH/44 Wrote a long comment about why errors_to addresses are verified as
1746 recipients, not senders.
1748 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1749 the ratelimit ACL was added.
1751 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1753 PH/46 Added extra information about PostgreSQL errors to the error string.
1755 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1756 avoiding the need to install special zones in a real server. This is
1757 backwards compatible; if it can't find the fake resolver, it drops back.
1758 Thus, both old and new test suites can be run.
1760 TF/02 Added util/ratelimit.pl
1762 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1763 clock is set back in time.
1765 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1766 Candler <B.Candler@pobox.com>.
1768 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1769 OK for addresses that are the result of redirections.
1771 PH/48 A number of further additions for the benefit of the new test suite,
1772 including a fake gethostbyname() that interfaces to the fake DNS resolver
1775 TF/06 The fix for widen_domains has also been applied to qualify_single and
1776 search_parents which are the other dnslookup options that can cause
1779 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1782 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1784 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1785 service" if it encountered an MX record with an empty target hostname.
1786 The message is now "an MX or SRV record indicated no SMTP service".
1788 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1789 if the DNS of the sending site is misconfigured. This is quite a
1790 common situation. This change restores the behaviour of treating a
1791 helo verification defer as a failure.
1793 PH/51 If self=fail was set on a router, the bounce message did not include the
1794 actual error message.
1800 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1802 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1803 command itself ended in error, the underlying message about the transport
1804 filter timeout was being overwritten with the pipe command error. Now the
1805 underlying error message should be appended to the second error message.
1807 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1809 PH/02 Reduce the amount of output that "make" produces by default. Full output
1810 can still be requested.
1812 PH/03 The warning log line about a condition test deferring for a "warn" verb
1813 was being output only once per connection, rather than after each
1814 occurrence (because it was using the same function as for successful
1815 "warn" verbs). This seems wrong, so I have changed it.
1817 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1818 it should not have, which might have caused a crash in the right
1819 circumstances, but probably never did.
1821 PH/04 Installed a modified version of Tony Finch's patch to make submission
1822 mode fix the return path as well as the Sender: header line, and to
1823 add a /name= option so that you can make the user's friendly name appear
1826 TF/03 Added the control = fakedefer ACL modifier.
1828 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1829 Mark Lowes for thorough testing.
1831 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1833 TK/03 Merged latest SRS patch from Miles Wilton.
1835 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1836 with the definition in sysexits.h (which is #included earlier).
1837 Fortunately, Exim does not actually use EX_OK. The code used to try to
1838 preserve the sysexits.h value, by assumimg that macro definitions were
1839 scanned for macro replacements. I have been disabused of this notion,
1840 so now the code just undefines EX_OK before #including unistd.h.
1842 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1843 in the smtp transport. When a block could not be written in a single
1844 write() function, the timeout was being re-applied to each part-write.
1845 This seems wrong - if the receiver was accepting one byte at a time it
1846 would take for ever. The timeout is now adjusted when this happens. It
1847 doesn't have to be particularly precise.
1849 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1850 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1852 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1853 without a domain if the check to <postmaster@domain> fails.
1855 SC/01 Eximstats: added -xls and the ability to specify output files
1856 (patch written by Frank Heydlauf).
1858 SC/02 Eximstats: use FileHandles for outputing results.
1860 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1862 SC/04 Eximstats: fixed display of large numbers with -nvr option
1864 SC/05 Eximstats: fixed merging of reports with empty tables.
1866 SC/06 Eximstats: added the -include_original_destination flag
1868 SC/07 Eximstats: removed tabs and trailing whitespace.
1870 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1872 TK/06 MBOX spool code: Add real "From " MBOX separator line
1873 so the .eml file is really in mbox format (even though
1874 most programs do not really care). Patch from Alex Miller.
1876 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1877 The latter is generated from $received_to and is only set if the
1878 message has one envelope recipient. SA can use these headers,
1879 obviously out-of-the-box. Patch from Alex Miller.
1881 PH/08 The ${def test on a variable was returning false if the variable's
1882 value was "0", contrary to what the specification has always said!
1883 The result should be true unless the variable is empty.
1885 PH/09 The syntax error of a character other than { following "${if
1886 def:variable_name" (after optional whitespace) was not being diagnosed.
1887 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1888 accidental colon was present, for example, could give incorrect results.
1890 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1891 result is used (not including appendfile, where other changes are about
1894 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1895 This involved changing a lot of size variables from int to off_t. It
1896 should work with maildirs and everything.
1898 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1899 spamd dying while we are connected to it.
1901 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1904 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1905 function for BDB 4.3.
1907 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1909 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1910 the sources for PCRE have been re-arranged and more files are now
1913 PH/15 The code I had for printing potentially long long variables in PH/11
1914 above was not the best (it lost precision). The length of off_t variables
1915 is now inspected at build time, and an appropriate printing format (%ld
1916 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1917 to be "long long int" or "long int". This is needed for the internal
1918 formatting function string_vformat().
1920 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1921 the configuration file to be ":syslog", then the script "guesses" where
1922 the logs files are, rather than using the compiled in default. In our
1923 case the guess is not the same as the compiled default, so the script
1924 suddenly stopped working when I started to use syslog. The patch checks
1925 to see if log_file_path is "". If so, it attempts to read it from exim
1926 with no configuration file to get the compiled in version, before it
1927 falls back to the previous guessing code."
1929 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1930 implementing BATV in an Exim configuration. See NewStuff for the gory
1933 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1934 Makefile that are specific to HP-UX.
1936 PH/18 If the "use_postmaster" option was set for a recipient callout together
1937 with the "random" option, the postmaster address was used as the MAIL
1938 FROM address for the random test, but not for the subsequent recipient
1939 test. It is now used for both.
1941 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1942 patch removes a few documentation additions to RFC 3028, because the
1943 latest draft now contains them. It adds the new en;ascii-case comparator
1944 and a new error check for 8bit text in MIME parts. Comparator and
1945 require names are now matched exactly. I enabled the subaddress
1946 extension, but it is not well tested yet (read: it works for me)."
1948 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1949 rework some of the code of TK/09 above to avoid the hardwired use of
1950 "%lld" and "long long". Replaced the call to snprintf() with a call to
1953 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1954 records point to non-existent hosts", "retry timeout exceeded", and
1955 "retry time not reached for any host after a long failure period".
1957 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1958 experimental DomainKeys support:
1960 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1961 (2) On an error such as an illegally used "control", the wrong name for
1962 the control was given.
1964 These problems did NOT occur unless DomainKeys support was compiled.
1966 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1968 PH/24 Added ${if match_ip condition.
1970 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1971 databases so that it will be absolutely obvious if a crash occurs in the
1972 DB library. This is a regular occurrence (often caused by mis-matched
1975 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1976 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1977 picked up on a user's system that detects such things. There doesn't seem
1978 to be a gcc warning option for this - only an attribute that has to be
1979 put on the function's prototype. It seems that in Fedora Core 4 they have
1980 set this on a number of new functions. No doubt there will be more in due
1983 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1984 specify a transport. However, if an address that was verified by such a
1985 router was the subject of a callout, Exim crashed because it tried to
1986 read the rcpt_include_affixes from the non-existent transport. Now it
1987 just assumes that the setting of that option is false. This bug was
1988 introduced by 4.51/PH/31.
1990 PH/28 Changed -d+all to exclude +memory, because that information is very
1991 rarely of interest, but it makes the output a lot bigger. People tend to
1992 do -d+all out of habit.
1994 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1995 code in os-type was giving problems when libc.so lives in lib64, like on
1998 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1999 aren't the modern standard, and it seems that some systems' include files
2000 don't always have them. Exim was already checking for some of the newer
2001 ones like T_AAAA, and defining it itself. I've added checks for all the
2002 record types that Exim uses.
2004 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2005 not automatically generating a new one, as it is supposed to. This
2006 prevented TLS from working. If the file did exist, but contained invalid
2007 data, a new version was generated, as expected. It was only the case of a
2008 non-existent file that was broken.
2010 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2011 with a change in libdomainkeys > 0.64.
2013 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2014 from DNS. If the selector record carries the flag, it now has
2015 precedence over the domain-wide flag.
2017 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2019 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2020 the use of a port name that isn't defined in /etc/services) occurred, the
2021 message was deferred as in a normal delivery, and thus remained on the
2022 spool, instead of being failed because of the mua_wrapper setting. This
2023 is now fixed, and I tidied up some of the mua_wrapper messages at the
2026 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2027 the messages in a hash of arrays rather than using individual hashes.
2028 This is a bit cleaner and results in dramatic memory savings, albeit
2029 at a slight CPU cost.
2031 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2032 as requested by Marc Sherman.
2034 SC/10 Eximstats: added histograms for user specified patterns as requested
2037 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2039 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2040 fopen() in the content-scanning modules that did not already have it.
2046 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2047 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2049 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2051 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2053 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2054 to test Sieve filters that use "vacation".
2056 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2057 that changes the way the GnuTLS parameters are stored in the cache file.
2058 The new format can be generated externally. For backward compatibility,
2059 if the data in the cache doesn't make sense, Exim assumes it has read an
2060 old-format file, and it generates new data and writes a new file. This
2061 means that you can't go back to an older release without removing the
2064 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2065 work if there are any delivery delays because "one_time" forces the
2066 parent to be marked "delivered", so its unseen clone is never tried
2067 again. For this reason, Exim now forbids the simultaneous setting of
2070 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2071 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2072 entry if you want to know the details. The fix, however, neglected to
2073 consider the case when local delivery batching is involved. The test for
2074 "previously delivered" was not happening when checking to see if an
2075 address could be batched with a previous (undelivered) one; under
2076 certain circumstances this could lead to multiple deliveries to the same
2079 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2080 in its include files, and this causes problems building Exim.
2082 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2083 header_syntax) but Exim was just ignoring anything given after a slash.
2084 In particular, this caused confusion with an attempt to use "verify =
2085 reverse_host_lookup/defer_ok". An error is now given when options are
2086 supplied for verify items that do not have them. (Maybe reverse_host_
2087 lookup should have a defer_ok option, but that's a different point.)
2089 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2090 defined by RFC 821) to 2048, because there were problems with some AUTH
2091 commands, and RFC 1869 says the size should be increased for extended
2092 SMTP commands that take arguments.
2094 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2097 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2098 "unknown" error; now it says that the functionality isn't in the binary.
2100 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2101 an address' error message when a string expansion fails (syntax or
2102 whatever). Otherwise the password may appear in the log. Following change
2103 PH/42 below, there is no longer a chance of it appearing in a bounce
2106 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2108 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2109 of its addresses was ever tried. (Bugzilla bug #2.)
2111 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2112 the result incorrectly in the debug output. (It correctly added a newline
2113 to what was transported.)
2115 TF/01 Added $received_time.
2117 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2118 commented out examples of how to interface to a virus scanner and to
2119 SpamAssassin. Also added commented examples of av_scanner and
2120 spamd_address settings.
2122 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2123 and controls are allowed in which ACLs. There were a couple of minor
2124 errors. Some of the entries in the conditions table (which is a table of
2125 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2126 negation of where the condition IS allowed.
2128 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2130 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2131 header file does not have a version number, so I've had to invent a new
2132 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2133 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2134 radiusclient), but it was contributed by a Radius user.
2136 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2137 files or directories, for OpenSSL.
2139 PH/19 When an Exim process that is running as root has to create an Exim log
2140 file, it does so in a subprocess that runs as exim:exim so as to get the
2141 ownership right at creation (otherwise, other Exim processes might see
2142 the file with the wrong ownership). There was no test for failure of this
2143 fork() call, which would lead to the process getting stuck as it waited
2144 for a non-existent subprocess. Forks do occasionally fail when resources
2145 run out. I reviewed all the other calls to fork(); they all seem to check
2148 PH/20 When checking for unexpected SMTP input at connect time (before writing
2149 the banner), Exim was not dealing correctly with a non-positive return
2150 from the read() function. If the client had disconnected by this time,
2151 the result was a log entry for a synchronization error with an empty
2152 string after "input=" when read() returned zero. If read() returned -1
2153 (an event I could not check), uninitialized data bytes were printed.
2154 There were reports of junk text (parts of files, etc) appearing after
2157 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2159 PH/22 Added support for macro redefinition, and (re)definition in between
2160 driver and ACL definitions.
2162 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2163 forgetting to use the resulting value; it was using the unexpanded value.
2165 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2166 hadn't been configured. The fix is from Juergen Kreileder, who
2167 understands it better than I do:
2169 "Here's what I see happening with three configured cyrus_sasl
2170 authenticators configured (plain, login, cram-md5):
2172 On startup auth_cyrus_sasl_init() gets called for each of these.
2173 This means three calls to sasl_listmech() without a specified mech_list.
2174 => SASL tests which mechs of all available mechs actually work
2175 => three warnings about OTP not working
2176 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2178 With the patch, sasl_listmech() also gets called three times. But now
2179 SASL's mech_list option is set to the server_mech specified in the the
2180 authenticator. Or in other words, the answer from sasl_listmech()
2181 gets limited to just the mech you're testing for (which is different
2183 => the return list contains just 'plain' or 'login', 'cram-md5' or
2184 nothing depending on the value of ob->server_mech.
2186 I've just tested the patch: Authentication still works fine,
2187 unavailable mechs specified in the exim configuration are still
2188 caught, and the auth.log warnings about OTP are gone."
2190 PH/25 When debugging is enabled, the contents of the command line are added
2191 to the debugging output, even when log_selector=+arguments is not
2194 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2195 answer is "GNU", and only if the return is "GNU/something" is the answer
2198 PH/27 $acl_verify_message is now set immediately after the failure of a
2199 verification in an ACL, and so is available in subsequent modifiers. In
2200 particular, the message can be preserved by coding like this:
2202 warn !verify = sender
2203 set acl_m0 = $acl_verify_message
2205 Previously, $acl_verify_message was set only while expanding "message"
2206 and "log_message" when a very denied access.
2208 PH/28 Modified OS/os.c-Linux with
2210 -#ifndef OS_LOAD_AVERAGE
2211 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2213 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2214 nomenclature these days.)
2216 PH/29 Installed patch from the Sieve maintainer that adds the options
2217 sieve_useraddress and sieve_subaddress to the redirect router.
2219 PH/30 In these circumstances:
2220 . Two addresses routed to the same list of hosts;
2221 . First host does not offer TLS;
2222 . First host accepts first address;
2223 . First host gives temporary error to second address;
2224 . Second host offers TLS and a TLS session is established;
2225 . Second host accepts second address.
2226 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2227 and peerdn, if requested) that were in fact used only for the second
2230 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2231 attention to any local part prefix or suffix that was matched by the
2232 router that accepted the address. It now behaves in the same way as it
2233 does for delivery: the affixes are removed from the local part unless
2234 rcpt_include_affixes is set on the transport.
2236 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2237 timeout during the DATA phase of an incoming message.
2239 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2240 applied a patch sanctioned by the Sieve maintainer.
2242 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2243 the uid or gid is negative. A case of a negative gid caused this to be
2244 noticed. The fix allows for either to be negative.
2246 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2247 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2250 PH/36 The change PH/12 above was broken. Fixed it.
2252 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2253 the grounds that routing the same address twice would always produce the
2254 same answer. This might have been true once, but it is certainly no
2255 longer true now. Routing a child address may depend on the previous
2256 routing that produced that child. Some complicated redirection strategies
2257 went wrong when messages had multiple recipients, and made Exim's
2258 behaviour dependent on the order in which the addresses were given.
2260 I have moved the duplicate checking until after the routing is complete.
2261 Exim scans the addresses that are assigned to local and remote
2262 transports, and removes any duplicates. This means that more work will be
2263 done, as duplicates will always all be routed, but duplicates are
2264 presumably rare, so I don't expect this is of any significance.
2266 For deliveries to pipes, files, and autoreplies, the duplicate checking
2267 still happens during the routing process, since they are not going to be
2270 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2271 It corrects a timeout issue with spamd. This is Ian's comment: "The
2272 background is that sometimes spamd either never reads data from a
2273 connection it has accepted, or it never writes response data. The exiscan
2274 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2275 blindly assumes that writes won't block so it may never time out."
2277 PH/39 Allow G after quota size as well as K and M.
2279 PH/40 The value set for $authenticated_id in an authenticator may not contain
2280 binary zeroes or newlines because the value is written to log lines and
2281 to spool files. There was no check on this. Now the value is run through
2282 the string_printing() function so that such characters are converted to
2283 printable escape sequences.
2285 PH/41 $message_linecount is a new variable that contains the total number of
2286 lines in the message. Compare $body_linecount, which is the count for the
2289 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2290 bounce and delay warning messages, except in certain special cases, which
2293 (a) An SMTP error message from a remote host;
2294 (b) A message specified in a :fail: redirection;
2295 (c) A message specified in a "fail" command in a system filter;
2296 (d) A message specified in a FAIL return from the queryprogram router;
2297 (e) A message specified by the cannot_route_message router option.
2299 In these cases only, Exim does include the error details in bounce and
2300 warning messages. There are also a few cases where bland messages such
2301 as "unrouteable address" or "local delivery error" are given.
2303 PH/43 $value is now also set for the "else" part of a ${run expansion.
2305 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2306 being worked on, but at least Exim now implements the latest version to
2309 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2310 process to complete was treated as a delivery failure, a timeout while
2311 writing the message to the pipe was logged, but erroneously treated as a
2312 successful delivery. Such timeouts include transport filter timeouts. For
2313 consistency with the overall process timeout, these timeouts are now
2314 treated as errors, giving rise to delivery failures by default. However,
2315 there is now a new Boolean option for the pipe transport called
2316 timeout_defer, which, if set TRUE, converts the failures into defers for
2317 both kinds of timeout. A transport filter timeout is now identified in
2320 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2321 systems where "make" and "gmake" are different, calling "gmake" at top
2322 level broke things. I've arranged for the value of $(MAKE) to be passed
2323 from the Makefile to this script so that it can call the same version of
2327 A note about Exim versions 4.44 and 4.50
2328 ----------------------------------------
2330 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2331 changes of various kinds. As a consequence, a big documentation update was
2332 needed. This delayed the release for rather longer than seemed good, especially
2333 in the light of a couple of (minor) security issues. Therefore, the changes
2334 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2335 release. So 4.44 and 4.50 are in effect two different branches that both start
2338 I have left the 4.50 change log unchanged; it contains all the changes since
2339 4.43. The change log for 4.44 is below; many of its items are identical to
2340 those for 4.50. This seems to be the most sensible way to preserve the
2341 historical information.
2347 1. Minor wording change to the doc/README.SIEVE file.
2349 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2350 computation of the current number of files was incorrect.
2352 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2353 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2356 4. Give more explanation in the error message when the command for a transport
2357 filter fails to execute.
2359 5. There are several places where Exim runs a non-Exim command in a
2360 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2361 was being done only for the command run by the queryprogram router. It is
2362 now done for all such subprocesses. The other cases are: ${run, transport
2363 filters, and the commands run by the lmtp and pipe transports.
2365 6. Added CONFIGURE_GROUP build-time option.
2367 7. Some older OS have a limit of 256 on the maximum number of file
2368 descriptors. Exim was using setrlimit() to set 1000 as a large value
2369 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2370 systems. I've change it so that if it can't get 1000, it tries for 256.
2372 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2373 was an oversight, and furthermore, ever since the addition of extra
2374 controls (e.g. 4.43/32), the checks on when to allow different forms of
2375 "control" were broken. There should now be diagnostics for all cases when a
2376 control that does not make sense is encountered.
2378 9. Added the /retain_sender option to "control=submission".
2380 10. $recipients is now available in the predata ACL (oversight).
2382 11. Tidy the search cache before the fork to do a delivery from a message
2383 received from the command line. Otherwise the child will trigger a lookup
2384 failure and thereby defer the delivery if it tries to use (for example) a
2385 cached ldap connection that the parent has called unbind on.
2387 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2388 of $address_data from the recipient verification was clobbered by the
2389 sender verification.
2391 13. The value of address_data from a sender verification is now available in
2392 $sender_address_data in subsequent conditions in the ACL statement.
2394 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2396 15. Added a new option "connect=<time>" to callout options, to set a different
2399 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2400 was its contents. (It was OK if the option was not defined at all.)
2402 17. A "Completed" log line is now written for messages that are removed from
2403 the spool by the -Mrm option.
2405 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2406 information about exactly what failed.
2408 19. Added -dd to debug only the daemon process.
2410 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2411 handles timeouts, both on the server side and network timeouts. Renamed the
2412 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2414 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2415 It is now set to "smtps".
2417 22. $host_address is now set to the target address during the checking of
2418 ignore_target_hosts.
2420 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2421 being passed; this would have caused $sender_host_name to have been used if
2422 matching the list had actually called for a host name (not very likely,
2423 since this list is usually IP addresses). A host name is now passed as
2426 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2427 code that specifies a non-restarting handler (typically sigaction() in
2428 modern systems) in an attempt to fix a rare and obscure crash bug.
2430 25. Narrowed the window for a race in the daemon that could cause it to ignore
2431 SIGCHLD signals. This is not a major problem, because they are used only to
2432 wake it up if nothing else does.
2434 26. A malformed maildirsize file could cause Exim to calculate negative values
2435 for the mailbox size or file count. Odd effects could occur as a result.
2436 The maildirsize information is now recalculated if the size or filecount
2439 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2440 support for a long time. Removed HAVE_SYS_VFS_H.
2442 28. Installed the latest version of exipick from John Jetmore.
2444 29. In an address list, if the pattern was not a regular expression, an empty
2445 subject address (from a bounce message) matched only if the pattern was an
2446 empty string. Non-empty patterns were not even tested. This was the wrong
2447 because it is perfectly reasonable to use an empty address as part of a
2448 database query. An empty address is now tested by patterns that are
2449 lookups. However, all the other forms of pattern expect the subject to
2450 contain a local part and a domain, and therefore, for them, an empty
2451 address still always fails if the pattern is not itself empty.
2453 30. Exim went into a mad DNS loop when attempting to do a callout where the
2454 host was specified on an smtp transport, and looking it up yielded more
2455 than one IP address.
2457 31. Re-factored the code for checking spool and log partition space into a
2458 function that finds that data and another that does the check. The former
2459 is then used to implement four new variables: $spool_space, $log_space,
2460 $spool_inodes, and $log_inodes.
2462 32. The RFC2047 encoding function was originally intended for short strings
2463 such as real names; it was not keeping to the 75-character limit for
2464 encoded words that the RFC imposes. It now respects the limit, and
2465 generates multiple encoded words if necessary. To be on the safe side, I
2466 have increased the buffer size for the ${rfc2047: expansion operator from
2469 33. It is now permitted to omit both strings after an "if" condition; if the
2470 condition is true, the result is "true". As before, when the second string
2471 is omitted, a false condition yields an empty string. This makes it less
2472 cumbersome to write custom ACL and router conditions.
2474 34. Failure to deliver a bounce message always caused it to be frozen, even if
2475 there was an errors_to setting on the router. The errors_to setting is now
2478 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2479 canonical form (fully expanded) before being placed in
2480 $sender_host_address.
2482 36. The table in the code that translates DNS record types into text (T_A to
2483 "A" for instance) was missing entries for NS and CNAME. It is just possible
2484 that this could have caused confusion if both these types were looked up
2485 for the same domain, because the text type is used as part of Exim's
2486 per-process caching. But the chance of anyone hitting this buglet seems
2489 37. The dnsdb lookup has been extended in a number of ways.
2491 (1) There is a new type, "zns", which walks up the domain tree until it
2492 finds some nameserver records. It should be used with care.
2494 (2) There is a new type, "mxh", which is like "mx" except that it returns
2495 just the host names, not the priorities.
2497 (3) It is now possible to give a list of domains (or IP addresses) to be
2498 looked up. The behaviour when one of the lookups defers can be
2499 controlled by a keyword.
2501 (4) It is now possible to specify the separator character for use when
2502 multiple records are returned.
2504 38. The dnslists ACL condition has been extended: it is now possible to supply
2505 a list of IP addresses and/or domains to be looked up in a particular DNS
2508 39. Added log_selector=+queue_time_overall.
2510 40. When running the queue in the test harness, wait just a tad after forking a
2511 delivery process, to get repeatability of debugging output.
2513 41. Include certificate and key file names in error message when GnuTLS fails
2514 to set them up, because the GnuTLS error message doesn't include the name
2515 of the failing file when there is a problem reading it.
2517 42. Allow both -bf and -bF in the same test run.
2519 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2521 44. The "Exiscan patch" is now merged into the mainline Exim source.
2523 45. Sometimes the final signoff response after QUIT could fail to get
2524 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2525 < 0 before doing a fflush(). This bug looks as though it goes back to the
2526 introduction of TLS in release 3.20, but "sometimes" must have been rare
2527 because the tests only now provoked it.
2529 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2530 (this can affect the format of dates).
2532 47. exim_tidydb, when checking for the continued existence of a message for
2533 which it has found a message-specific retry record, was not finding
2534 messages that were in split spool directories. Consequently, it was
2535 deleting retry records that should have stayed in existence.
2537 48. Steve fixed some bugs in eximstats.
2539 49. The SPA authentication driver was not abandoning authentication and moving
2540 on to the next authenticator when an expansion was forced to fail,
2541 contradicting the general specification for all authenticators. Instead it
2542 was generating a temporary error. It now behaves as specified.
2544 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2545 (the order specifies the preference for clients). The order is now AES256,
2546 AES128, 3DES, ARCFOUR128.
2548 51. Small patch to Sieve code - explicitly set From: when generating an
2551 52. Exim crashed if a remote delivery caused a very long error message to be
2552 recorded - for instance if somebody sent an entire SpamAssassin report back
2553 as a large number of 550 error lines. This bug was coincidentally fixed by
2554 increasing the size of one of Exim's internal buffers (big_buffer) that
2555 happened as part of the Exiscan merge. However, to be on the safe side, I
2556 have made the code more robust (and fixed the comments that describe what
2559 53. Now that there can be additional text after "Completed" in log lines (if
2560 the queue_time_overall log selector is set), a one-byte patch to exigrep
2561 was needed to allow it to recognize "Completed" as not the last thing in
2564 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2565 patch that reportedly fixes this has been added. I am not expert enough to
2566 create a test for it. This is what the patch creator wrote:
2568 "I found a little strange behaviour of ldap code when working with
2569 Windows 2003 AD Domain, where users was placed in more than one
2570 Organization Units. When I tried to give exim partial DN, the exit code
2571 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2572 But simultaneously result of request was absolutely normal ldap result,
2573 so I produce this patch..."
2575 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2576 REFERENCE, so I have modified the code to exclude the patch when that macro
2579 55. Some experimental protocols are using DNS PTR records for new purposes. The
2580 keys for these records are domain names, not reversed IP addresses. The
2581 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2582 leaves it alone. Component reversal etc. now happens only for IP addresses.
2585 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2587 57. Double the size of the debug message buffer (to 2048) so that more of very
2588 long debug lines gets shown.
2590 58. The exicyclog utility now does better if the number of log files to keep
2591 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2593 59. Two changes related to the smtp_active_hostname option:
2595 (1) $smtp_active_hostname is now available as a variable.
2596 (2) The default for smtp_banner uses $smtp_active_hostname instead
2597 of $primary_hostname.
2599 60. The host_aton() function is supposed to be passed a string that is known
2600 to be a valid IP address. However, in the case of IPv6 addresses, it was
2601 not checking this. This is a hostage to fortune. Exim now panics and dies
2602 if the condition is not met. A case was found where this could be provoked
2603 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2604 components; fortuitously, this particular loophole had already been fixed
2605 by change 4.50/55 above.
2607 If there are any other similar loopholes, the new check in host_aton()
2608 itself should stop them being exploited. The report I received stated that
2609 data on the command line could provoke the exploit when Exim was running as
2610 exim, but did not say which command line option was involved. All I could
2611 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2612 running as the user.
2615 61. There was a buffer overflow vulnerability in the SPA authentication code
2616 (which came originally from the Samba project). I have added a test to the
2617 spa_base64_to_bits() function which I hope fixes it.
2620 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2621 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2623 63. The daemon start-up calls getloadavg() while still root for those OS that
2624 need the first call to be done as root, but it missed one case: when
2625 deliver_queue_load_max is set with deliver_drop_privilege. This is
2626 necessary for the benefit of the queue runner, because there is no re-exec
2627 when deliver_drop_privilege is set.
2629 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2630 This has been fixed.
2632 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2633 was in use, was not putting the data itself into the right store pool;
2634 consequently, it could be overwritten for a subsequent message in the same
2635 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2638 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2640 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2641 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2642 the function were treating the return as a boolean value, which happened to
2643 work because 0=false and not-0=true, but is not correct code.
2645 68. The host_aton() function was not handling scoped IPv6 addresses (those
2646 with, for example, "%eth0" on the end) correctly.
2648 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2649 negated items (that is, ~something) in unsigned ints. Some compilers
2650 apparently mutter when there is no cast.
2652 70. If an address verification called from an ACL failed, and did not produce a
2653 user-specific message (i.e. there was only a "system" message), nothing was
2654 put in $acl_verify_message. In this situation, it now puts the system
2657 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2658 session; change 4.31/43 added the unwanted input to the log line - except
2659 that it did not do this in the start of session case. It now does.
2661 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2662 This is wrong and can cause the other end to generate a synchronization
2663 error if it is another Exim or anything else that does the synchronization
2664 check. A QUIT command is no longer sent after a timeout.
2666 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2667 during host lookups.
2669 74. The defer_ok option of callout verification was not working if it was used
2670 when verifying addresses in header lines, that is, for this case:
2672 verify = header_sender/callout=defer_ok
2674 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2675 those file descriptors could be used for SMTP connections. If anything
2676 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2677 could be sent to the SMTP client, causing chaos. The daemon now opens
2678 stdin, stdout, and stderr to /dev/null when it puts itself into the
2681 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2682 log by default. The user can override this with suitable Perl magic.
2684 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2685 the log message when discard triggers, was not working for the DATA ACL or
2686 for the non-SMTP ACL.
2688 78. Error message wording change in sieve.c.
2690 79. If smtp_accept_max_per_host was set, the number of connections could be
2691 restricted to fewer than expected, because the daemon was trying to set up
2692 a new connection before checking whether the processes handling previous
2693 connections had finished. The check for completed processes is now done
2694 earlier. On busy systems, this bug wouldn't be noticed because something
2695 else would have woken the daemon, and it would have reaped the completed
2698 80. If a message was submitted locally by a user whose login name contained one
2699 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2700 It caused a spool format error. I have fixed the spool reading code. A
2701 related problem was that the "from" clause in the Received: line became
2702 illegal because of the space(s). It is now covered by ${quote_local_part.
2704 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2707 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2709 83. Patch from Sieve maintainer to fix unterminated string problem in
2710 "vacation" handling.
2712 84. Some minor changes to the Linux configuration files to help with other
2713 OS variants using glibc.
2715 85. One more patch for Sieve to update vacation handling to latest spec.
2718 ----------------------------------------------------
2719 See the note above about the 4.44 and 4.50 releases.
2720 ----------------------------------------------------
2726 1. Change 4.43/35 introduced a bug that caused file counts to be
2727 incorrectly computed when quota_filecount was set in an appendfile
2730 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2731 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2734 3. Give more explanation in the error message when the command for a transport
2735 filter fails to execute.
2737 4. There are several places where Exim runs a non-Exim command in a
2738 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2739 was being done only for the command run by the queryprogram router. It is
2740 now done for all such subprocesses. The other cases are: ${run, transport
2741 filters, and the commands run by the lmtp and pipe transports.
2743 5. Some older OS have a limit of 256 on the maximum number of file
2744 descriptors. Exim was using setrlimit() to set 1000 as a large value
2745 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2746 systems. I've change it so that if it can't get 1000, it tries for 256.
2748 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2749 was an oversight, and furthermore, ever since the addition of extra
2750 controls (e.g. 4.43/32), the checks on when to allow different forms of
2751 "control" were broken. There should now be diagnostics for all cases when a
2752 control that does not make sense is encountered.
2754 7. $recipients is now available in the predata ACL (oversight).
2756 8. Tidy the search cache before the fork to do a delivery from a message
2757 received from the command line. Otherwise the child will trigger a lookup
2758 failure and thereby defer the delivery if it tries to use (for example) a
2759 cached ldap connection that the parent has called unbind on.
2761 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2762 of $address_data from the recipient verification was clobbered by the
2763 sender verification.
2765 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2766 was its contents. (It was OK if the option was not defined at all.)
2768 11. A "Completed" log line is now written for messages that are removed from
2769 the spool by the -Mrm option.
2771 12. $host_address is now set to the target address during the checking of
2772 ignore_target_hosts.
2774 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2775 being passed; this would have caused $sender_host_name to have been used if
2776 matching the list had actually called for a host name (not very likely,
2777 since this list is usually IP addresses). A host name is now passed as
2780 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2781 code that specifies a non-restarting handler (typically sigaction() in
2782 modern systems) in an attempt to fix a rare and obscure crash bug.
2784 15. Narrowed the window for a race in the daemon that could cause it to ignore
2785 SIGCHLD signals. This is not a major problem, because they are used only to
2786 wake it up if nothing else does.
2788 16. A malformed maildirsize file could cause Exim to calculate negative values
2789 for the mailbox size or file count. Odd effects could occur as a result.
2790 The maildirsize information is now recalculated if the size or filecount
2793 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2794 support for a long time. Removed HAVE_SYS_VFS_H.
2796 18. Updated exipick to current release from John Jetmore.
2798 19. Allow an empty sender to be matched against a lookup in an address list.
2799 Previously the only cases considered were a regular expression, or an
2802 20. Exim went into a mad DNS lookup loop when doing a callout where the
2803 host was specified on the transport, if the DNS lookup yielded more than
2806 21. The RFC2047 encoding function was originally intended for short strings
2807 such as real names; it was not keeping to the 75-character limit for
2808 encoded words that the RFC imposes. It now respects the limit, and
2809 generates multiple encoded words if necessary. To be on the safe side, I
2810 have increased the buffer size for the ${rfc2047: expansion operator from
2813 22. Failure to deliver a bounce message always caused it to be frozen, even if
2814 there was an errors_to setting on the router. The errors_to setting is now
2817 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2818 canonical form (fully expanded) before being placed in
2819 $sender_host_address.
2821 24. Updated eximstats to version 1.33
2823 25. Include certificate and key file names in error message when GnuTLS fails
2824 to set them up, because the GnuTLS error message doesn't include the name
2825 of the failing file when there is a problem reading it.
2827 26. Expand error message when OpenSSL has problems setting up cert/key files.
2830 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2831 (this can affect the format of dates).
2833 28. exim_tidydb, when checking for the continued existence of a message for
2834 which it has found a message-specific retry record, was not finding
2835 messages that were in split spool directories. Consequently, it was
2836 deleting retry records that should have stayed in existence.
2838 29. eximstats updated to version 1.35
2839 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2840 1.35 - bugfix such that pie charts by volume are generated correctly
2842 30. The SPA authentication driver was not abandoning authentication and moving
2843 on to the next authenticator when an expansion was forced to fail,
2844 contradicting the general specification for all authenticators. Instead it
2845 was generating a temporary error. It now behaves as specified.
2847 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2848 (the order specifies the preference for clients). The order is now AES256,
2849 AES128, 3DES, ARCFOUR128.
2851 31. Small patch to Sieve code - explicitly set From: when generating an
2854 32. Exim crashed if a remote delivery caused a very long error message to be
2855 recorded - for instance if somebody sent an entire SpamAssassin report back
2856 as a large number of 550 error lines. This bug was coincidentally fixed by
2857 increasing the size of one of Exim's internal buffers (big_buffer) that
2858 happened as part of the Exiscan merge. However, to be on the safe side, I
2859 have made the code more robust (and fixed the comments that describe what
2862 33. Some experimental protocols are using DNS PTR records for new purposes. The
2863 keys for these records are domain names, not reversed IP addresses. The
2864 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2865 leaves it alone. Component reversal etc. now happens only for IP addresses.
2868 34. The host_aton() function is supposed to be passed a string that is known
2869 to be a valid IP address. However, in the case of IPv6 addresses, it was
2870 not checking this. This is a hostage to fortune. Exim now panics and dies
2871 if the condition is not met. A case was found where this could be provoked
2872 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2873 components; fortuitously, this particular loophole had already been fixed
2874 by change 4.50/55 or 4.44/33 above.
2876 If there are any other similar loopholes, the new check in host_aton()
2877 itself should stop them being exploited. The report I received stated that
2878 data on the command line could provoke the exploit when Exim was running as
2879 exim, but did not say which command line option was involved. All I could
2880 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2881 running as the user.
2884 35. There was a buffer overflow vulnerability in the SPA authentication code
2885 (which came originally from the Samba project). I have added a test to the
2886 spa_base64_to_bits() function which I hope fixes it.
2889 36. The daemon start-up calls getloadavg() while still root for those OS that
2890 need the first call to be done as root, but it missed one case: when
2891 deliver_queue_load_max is set with deliver_drop_privilege. This is
2892 necessary for the benefit of the queue runner, because there is no re-exec
2893 when deliver_drop_privilege is set.
2895 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2896 was in use, was not putting the data itself into the right store pool;
2897 consequently, it could be overwritten for a subsequent message in the same
2898 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2901 38. Sometimes the final signoff response after QUIT could fail to get
2902 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2903 < 0 before doing a fflush(). This bug looks as though it goes back to the
2904 introduction of TLS in release 3.20, but "sometimes" must have been rare
2905 because the tests only now provoked it.
2911 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2912 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2913 Now it returns an integer. A number of calls were still expecting a T/F
2914 return. Fortuitously, in all cases, the tests worked in OK situations,
2915 which is the norm. However, things would have gone wrong on any write
2916 failures on the smtp file descriptor. This function is used when sending
2917 messages over SMTP and also when doing verify callouts.
2919 2. When Exim is called to do synchronous delivery of a locally submitted
2920 message (the -odf or -odi options), it no longer closes stderr before doing
2923 3. Implemented the mua_wrapper option.
2925 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2927 5. Implemented the functions header_remove(), header_testname(),
2928 header_add_at_position(), and receive_remove_recipient(), and exported them
2931 6. If an ACL "warn" statement specified the addition of headers, Exim already
2932 inserted X-ACL-Warn: at the start if there was no header name. However, it
2933 was not making this test for the second and subsequent header lines if
2934 there were newlines in the string. This meant that an invalid header could
2935 be inserted if Exim was badly configured.
2937 7. Allow an ACL "warn" statement to add header lines at the start or after all
2938 the Received: headers, as well as at the end.
2940 8. Added the rcpt_4xx retry error code.
2942 9. Added postmaster_mailfrom=xxx to callout verification option.
2944 10. Added mailfrom=xxxx to the callout verification option, for verify=
2947 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2948 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2950 12. Inserted some casts to stop certain compilers warning when using pointer
2951 differences as field lengths or precisions in printf-type calls (mostly
2952 affecting debugging statements).
2954 13. Added optional readline() support for -be (dynamically loaded).
2956 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2957 same clock tick as a message's arrival, so that its received time was the
2958 same as the "first fail" time on the retry record, and that message
2959 remained on the queue past the ultimate address timeout, every queue runner
2960 would try a delivery (because it was past the ultimate address timeout) but
2961 after another failure, the ultimate address timeout, which should have then
2962 bounced the address, did not kick in. This was a "< instead of <=" error;
2963 in most cases the first failure would have been in the next clock tick
2964 after the received time, and all would be well.
2966 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2967 being recognized when the domain list was tested by the match_domain
2968 condition in an expansion string.
2970 16. Added the ${str2b64: operator.
2972 17. Exim was always calling setrlimit() to set a large limit for the number of
2973 processes, without checking whether the existing limit was already
2974 adequate. (It did check for the limit on file descriptors.) Furthermore,
2975 errors from getrlimit() and setrlimit() were being ignored. Now they are
2976 logged to the main and panic logs, but Exim does carry on, to try to do its
2977 job under whatever limits there are.
2979 18. Imported PCRE 5.0.
2981 19. Trivial typo in log message " temporarily refused connection" (the leading
2984 20. If the log selector return_path_on_delivery was set and an address was
2985 redirected to /dev/null, the delivery process crashed because it assumed
2986 that a return path would always be set for a "successful" delivery. In this
2987 case, the whole delivery is bypassed as an optimization, and therefore no
2990 21. Internal re-arrangement: the function for sending a challenge and reading
2991 a response while authentication was assuming a zero-terminated challenge
2992 string. It's now changed to take a pointer and a length, to allow for
2993 binary data in such strings.
2995 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2997 23. Exim was not respecting finduser_retries when seeking the login of the
2998 uid under which it was called; it was always trying 10 times. (The default
2999 setting of finduser_retries is zero.) Also, it was sleeping after the final
3000 failure, which is pointless.
3002 24. Implemented tls_on_connect_ports.
3004 25. Implemented acl_smtp_predata.
3006 26. If the domain in control=submission is set empty, Exim assumes that the
3007 authenticated id is a complete email address when it generates From: or
3008 Sender: header lines.
3010 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3011 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3012 chown and chgrp in /bin and hostname in /usr/bin.
3014 28. Exim was keeping the "process log" file open after each use, just as it
3015 does for the main log. This opens the possibility of it remaining open for
3016 long periods when the USR1 signal hits a daemon. Occasional processlog
3017 errors were reported, that could have been caused by this. Anyway, it seems
3018 much more sensible not to leave this file open at all, so that is what now
3021 29. The long-running daemon process does not normally write to the log once it
3022 has entered its main loop, and it closes the log before doing so. This is
3023 so that log files can straightforwardly be renamed and moved. However,
3024 there are a couple of unusual error situations where the daemon does write
3025 log entries, and I had neglected to close the log afterwards.
3027 30. The text of an SMTP error response that was received during a remote
3028 delivery was being truncated at 512 bytes. This is too short for some of
3029 the long messages that one sometimes sees. I've increased the limit to
3032 31. It is now possible to make retry rules that apply only when a message has a
3033 specific sender, in particular, an empty sender.
3035 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3036 it possible to be selective about when SMTP synchronization is enforced.
3038 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3040 32. Implemented hosts_connection_nolog.
3042 33. Added an ACL for QUIT.
3044 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3047 35. Added mailbox_size and mailbox_filecount to appendfile.
3049 36. Added control = no_multiline_responses to ACLs.
3051 37. There was a bug in the logic of the code that waits for the clock to tick
3052 in the case where the clock went backwards by a substantial amount such
3053 that the microsecond fraction of "now" was more than the microsecond
3054 fraction of "then" (but the whole seconds number was less).
3056 38. Added support for the libradius Radius client library this is found on
3057 FreeBSD (previously only the radiusclient library was supported).
3063 1. When certain lookups returned multiple values in the form name=value, the
3064 quoting of the values was not always being done properly. Specifically:
3065 (a) If the value started with a double quote, but contained no whitespace,
3067 (b) If the value contained whitespace other than a space character (i.e.
3068 tabs or newlines or carriage returns) it was not quoted.
3069 This fix has been applied to the mysql and pgsql lookups by writing a
3070 separate quoting function and calling it from the lookup code. The fix
3071 should probably also be applied to nisplus, ibase and oracle lookups, but
3072 since I cannot test any of those, I have not disturbed their existing code.
3074 2. A hit in the callout cache for a specific address caused a log line with no
3075 reason for rejecting RCPT. Now it says "Previous (cached) callout
3076 verification failure".
3078 3. There was an off-by-one bug in the queryprogram router. An over-long
3079 return line was truncated at 256 instead of 255 characters, thereby
3080 overflowing its buffer with the terminating zero. As well as fixing this, I
3081 have increased the buffer size to 1024 (and made a note to document this).
3083 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3084 when Exim is waiting for an SMTP response from a remote server, Exim
3085 restarts its select() call on the socket, thereby resetting its timeout.
3086 This is not a problem when such interrupts are rare. Somebody set up a cron
3087 job to run exiwhat every 2 minutes, which is less than the normal select()
3088 timeout (5 or 10 minutes). This meant that the select() timeout never
3089 kicked in because it was always reset. I have fixed this by comparing the
3090 time when an interrupt arrives with the time at the start of the first call
3091 to select(). If more time than the timeout has elapsed, the interrupt is
3092 treated as a timeout.
3094 5. Some internal re-factoring in preparation for the addition of Sieve
3095 extensions (by MH). In particular, the "personal" test is moved to a
3096 separate function, and given an option for scanning Cc: and Bcc: (which is
3097 not set for Exim filters).
3099 6. When Exim created an email address using the login of the caller as the
3100 local part (e.g. when creating a From: or Sender: header line), it was not
3101 quoting the local part when it contained special characters such as @.
3103 7. Installed new OpenBSD configuration files.
3105 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3106 try to make them clearer.
3108 9. Callout options, other than the timeout value, were being ignored when
3109 verifying sender addresses in header lines. For example, when using
3111 verify = header_sender/callout=no_cache
3113 the cache was (incorrectly) being used.
3115 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3116 only the Cygwin environment.
3118 11. When return_path_on_delivery was set as a log selector, if different remote
3119 addresses in the same message used different return paths and parallel
3120 remote delivery occurred, the wrong values would sometimes be logged.
3121 (Whenever a remote delivery process finished, the return path value from
3122 the most recently started remote delivery process was logged.)
3124 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3125 lines when AUTH and/or TLS are in use. This is the "received protocol"
3126 field. Exim used to use "asmtp" for authenticated SMTP, without any
3127 indication (in the protocol name) for TLS use. Now it follows the RFC and
3128 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3129 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3130 names appear in log lines as well as in Received: header lines.
3132 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3133 extensions, and comparison tests, and to fix some bugs.
3135 14. Changes to the "personal" filter test:
3137 (1) The test was buggy in that it was just doing the equivalent of
3138 "contains" tests on header lines. For example, if a user's address was
3139 anne@some.where, the "personal" test would incorrectly be true for
3141 To: susanne@some.where
3143 This test is now done by extracting each address from the header in turn,
3144 and checking the entire address. Other tests that are part of "personal"
3145 are now done using regular expressions (for example, to check local parts
3146 of addresses in From: header lines).
3148 (2) The list of non-personal local parts in From: addresses has been
3149 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3150 taken from the Sieve specification recommendations.
3152 (3) If the message contains any header line starting with "List-" it is
3153 treated as non-personal.
3155 (4) The test for "circular" in the Subject: header line has been removed
3156 because it now seems ill-conceived.
3158 15. Minor typos in src/EDITME comments corrected.
3160 16. Installed latest exipick from John Jetmore.
3162 17. If headers_add on a router specified a text string that was too long for
3163 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3164 of string_sprintf() is now avoided.
3166 18. $message_body_size was not set (it was always zero) when running the DATA
3167 ACL and the local_scan() function.
3169 19. For the "mail" command in an Exim filter, no default was being set for
3170 the once_repeat time, causing a random time value to be used if "once" was
3171 specified. (If the value happened to be <= 0, no repeat happened.) The
3172 default is now 0s, meaning "never repeat". The "vacation" command was OK
3173 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3174 (I found it when inspecting the code).
3176 20. There is now an overall timeout for performing a callout verification. It
3177 defaults to 4 times the callout timeout, which applies to individual SMTP
3178 commands during the callout. The overall timeout applies when there is more
3179 than one host that can be tried. The timeout is checked before trying the
3180 next host. This prevents very long delays if there are a large number of
3181 hosts and all are timing out (e.g. when the network connections are timing
3182 out). The value of the overall timeout can be changed by specifying an
3183 additional sub-option for "callout", called "maxwait". For example:
3185 verify = sender/callout=5s,maxwait=20s
3187 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3188 to the end before writing, but this should make it even safer).
3190 22. Exim was forgetting that it had advertised PIPELINING for the second and
3191 subsequent messages on an SMTP connection. It was also not resetting its
3192 memory on STARTTLS and an internal HELO.
3194 23. When Exim logs an SMTP synchronization error within a session, it now
3195 records whether PIPELINING has been advertised or not.
3197 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3198 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3199 rather than long int.
3201 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3203 26. Added the never_mail option to autoreply.
3209 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3210 crash if the getsockname() call failed; this can happen if a connection is
3211 closed very soon after it is established. The problem was simply in the
3212 order in which certain operations were done, causing Exim to try to write
3213 to the SMTP stream before it had set up the file descriptor. The bug has
3214 been fixed by making things happen in the correct order.
3220 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3221 before the connection was closed, thus losing the rejection response.
3223 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3224 some early Solaris releases, but causes trouble in current releases where
3225 socklen_t is defined.
3227 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3230 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3233 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3234 that releasing the top of it at the end releases what was used for sub-
3235 expansions (unless the block got too big). However, discard this block if
3236 the first thing is a variable or header, so that we can use its block when
3237 it is dynamic (useful for very large $message_headers, for example).
3239 6. Lookups now cache *every* query, not just the most recent. A new, separate
3240 store pool is used for this. It can be recovered when all lookup caches are
3241 flushed. Lookups now release memory at the end of their result strings.
3242 This has involved some general refactoring of the lookup sources.
3244 7. Some code has been added to the store_xxx() functions to reduce the amount
3245 of flapping under certain conditions.
3247 8. log_incoming_interface used to affect only the <= reception log lines. Now
3248 it causes the local interface and port to be added to several more SMTP log
3249 lines, for example "SMTP connection from", and rejection lines.
3251 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3253 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3255 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3256 could be overwritten at the end of the current message (or the start of a
3257 new message if it was set in a HELO ACL). The value is now preserved for
3258 the duration of the SMTP connection.
3260 12. If a transport had a headers_rewrite setting, and a matching header line
3261 contained an unqualified address, that address was qualified, even if it
3262 did not match any rewriting rules. The underlying bug was that the values
3263 of the flags that permit the existence of unqualified sender and recipient
3264 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3265 non-local messages, and by -bnq for local messages) were not being
3266 preserved with the message after it was received.
3268 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3269 "next input=" as part of the text comprising the host identity instead of
3270 the correct text. The code was using the same buffer for two different
3271 strings. However, depending on which order the printing function evaluated
3272 its arguments, the bug did not always show up. Under Linux, for example, my
3273 test suite worked just fine.
3275 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3276 doesn't work with some older versions of Perl. It has been changed to "my",
3277 which in any case is probably the better facility to use.
3279 15. A really picky compiler found some instances of statements for creating
3280 error messages that either had too many or two few arguments for the format
3283 16. The size of the buffer for calls to the DNS resolver has been increased
3284 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3285 for addresses that have a lot of PTR records. This alleviates a problem; it
3286 does not fully solve it.
3288 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3289 buffer now truncates the list and logs the incident, which is the same
3290 action as happens when Exim is looking up a host name and its aliases.
3291 Previously in this situation something unpredictable would happen;
3292 sometimes it was "internal error: store_reset failed".
3294 18. If a server dropped the connection unexpectedly when an Exim client was
3295 using GnuTLS and trying to read a response, the client delivery process
3296 crashed while trying to generate an error log message.
3298 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3299 string, for example:
3301 warn message = H1: something\nH2: something
3303 the text was added as a single header line from Exim's point of view
3304 though it ended up OK in the delivered message. However, searching for the
3305 second and subsequent header lines using $h_h2: did not work. This has been
3306 fixed. Similarly, if a system filter added multiple headers in this way,
3307 the routers could not see them.
3309 20. Expanded the error message when iplsearch is called with an invalid key to
3310 suggest using net-iplsearch in a host list.
3312 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3313 ACLs are no longer actually imposed (and a message to that effect is
3316 22. If a "gecos" field in a passwd entry contained escaped characters, in
3317 particular, if it contained a \" sequence, Exim got it wrong when building
3318 a From: or a Sender: header from that name. A second bug also caused
3319 incorrect handling when an unquoted " was present following a character
3320 that needed quoting.
3322 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3323 was not being matched caselessly.
3325 24. Arranged for all hyphens in the exim.8 source to be escaped with
3328 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3329 properly. Recipient callout cache records were still being keyed to include
3330 the sender, even when use_sender was set false. This led to far more
3331 callouts that were necessary. The sender is no longer included in the key
3332 when use_sender is false.
3334 26. Added "control = submission" modifier to ACLs.
3336 27. Added the ${base62d: operator to decode base 62 numbers.
3338 28. dnsdb lookups can now access SRV records.
3340 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3341 the configuration file.
3343 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3344 (-v) mode. This makes the output for a verbose queue run more intelligible.
3346 31. Added a use_postmaster feature to recipient callouts.
3348 32. Added the $body_zerocount variable, containing the number of binary zero
3349 bytes in the message body.
3351 33. The time of last modification of the "new" subdirectory is now used as the
3352 "mailbox time last read" when there is a quota error for a maildir
3355 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3357 35. Added +ignore_unknown as a special item in host lists.
3359 36. Code for decoding IPv6 addresses in host lists is now included, even if
3360 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3361 address was recognized as an IP address, but was then not correctly decoded
3362 into binary, causing unexpected and incorrect effects when compared with
3369 1. Very minor rewording of debugging text in manualroute to say "list of
3370 hosts" instead of "hostlist".
3372 2. If verify=header_syntax was set, and a header line with an unqualified
3373 address (no domain) and a large number of spaces between the end of the
3374 name and the colon was received, the reception process suffered a buffer
3375 overflow, and (when I tested it) crashed. This was caused by some obsolete
3376 code that should have been removed. The fix is to remove it!
3378 3. When running in the test harness, delay a bit after writing a bounce
3379 message to get a bit more predictability in the log output.
3381 4. Added a call to search_tidyup() just before forking a reception process. In
3382 theory, someone could use a lookup in the expansion of smtp_accept_max_
3383 per_host which, without the tidyup, could leave open a database connection.
3385 5. Added the variables $recipient_data and $sender_data which get set from a
3386 lookup success in an ACL "recipients" or "senders" condition, or a router
3387 "senders" option, similar to $domain_data and $local_part_data.
3389 6. Moved the writing of debug_print from before to after the "senders" test
3392 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3393 problems for message scanning, either using a data ACL, or using
3394 local_scan() because the Received: header was not generated till after they
3395 were called (in order to set the time as the time of reception completion).
3396 I have revised the way this works. The header is now generated after the
3397 body is received, but before the ACL or local_scan() are called. After they
3398 are run, the timestamp in the header is updated.
3404 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3405 before starting a queue runner without re-exec. This happened only when
3406 deliver_drop_privilege was set or when the Exim user was set to root. The
3407 effect of the bug was that timeouts during subsequent deliveries caused
3408 crashes instead of being properly handled. The handler is now left at its
3409 default (and expected) setting.
3411 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3412 message, again when deliver_drop_privilege is set or Exim is run as root.
3413 The bug described in (1) was not present in this case, but the tidying up
3414 of the other signals was missing. I have made the two cases consistent.
3416 3. The ignore_target_hosts setting on a manualroute router was being ignored
3417 for hosts that were looked up using the /MX notation.
3419 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3422 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3423 operated on the sender address. After changing the $sender_address to <>
3424 for the sender address verify, Exim was re-instated it as the original
3425 (before rewriting) address, but remembering that it had rewritten it, so it
3426 wasn't rewriting it again. This bug also had the effect of breaking the
3427 sender address verification caching when the sender address was rewritten.
3429 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3430 This has been changed so that if the ip literal address matches
3431 ignore_target_hosts, the router declines.
3433 7. Added expansion conditions match_domain, match_address, and match_local_
3434 part (NOT match_host).
3436 8. The placeholder for the Received: header didn't have a length field set.
3438 9. Added code to Exim itself and to exim_lock to test for a specific race
3439 condition that could lead to file corruption when using MBX delivery. The
3440 issue is with the lockfile that is created in /tmp. If this file is removed
3441 after a process has opened it but before that process has acquired a lock,
3442 there is the potential for a second process to recreate the file and also
3443 acquire a lock. This could lead to two Exim processes writing to the file
3444 at the same time. The added code performs the same test as UW imapd; it
3445 checks after acquiring the lock that its file descriptor still refers to
3446 the same named file.
3448 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3449 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3452 11. Added the smtp_active_hostname option. If used, this will typically be made
3453 to depend on the incoming interface address. Because $interface_address is
3454 not set up until the daemon has forked a reception process, error responses
3455 that can happen earlier (such as "too many connections") no longer contain
3458 12. If an expansion in a condition on a "warn" statement fails because a lookup
3459 defers, the "warn" statement is abandoned, and the next ACL statement is
3460 processed. Previously this caused the whole ACL to be aborted.
3462 13. Added the iplsearch lookup type.
3464 14. Added ident_timeout as a log selector.
3466 15. Added tls_certificate_verified as a log selector.
3468 16. Added a global option tls_require_ciphers (compare the smtp transport
3469 option of the same name). This controls incoming TLS connections.
3471 17. I finally figured out how to make tls_require_ciphers do a similar thing
3472 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3473 before starting the TLS session.
3475 18. Tabs are now shown as \t in -bP output.
3477 19. If the log selector return_path_on_delivery was set, Exim crashed when
3478 bouncing a message because it had too many Received: header lines.
3480 20. If two routers both had headers_remove settings, and the first one included
3481 a superfluous trailing colon, the final name in the first list and the
3482 first name in the second list were incorrectly joined into one item (with a
3483 colon in the middle).
3489 1. Added -C and -D options to the exinext utility, mainly to make it easier
3490 to include in the automated testing, but these could be helpful when
3491 multiple configurations are in use.
3493 2. The exinext utility was not formatting the output nicely when there was
3494 an alternate port involved in the retry record key, nor when there was a
3495 message id as well (for retries that were specific to a specific message
3496 and a specific host). It was also confused by IPv6 addresses, because of
3497 the additional colons they contain. I have fixed the IPv4 problem, and
3498 patched it up to do a reasonable job for IPv6.
3500 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3501 delivery, the log line now contains "pipelined" if PIPELINING was used.
3503 4. An SMTP transport process used to panic and die if the bind() call to set
3504 an explicit outgoing interface failed. This has been changed; it is now
3505 treated in the same way as a connect() failure.
3507 5. A reference to $sender_host_name in the part of a conditional expansion
3508 that was being skipped was still causing a DNS lookup. This no longer
3511 6. The def: expansion condition was not recognizing references to header lines
3512 that used bh_ and bheader_.
3514 7. Added the _cache feature to named lists.
3516 8. The code for checking quota_filecount in the appendfile transport was
3517 allowing one more file than it should have been.
3519 9. For compatibility with Sendmail, the command line option
3527 and sets the incoming protocol and host name (for trusted callers). The
3528 host name and its colon can be omitted when only the protocol is to be set.
3529 Note the Exim already has two private options, -pd and -ps, that refer to
3530 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3531 "s", but I don't think that's a major issue.
3533 10. A number of refactoring changes to the code, none of which should affect
3536 (a) The number of logging options was getting close to filling up the
3537 32-bit word that was used as a bit map. I have split them into two classes:
3538 those that are passed in the argument to log_write(), and those that are
3539 only ever tested independently outside of that function. These are now in
3540 separate 32-bit words, so there is plenty of room for expansion again.
3541 There is no change in the user interface or the logging behaviour.
3543 (b) When building, for example, log lines, the code previously used a
3544 macro that called string_cat() twice, in order to add two strings. This is
3545 not really sufficiently general. Furthermore, there was one instance where
3546 it was actually wrong because one of the argument was used twice, and in
3547 one call a function was used. (As it happened, calling the function twice
3548 did not affect the overall behaviour.) The macro has been replaced by a
3549 function that can join an arbitrary number of extra strings onto a growing
3552 (c) The code for expansion conditions now uses a table and a binary chop
3553 instead of a serial search (which was left over from when there were very
3554 few conditions). Also, it now recognizes conditions like "pam" even when
3555 the relevant support is not compiled in: a suitably worded error message is
3556 given if an attempt is made to use such a condition.
3558 11. Added ${time_interval:xxxxx}.
3560 12. A bug was causing one of the ddress fields not to be passed back correctly
3561 from remote delivery subprocesses. The field in question was not being
3562 subsequently used, so this caused to problems in practice.
3564 13. Added new log selectors queue_time and deliver_time.
3566 14. Might have fixed a bug in maildirsizefile handling that threw up
3567 "unexpected character" debug warnings, and recalculated the data
3568 unnecessarily. In any case, I expanded the warning message to give more
3571 15. Added the message "Restricted characters in address" to the statements in
3572 the default ACL that block characters like @ and % in local parts.
3574 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3575 Three changes have been made:
3577 (a) There was a serious bug; a negative response to MAIL caused the whole
3578 recipient domain to be cached as invalid, thereby blocking all messages
3579 to all local parts at the same domain, from all senders. This bug has
3580 been fixed. The domain is no longer cached after a negative response to
3581 MAIL if the sender used is not empty.
3583 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3586 (c) A new callout option, "use_sender" has been added for people who want
3587 the modified behaviour.
3593 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3596 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3597 indeed breaks things for older releases.
3599 3. Added additional logging to the case where there is a problem reading data
3600 from a filter that is running in a subprocess using a pipe, in order to
3601 try to track down a specific problem.
3603 4. Testing facility fudge: when running in the test harness and attempting
3604 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3605 getting "No route to host". Convert this to a timeout.
3607 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3610 6. Some OS don't have socklen_t but use size_t instead. This affects the
3611 fifth argument of getsockopt() amongst other things. This is now
3612 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3613 can be set for individual OS. I have set it for SunOS5, OSF1, and
3614 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3615 some earlier ones do not.
3617 7. Change 4.30/15 was not doing the test caselessly.
3619 8. The standard form for an IPv6 address literal was being rejected by address
3620 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3621 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3622 this, as well as the form without the "IPv6" on the front (but only when
3623 address literals are enabled, of course).
3625 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3627 10. Exim crashed if a message with an empty sender address specified by -f
3628 encountered a router with an errors_to setting. This could be provoked only
3629 by a command such as
3633 where an empty string was supplied; "<>" did not hit this bug.
3635 11. Installed PCRE release 4.5.
3637 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3638 remained set. It is now erased.
3640 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3641 times from message ids (which are base 36 rather than the normal 62).
3643 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3644 were being counted as actual protocol errors, and logged if the log
3645 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3646 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3647 and DATA following a set of rejected RCPTs do not count as protocol errors.
3648 In other words, Exim assumes they were pipelined, though this may not
3649 actually be the case. Of course, in all cases the client gets an
3650 appropriate error code.
3652 15. If a lookup fails in an ACL condition, a message about the failure may
3653 be available; it is used if testing the ACL cannot continue, because most
3654 such messages specify what the cause of the deferral is. However, some
3655 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3656 that caused an old message to be retained and used if a later statement
3657 caused a defer, replacing the real cause of the deferral.
3659 16. If an IP address had so many PTR records that the DNS lookup buffer
3660 was not large enough to hold them, Exim could crash while trying to process
3661 the truncated data. It now detects and logs this case.
3663 17. Further to 4.21/58, another change has been made: if (and only if) the
3664 first line of a message (the first header line) ends with CRLF, a bare LF
3665 in a subsequent header line has a space inserted after it, so as not to
3666 terminate the header.
3668 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3669 unnecessarily, used atoi() instead of strtol(), and didn't check the
3670 termination when getting file sizes from file names by regex.
3672 19. Completely re-implemented the support for maildirsize files, in the light
3673 of a number of problems with the previous contributed implementation
3674 (4.30/29). In particular:
3676 . If the quota is zero, the maildirsize file is maintained, but no quota is
3679 . If the maildir directory does not exist, it is created before any attempt
3680 to write a maildirsize file.
3682 . The quota value in the file is just a cache; if the quota is changed in
3683 the transport, the new value overrides.
3685 . A regular expression is available for excluding directories from the
3688 20. The autoreply transport checks the characters in options that define the
3689 message's headers; it allows continued headers, but it was checking with
3690 isspace() after an embedded newline instead of explicitly looking for a
3693 21. If all the "regular" hosts to which an address was routed had passed their
3694 expiry times, and had not reached their retry times, the address was
3695 bounced, even if fallback hosts were defined. Now Exim should go on to try
3698 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3699 equivalent code in the SMTP transport. Some hosts send humungous responses
3700 to HELO/EHLO, more than 1024 it seems.
3702 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3703 gives compiler warnings in some environments. I've now done it "properly",
3706 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3707 (because of problems with the built-in one) was declared to return uschar *
3708 instead of char *, causing compiler failure.
3710 25. Fixed a file descriptor leak when processing alias/forward files.
3712 26. Fixed a minor format string issue in dbfn.c.
3714 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3716 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3717 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3720 29. When a sender address is verified, it is cached, to save repeating the test
3721 when there is more than one recipient in a message. However, when the
3722 verification involves a callout, it is possible for different callout
3723 options to be set for different recipients. It is too complicated to keep
3724 track of this in the cache, so now Exim always runs a verification when a
3725 callout is required, relying on the callout cache for the optimization.
3726 The overhead is duplication of the address routing, but this should not be
3729 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3730 to be verified with callout=postmaster, and the main callout worked but the
3731 postmaster check failed, the verification correctly failed. However, if a
3732 subsequent RCPT command asked for sender verification *without* the
3733 postmaster check, incorrect caching caused this verification also to fail,
3736 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3737 it was not caching the DNS options (qualify_single, search_parents) that
3738 were used when the lookup failed. A subsequent lookup with different
3739 options therefore always gave the same answer, though there were cases
3740 where it should not have. (Example: a "domains = !$mx_any" option on a
3741 dnslookup router: the "domains" option is always processed without any
3742 widening, but the router might have qualify_single set.) Now Exim uses the
3743 cached value only when the same options are set.
3745 32. Added John Jetmore's "exipick" utility to the distribution.
3747 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3748 than a timeout (e.g. a certificate is required, and is not provided), an
3749 Exim server now closes the connection immediately. Previously it waited for
3750 the client to close - but if the client is SSL, it seems that they each
3751 wait for each other, leading to a delay before one of them times out.
3753 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3754 maintained 0.8.x compatibility because I don't think many are using it, and
3755 it is clearly obsolete.
3757 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3760 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3761 client certificate was expired. A simple patch fixes this, though I don't
3762 understand the full logic of why the verify callback is called multiple
3765 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3766 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3767 which causes problems with some clients (such as the Certicom SSL Plus
3768 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3769 disables the coutermeasure allowing Eudora to connect."
3771 38. Exim was not checking that a write() to a log file succeeded. This could
3772 lead to Bad Things if a log got too big, in particular if it hit a file
3773 size limit. Exim now panics and dies if it cannot write to a log file, just
3774 as it does if it cannot open a log file.
3776 39. Modified OS/Makefile-Linux so that it now contains
3778 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3780 The two -D definitions ensure that Exim is compiled with large file
3781 support, which makes it possible to handle log files that are bigger than
3784 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3785 instance) a domain was checked against a named list that involved a lookup,
3786 causing $domain_data to be set, then another domain was checked against the
3787 same list, then the first domain was re-checked, the value of $domain_data
3788 after the final check could be wrong. In particular, if the second check
3789 failed, it could be set empty. This bug probably also applied to
3792 41. The strip_trailing_dot option was not being applied to the address given
3793 with the -f command-line option.
3795 42. The code for reading a message's header from the spool was incrementing
3796 $received_count, but never initializing it. This meant that the value was
3797 incorrect (doubled) while delivering a message in the same process in which
3798 it was received. In the most common configuration of Exim, this never
3799 happens - a fresh exec is done - but it can happen when
3800 deliver_drop_privilege is set.
3802 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3803 it now includes up to 150 characters of the unexpected data in the log
3806 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3807 and building data strings. The size of both of these buffers was 10 000
3808 bytes - far larger than anybody would *ever* want, thought I. Needless to
3809 say, somebody hit the limit. I have increased the maximum line length to
3810 20 000 and the maximum data length of concatenated lines to 100 000. I have
3811 also fixed two bugs, because there was no checking on these buffers. Tsk,
3812 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3813 buffer is too small.
3815 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3816 lsearch lookups. Now it does.
3818 46. When parsing a route_list item in a manualroute router, a fixed-length
3819 buffer was used for the list of hosts. I made this 1024 bytes long,
3820 thinking that nobody would ever have a list of hosts that long. Wrong.
3821 Somebody had a whole pile of complicated expansion conditions, and the
3822 string was silently truncated, leading to an expansion error. It turns out
3823 that it is easier to change to an unlimited length (owing to other changes
3824 that have happened since this code was originally written) than to build
3825 structure for giving a limitation error. The length of the item that
3826 expands into the list of hosts is now unlimited.
3828 47. The lsearch lookup could not handle data where the length of text line was
3829 more than 4095 characters. Such lines were truncated, leading to shortened
3830 data being returned. It should now handle lines of any length.
3832 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3833 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3836 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3837 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3838 no longer happens in comment lines. A list of replacements is now placed
3839 at the head of all of the source files, except those whose only change is
3840 to replace PERL_COMMAND in the very first #! line.
3842 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3843 messages on the queue, with a bottom-up merge sort, using code contributed
3844 by Michael Haardt. This should make operations like -bp somewhat faster on
3845 large queues. It won't affect queue runners, except when queue_run_in_order
3848 51. Installed eximstats 1.31 in the distribution.
3850 52. Added support for SRV lookups to the dnslookup router.
3852 53. If an ACL referred to $message_body or $message_body_end, the value was not
3853 reset for any messages that followed in the same SMTP session.
3855 54. The store-handling optimization for building very long strings was not
3856 differentiating between the different store pools. I don't think this
3857 actually made any difference in practice, but I've tidied it.
3859 55. While running the routers to verify a sender address, $sender_address
3860 was still set to the sender address. This is wrong, because when routing to
3861 send a bounce to the sender, it would be empty. Therefore, I have changed
3862 it so that, while verifying a sender address, $sender_address is set to <>.
3863 (There is no change to what happens when verifying a recipient address.)
3865 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3866 target A or AAAA records (if not already returned) without resetting the
3867 qualify_single or search_parents options of the DNS resolver. These are
3868 inappropriate in this case because the targets of MX and SRV records must
3869 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3870 target that, when qualified, matched something in the local domain. These
3871 two options are now turned off when doing these lookups.
3873 57. It seems that at least some releases of Reiserfs (which does not have the
3874 concept of a fixed number of inodes) returns zero and not -1 for the
3875 number of available inodes. This interacted badly with check_spool_inodes,
3876 which assumed that -1 was the "no such thing" setting. What I have done is
3877 to check that the total number of inodes is greater than zero before doing
3878 the test of how many are available.
3880 58. When a "warn" ACL statement has a log_message modifier, the message is
3881 remembered, and not repeated. This is to avoid a lot of repetition when a
3882 message has many recipients that cause the same warning to be written.
3883 Howewer, Exim was preserving the list of already written lines for an
3884 entire SMTP session, which doesn't seem right. The memory is now reset if a
3885 new message is started.
3887 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3888 debugging output unless log_rewrite was also set.
3890 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3891 of (int)(handle) when we know that handle contains (void *)(-1).
3893 61. The Exim daemon panic-logs an error return when it closes the incoming
3894 connection. However "connection reset by peer" seems to be common, and
3895 isn't really an error worthy of noting specially, so that particular error
3898 62. When Exim is trying to find all the local interfaces, it used to panic and
3899 die if the ioctl to get the interface flags failed. However, it seems that
3900 on at least one OS (Solaris 9) it is possible to have an interface that is
3901 included in the list of interfaces, but for which you get a failure error
3902 for this call. This happens when the interface is not "plumbed" into a
3903 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3904 failure of the "get flags" call assumes that the interface is down.
3906 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3907 makes life easier for people who are doing arithmetic on fields extracted
3908 from dates, where you often get leading zeros that should not be
3909 interpreted as octal.
3911 64. Added qualify_domain to the redirect router, to override the global
3914 65. If a pathologically long header line contained very many addresses (the
3915 report of this problem mentioned 10 000) and each of them was rewritten,
3916 Exim could use up a very large amount of memory. (It kept on making new
3917 copies of the header line as it rewrote, and never released the old ones.)
3918 At the expense of a bit more processing, the header rewriting function has
3919 been changed so that it no longer eats memory in this way.
3921 66. The generation of the Received: header has been moved from the time that a
3922 message starts to be received, to the time that it finishes. The timestamp
3923 in the Received: header should now be very close to that of the <= log
3924 line. There are two side-effects of this change:
3926 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3927 logged header lines no longer include the local Received: line, because
3928 it has not yet been created. The same applies to a copy of the message
3929 that is returned to a non-SMTP sender when a message is rejected.
3931 (b) When a filter file is tested using -bf, no additional Received: header
3932 is added to the test message. After some thought, I decided that this
3935 This change does not affect the value of $received_for. It is still set
3936 after address rewriting, but before local_scan() is called.
3938 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3940 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3941 gave an unhelpful panic error message, and a defer error. I have managed to
3942 change this behaviour so that it now rejects any supplied certificate,
3943 which seems right, as the list of acceptable certificates is empty.
3945 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3946 gave an unhelpful defer error. I have not managed to make this reject any
3947 supplied certificates, but the error message it gives is "no certificate
3948 supplied", which is not helpful.
3950 70. exigrep's output now also includes lines that are not associated with any
3951 message, but which match the given pattern. Implemented by a patch from
3952 Martin Sluka, which also tidied up the Perl a bit.
3954 71. Recipient callout verification, like sender verification, was using <> in
3955 the MAIL FROM command. This isn't really the right thing, since the actual
3956 sender may affect whether the remote host accepts the recipient or not. I
3957 have changed it to use the actual sender in the callout; this means that
3958 the cache record is now keyed on a recipient/sender pair, not just the
3959 recipient address. There doesn't seem to be a real danger of callout loops,
3960 since a callout by the remote host to check the sender would use <>.
3961 [SEE ABOVE: changed after hitting problems.]
3963 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3964 temporary errors. However, in the case of such a code being given after
3965 the end of a data transmission (i.e. after ".") Exim was failing to write
3966 a retry record for the message. (Yes, there was some broken host that was
3967 actually sending 8xx at this point.)
3969 73. An unknown lookup type in a host list could cause Exim to panic-die when
3970 the list was checked. (An example that provoked this was putting <; in the
3971 middle of a list instead of at the start.) If this happened during a DATA
3972 ACL check, a -D file could be left lying around. This kind of configuration
3973 error no longer causes Exim to die; instead it causes a defer errror. The
3974 incident is still logged to the main and panic logs.
3976 74. Buglet left over from Exim 3 conversion. The message "too many messages
3977 in one connection" was written to the rejectlog but not the mainlog, except
3978 when address rewriting (yes!) was being logged.
3980 75. Added write_rejectlog option.
3982 76. When a system filter was run not as root (that is, when system_filter_user
3983 was set), the values of the $n variables were not being returned to the
3984 main process; thus, they were not subsequently available in the $sn
3987 77. Added +return_path_on_delivery log selector.
3989 78. A connection timeout was being treated differently from recipients deferred
3990 when testing hosts_max_try with a message that was older than the host's
3991 retry timeout. (The host should not be counted, thus allowing all hosts to
3992 be tried at least once before bouncing.) This may have been the cause of an
3993 occasionally reported bug whereby a message would remain on the queue
3994 longer than the retry timeout, but would be bounced if a delivery was
3995 forced. I say "may" because I never totally pinned down the problem;
3996 setting up timeout/retry tests is difficult. See also the next item.
3998 79. The ultimate address timeout was not being applied to errors that involved
3999 a combination of host plus message (for example, a timeout on a MAIL
4000 command). When an address resolved to a number of possible hosts, and they
4001 were not all tried for each delivery (e.g. because of hosts_max_try), a
4002 message could remain on the queue longer than the retry timeout.
4004 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4007 81. Fixed an obscure SMTP outgoing bug which required at least the following
4008 conditions: (a) there was another message waiting for the same server;
4009 (b) the server returned 5xx to all RCPT commands in the first message so
4010 that the message was not completed; (c) the server dropped the connection
4011 or gave a negative response to the RSET that Exim sends to abort the
4012 transaction. The observed case was a dropped connection after DATA that had
4013 been sent in pipelining mode. That is, the server had advertised PIPELINING
4014 but was not implementing it correctly. The effect of the bug was incorrect
4015 behaviour, such as trying another host, and this could lead to a crash.
4021 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4022 and daemon.c were passed as pointers to ints; they should have been
4023 pointers to socklen_t variables (which are typically unsigned ints).
4025 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4028 3. Fixed a really odd bug that affected only the testing scheme; patching a
4029 certain fixed string in the binary changed the value of another string that
4030 happened to be identical to the end of the original first string.
4032 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4033 name", it returns that address as the IP address. On some operating
4034 systems (e.g. Solaris), it also passes back the IP address string as the
4035 "host name". However, on others (e.g. Linux), it passes back an empty
4036 string. Exim wasn't checking for this, and was changing the host name to an
4037 empty string, assuming it had been canonicized.
4039 5. Although rare, it is permitted to have more than one PTR record for a given
4040 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4041 all the names associated with an address, because they do in Solaris.
4042 However, it seems that they do not in Linux for data that comes from the
4043 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4044 I found this out when I moved to a new Linux workstation and tried to run
4045 the Exim test suite.
4047 To get round this problem I have changed the code so that it now does its
4048 own call to the DNS to look up PTR records when searching for a host name.
4049 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4050 addresses that are only in /etc/hosts are still found.
4052 This behaviour is, however, controlled by an option called host_lookup_
4053 order, which defaults to "bydns:byaddr". If people want to use the other
4054 order, or indeed, just use one or the other means of lookup, they can
4055 specify it in this variable.
4057 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4058 some operating systems, this comes back from gethostbyaddr() as an empty
4059 string, and this is what Exim used to test for. However, it seems that in
4060 other systems, "." is yielded. Exim now tests for this case too.
4062 7. The values of check_spool_space and check_log_space are now held internally
4063 as a number of kilobytes instead of an absolute number of bytes. If a
4064 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4065 kilobyte. This means that much larger values can be stored.
4067 8. Exim monitor: an attempt to get the action menu when not actually pointing
4068 at a message produces an empty menu entitled "No message selected". This
4069 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4070 no entries in it ("Shell widget menu has zero width and/or height"). So I
4071 have added a single, blank menu entry in this case.
4073 9. Added ${quote_local_part.
4075 10. MIME decoding is now applied to the contents of Subject: header lines when
4078 11. Now that a reference to $sender_host_address automatically causes a reverse
4079 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4080 host lookup before query-style lookups in lists that might use this
4081 variable. This has therefore been abolished, and the "net-" prefix is no
4082 longer necessary for query-style lookups.
4084 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4085 have been a typo for LFLAGS, so it has been changed.
4087 13. The install script calls Exim with "-C /dev/null" in order to find the
4088 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4089 to be output. Howeve, since Exim outputs its version number before the
4090 error, it didn't break the script. It just looked ugly. I fixed this by
4091 always allowing "-C /dev/null" if the caller is root.
4093 14. Ignore overlarge ACL variable number when reading spool file - insurance
4094 against a later release with more variables having written the file.
4096 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4097 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4098 this, as well as the form without the "IPv6" on the front.
4100 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4101 OS/Makefile-Darwin file.
4103 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4104 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4106 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4107 variable. However, some OS use ints for inodes. I've added cast to long int
4108 to get rid of the compiler warning.
4110 19. I had forgotten to lock out "/../" in configuration file names when
4111 ALT_CONFIG_PREFIX was set.
4113 20. Routers used for verification do not need to specify transports. However,
4114 if such a router generated a host list, and callout was configured, Exim
4115 crashed, because it could not find a port number from the (non-existent)
4116 transport. It now assumes port 25 in this circumstance.
4118 21. Added the -t option to exigrep.
4120 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4121 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4122 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4123 them is set, LOOKUP_LSEARCH is forced.
4125 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4127 24. Added sender and host information to the "rejected by local_scan()" log
4128 line; previously there was no indication of these.
4130 25. Added .include_if_exists.
4132 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4133 Linux. It turns out that not all file systems support this. Apparently some
4134 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4135 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4136 not supported on the file descriptor, is now ignored when Exim is trying to
4137 sync a directory. This applies only to Linux.
4139 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4141 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4142 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4145 29. Added support for maildirsize files from supplied patch (modified a bit).
4147 30. The use of :fail: followed by an empty string could lead Exim to respond to
4148 sender verification failures with (e.g.):
4150 550 Verification failed for <xxx>
4151 550 Sender verify failed
4153 where the first response line was missing the '-' that indicates it is not
4154 the final line of the response.
4156 31. The loop for finding the name of the user that called Exim had a hardwired
4157 limit of 10; it now uses the value of finduser_retries, which is used for
4158 all other user lookups.
4160 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4163 33. Exim was neglecting to zero errno before one call of strtol() when
4164 expanding a string and expecting an integer value. On some systems this
4165 resulted in spurious "integer overflow" errors. Also, it was casting the
4166 result into an int without checking.
4168 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4169 did not work. The code looks as if it has *never* worked, though it appears
4170 to have been documented since at least releast 1.62. I have made it work.
4172 35. The "timeout_DNS" error in retry rules, also documented since at least
4173 1.62, also never worked. As it isn't clear exactly what this means, and
4174 clearly it isn't a major issue, I have abolished the feature by treating it
4175 as "timeout", and writing a warning to the main and panic logs.
4177 36. The display of retry rules for -brt wasn't always showing the error code
4180 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4181 timeout_connect_A, timeout_connect_MX.
4183 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4184 to the empty sender.
4186 39. The daemon was not analysing the content of -oX till after it had closed
4187 stderr and disconnected from the controlling terminal. This meant that any
4188 syntax errors were only noted on the panic log, and the return code from
4189 the command was 0. By re-arranging the code a little, I've made the
4190 decoding happen first, so such errors now appear on stderr, and the return
4191 code is 1. However, the actual setting up of the sockets still happens in
4192 the disconnected process, so errors there are still only recorded on the
4195 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4196 connections (as happens on some IP stacks) was logged at start up time as
4197 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4198 it from "IPv6 and IPv4", which means that two separate sockets are being
4201 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4202 says whether AF_INET or AF_INET6 was passed as an argument.
4204 42. Exiwhat output was messed up when time zones were included in log
4207 43. Exiwhat now gives more information about the daemon's listening ports,
4208 and whether -tls-on-connect was used.
4210 44. The "port" option of the smtp transport is now expanded.
4212 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4213 silently ignored. Now an error message is written to the main and panic
4216 46. There's a new ACL modifier called "logwrite" which writes to a log file
4217 as soon as it is encountered.
4219 47. Added $local_user_uid and $local_user_gid at routing time.
4221 48. Exim crashed when trying to verify a sender address that was being
4224 49. Exim was recognizing only a space character after ".include". It now also
4225 recognizes a tab character.
4227 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4228 extracting the relevant information from the specification. The man page no
4229 longer contains scrambled data for the -d option, and I've added a section
4230 at the front about calling Exim under different names.
4232 51. Added "extra_headers" argument to the "mail" command in filter files.
4234 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4237 53. Installed eximstats 1.29.
4239 54. Added transport_filter_timeout as a generic transport option.
4241 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4242 Cc: header lines. This was required by RFC 822, but it not required by RFC
4245 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4246 incoming messages that did not have them. Now it does so only if the
4247 message originates locally, that is, if there is no associated remote host
4248 address. When Resent- header lines are present, this applies to the Resent-
4249 lines rather than the non-Resent- lines.
4251 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4252 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4254 58. Messages for configuration errors now include the name of the main
4255 configuration file - useful now that there may be more than one file in a
4256 list (.included file names were always shown).
4258 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4259 for those rare installations that do not start the daemon as root or run it
4260 setuid root. I've cut out the call to initgroups() if the daemon is not
4263 60. The Exim user and group can now be bound into the binary as text strings
4264 that are looked up at the start of Exim's processing.
4266 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4268 62. Added $mailstore_basename variable.
4270 63. Installed patch to sieve.c from Michael Haardt.
4272 64. When Exim failed to open the panic log after failing to open the main log,
4273 the original message it was trying to log was written to stderr and debug
4274 output, but if they were not available (the usual case in production), it
4275 was lost. Now it is written to syslog before the two lines that record the
4276 failures to open the logs.
4278 65. Users' Exim filters run in subprocesses under the user's uid. It is
4279 possible for a "deliver" command or an alias in a "personal" command to
4280 provoke an address rewrite. If logging of address rewriting is configured,
4281 this fails because the process is not running as root or exim. There may be
4282 a better way of dealing with this, but for the moment (because 4.30 needs
4283 to be released), I have disabled address rewrite logging when running a
4284 filter in a non-root, non-exim process.
4290 1. The buildconfig auxiliary program wasn't quoting the value set for
4291 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4292 not defined. This bug was masked in 4.22 by the effect that was fixed in
4295 2. Some messages that were rejected after a message id was allocated were
4296 shown as "incomplete" by exigrep. It no longer does this for messages that
4297 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4299 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4300 have allow_domain_literals set, the ID did not get logged in the <= line.
4301 Domain literals are now always recognized in Message-ID: header lines.
4303 4. The first argument for a ${extract expansion item is the key name or field
4304 number. Leading and trailing spaces in this item were not being ignored,
4305 causing some misleading effects.
4307 5. When deliver_drop_privilege was set, single queue runner processes started
4308 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4309 same command in the process it spins off) were not dropping privilege.
4311 6. When the daemon running as "exim" started a queue runner, it always
4312 re-executed Exim in the spun-off process. This is a waste of effort when
4313 deliver_drop_privilege is set. The new process now just calls the
4314 queue-runner function directly.
4320 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4323 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4324 ignored. Though the use of -oP was forcing the writing of a pid file, it
4325 was always written to the default place.
4327 3. If the message "no IP address found for host xxxx" is generated during
4328 incoming verification, it is now followed by identification of the incoming
4329 connection (so you can more easily find what provoked it).
4331 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4333 5. Added some features to "harden" Exim a bit more against certain attacks:
4335 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4336 be put in Local/Makefile. This is like the never_users runtime option,
4337 but it cannot be overridden. The default setting is "root".
4339 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4340 prefix string with which any file named in a -C command line option
4343 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4344 is retained for -C and -D only if the caller of Exim is root. Without
4345 it, the exim user may also use -C and -D and retain privilege.
4347 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4348 command line option is disabled.
4350 6. Macro names set by the -D option must start with an upper case letter, just
4351 like macro names defined in the configuration file.
4353 7. Added "dereference=" facility to LDAP.
4355 8. Two instances of the typo "uknown" in the source files are fixed.
4357 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4358 the Configure-Makefile script screwed up while processing it.
4360 10. Incorporated PCRE 4.4.
4362 11. The SMTP synchronization check was not operating right at the start of an
4363 SMTP session. For example, it could not catch a HELO sent before the client
4364 waited for the greeting. There is now a check for outstanding input at the
4365 point when the greeting is written. Because of the duplex, asynchronous
4366 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4367 way, but not yet received, when the check is performed.
4369 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4370 on TCP/IP sockets, because this apparently causes some broken clients to
4373 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4374 unchanged) from the Cygwin maintainer.
4376 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4377 was supported instead of testing for mbx. Effectively a typo.
4379 15. The spa authenticator server code was not checking that the input it
4380 received was valid base64.
4382 16. The debug output line for the "set" modifier in ACLs was not showing the
4383 name of the variable that was being set.
4385 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4387 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4388 lookup on its own. Something else was needed to trigger the lookup. For
4389 example, a match in host_lookup or the need for a host name in a host list.
4390 Now, if $sender_host_name is referenced and the host name has not yet been
4391 looked up, a lookup is performed. If the lookup fails, the variable remains
4392 empty, and $host_lookup_failed is set to "1".
4394 19. Added "eqi" as a case-independent comparison operator.
4396 20. The saslauthd authentication condition could segfault if neither service
4397 nor realm was specified.
4399 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4400 error message that was logged was misleading, and incoming SMTP
4401 connections were dropped. The message is now more accurate, and temporary
4402 errors are given to SMTP connections.
4404 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4405 (except RSET) with a 421 error, until QUIT is received. However, it was
4406 failing to send a response to QUIT.
4408 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4409 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4410 ACL is now run after the helo_try_verify_hosts code.
4412 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4413 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4414 case-independent, so other case variants are also recognized). Apparently
4415 some systems use these upper case variants.
4417 25. If more than two messages were waiting for the same host, and a transport
4418 filter was specified for the transport, Exim sent two messages over the
4419 same TCP/IP connection, and then failed with "socket operation on non-
4420 socket" when it tried to send the third.
4422 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4424 27. The extern definition of crypt16() in expand.c was not being excluded when
4425 the OS had its own crypt16() function.
4427 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4428 as a preferred synonym for return_size_limit, both as an option and as an
4431 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4433 30. Changed the default configuration ACL to relax the local part checking rule
4434 for addresses that are not in any local domains. For these addresses,
4435 slashes and pipe symbols are allowed within local parts, but the sequence
4436 /../ is explicitly forbidden.
4438 31. SPA server authentication was not clearing the challenge buffer before
4441 32. log_message in a "warn" ACL statement was writing to the reject log as
4442 well as to the main log, which contradicts the documentation and doesn't
4443 seem right (because no rejection is happening). So I have stopped it.
4445 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4446 However, I am unable to do any testing of this.
4448 34. Fixed an infelicity in the appendfile transport. When checking directories
4449 for a mailbox, to see if any needed to be created, it was accidentally
4450 using path names with one or more superfluous leading slashes; tracing
4451 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4453 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4454 contents are added to the log line that is written for every discarded
4455 recipient. (Previously a log_message setting was ignored.)
4457 36. The ${quote: operator now quotes the string if it is empty.
4459 37. The install script runs exim in order to find its version number. If for
4460 some reason other than non-existence or emptiness, which it checks, it
4461 could not run './exim', it was installing it with an empty version number,
4462 i.e. as "exim-". This error state is now caught, and the installation is
4465 38. An argument was missing from the function that creates an error message
4466 when Exim fails to connect to the socket for saslauthd authentication.
4467 This could cause Exim to crash, or give a corrupted message.
4469 39. Added isip, isip4, and isip6 to ${if conditions.
4471 40. The ACL variables $acl_xx are now saved with the message, and can be
4472 accessed later in routers, transports, and filters.
4474 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4475 strings in the file are not string-expanded.
4477 42. If a MAIL command specified a SIZE value that was too large to fit into an
4478 int variable, the check against message_size_limit failed. Such values are
4479 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4480 day this will have to be increased, but I don't think I want to be around
4481 when emails are that large.
4488 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4489 iconv() is not standard in FreeBSD.
4491 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4492 IPv6 enabled. The observed symptom was a segmentation fault on return from
4493 the function os_common_find_running_interfaces() in src/os.c.
4495 3. In the check_special_case() function in daemon.c I had used "errno" as an
4496 argument name, which causes warnings on some systems. This was basically a
4497 typo, since it was named "eno" in the comments!
4499 4. The code that waits for the clock to tick (at a resolution of some fraction
4500 of a second) so as to ensure message-id uniqueness was always waiting for
4501 at least one whole tick, when it could have waited for less. [This is
4502 almost certainly not relevant at current processor speeds, where it is
4503 unlikely to ever wait at all. But we try to future-proof.]
4505 5. The function that sleeps for a time interval that includes fractions of a
4506 second contained a race. It did not block SIGALRM between setting the
4507 timer, and suspending (a couple of lines later). If the interval was short
4508 and the sigsuspend() was delayed until after it had expired, the suspension
4509 never ended. On busy systems this could lead to processes getting stuck for
4512 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4513 process, before it forks any delivery processes. The open lookup caching
4514 mechanism meant that the open file or database connection was passed into
4515 the delivery process. The problem was that delivery processes always tidy
4516 up cached lookup data. This could cause a problem for the next delivery
4517 process started by the queue runner, because the external queue runner
4518 process does not know about the closure. So the next delivery process
4519 still has data in the lookup cache. In the case of a file lookup, there was
4520 no problem because closing a file descriptor in a subprocess doesn't affect
4521 the parent. However, if the lookup was caching a connection to a database,
4522 the connection was closed, and the second delivery process was likely to
4523 see errors such as "PGSQL: query failed: server closed the connection
4524 unexpectedly". The problem has been fixed by closing all cached lookups
4525 in a queue runner before running a delivery process.
4527 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4528 seem to have the "const" qualifier which it has on other OS. I've
4531 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4532 *of the same type* (client or server) with the same public name that an
4533 error should be diagnosed.
4535 9. When Exim looked up a host name for an IP address, but failed to find the
4536 original IP address when looking up the host name (a safety check), it
4537 output the message "<ip address> does not match any IP for NULL", which was
4538 confusing, to say the least. The bug was that the host name should have
4539 appeared instead of "NULL".
4541 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4542 user that is built into the binary, and the -C or -D options is used, root
4543 privilege is dropped before the configuration file is read. In addition,
4544 logging is switched to stderr instead of the normal log files. If the
4545 configuration then re-defines the Exim user, the unprivileged environment
4546 is probably not what is expected, so Exim logs a panic warning message (but
4549 However, if deliver_drop_privilege is set, the unprivileged state may well
4550 be exactly what is intended, so the warning has been cut out in that case,
4551 and Exim is allowed to try to write to its normal log files.
4557 1. smtp_return_error_details was not giving details for temporary sender
4558 or receiver verification errors.
4560 2. Diagnose a configuration error if two authenticators have the same public
4563 3. Exim used not to create the message log file for a message until the first
4564 delivery attempt. This could be confusing when incoming messages were held
4565 for policy or load reasons. The message log file is now created at the time
4566 the message is received, and an initial "Received" line is written to it.
4568 4. The automatically generated man page for command line options had a minor
4569 bug that caused no ill effects; however, a more serious problem was that
4570 the procedure for building the man page automatically didn't always
4571 operate. Consequently, release 4.20 contains an out-of-date version. This
4572 shouldn't happen again.
4574 5. When building Exim with embedded Perl support, the script that builds the
4575 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4576 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4578 6. The freeze_tell option was not being used for messages that were frozen on
4579 arrival, either by an ACL or by local_scan().
4581 7. Added the smtp_incomplete_transaction log selector.
4583 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4584 was accepting AUTH without a new EHLO.
4586 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4587 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4590 10. Logging of TCP/IP connections (when configured) now happens in the main
4591 daemon process instead of the child process, so that the TCP/IP connection
4592 count is more accurate (but it can never be perfect).
4594 11. The use of "drop" in a nested ACL was not being handled correctly in the
4595 outer ACL. Now, if condition failure induced by the nested "drop" causes
4596 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4597 or "require"), the connection is dropped.
4599 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4600 that yield "discard" can now be used with an "accept" or a "discard" verb,
4601 but an error is generated for any others (because I can't see a useful way
4602 to define what should happen).
4604 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4605 are now processed in the same way as lines in the Exim configuration file.
4606 In particular, continuation lines are supported.
4608 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4610 15. Added -ti meaning -t -i.
4612 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4613 domains, and warn by logging if others are found.
4615 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4616 returned from GIFCONF when Exim is trying to find the list of interfaces on
4617 a host. The code in os.c has been modified to copy each ifreq to an aligned
4618 structure in all cases.
4620 Also, in some cases, the returned ifreq's were being copied to a 'struct
4621 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4622 means the last couple of bytes of an IPv6 address could be chopped if the
4623 ifreq contained only a normal sockaddr (14 bytes storage).
4625 18. Named domain lists were not supported in the hosts_treat_as_local option.
4626 An entry such as +xxxx was not recognized, and was treated as a literal
4629 19. Ensure that header lines added by a DATA ACL are included in the reject log
4630 if the ACL subsequently rejects the message.
4632 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4633 MD5 (which is deprecated).
4635 21. When testing a filter file using -bf, Exim was writing a message when it
4636 took the sender from a "From " line in the message, but it was not doing so
4637 when it took $return_path from a Return-Path: header line. It now does.
4639 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4640 with a valid header line field name (a series of printing characters
4641 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4643 23. Changed "disc" in the source to "disk" to conform to the documentation and
4644 the book and for uniformity.
4646 24. Ignore Sendmail's -Ooption=value command line item.
4648 25. When execve() failed while trying to run a command in a pipe transport,
4649 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4650 could be confused with a return value of 69 from the command itself. This
4651 has been changed to 127, the value the shell returns if it is asked to run
4652 a non-existent command. The wording for the related log line suggests a
4653 non-existent command as the problem.
4655 26. If received_header_text expands to an empty string, do not add a Received:
4656 header line to the message. (Well, it adds a token one on the spool, but
4657 marks it "old" so that it doesn't get used or transmitted.)
4659 27. Installed eximstats 1.28 (addition of -nt option).
4661 28. There was no check for failure on the call to getsockname() in the daemon
4662 code. This can fail if there is a shortage of resources on the system, with
4663 ENOMEM, for example. A temporary error is now given on failure.
4665 29. Contrary to the C standard, it seems that in some environments, the
4666 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4667 program. Exim now does this explicitly; it affects the formatting of
4668 timestamps using strftime().
4670 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4671 complaints. I've now initialized all the variables, to avoid this.
4673 32. Header lines added by a system filter were not being "seen" during
4674 transport-time rewrites.
4676 33. The info_callback() function passed to OpenSSL is set up with type void
4677 (*)(SSL *, int, int), as described somewhere. However, when calling the
4678 function (actually a macro) that sets it up, the type void(*)() is
4679 expected. I've put in a cast to prevent warnings from picky compilers.
4681 34. If a DNS black list lookup found a CNAME record, but there were no A
4682 records associated with the domain it pointed at, Exim crashed.
4684 35. If a DNS black list lookup returned more than one A record, Exim ignored
4685 all but the first. It now scans all returned addresses if a particular IP
4686 value is being sought. In this situation, the contents of the
4687 $dnslist_value variable are a list of all the addresses, separated by a
4690 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4691 to accept a host name and all its aliases if the forward lookup for any of
4692 them yielded the IP address of the incoming connection. Now it accepts only
4693 those names whose forward lookup yields the correct IP address. Any other
4694 names are discarded. This closes a loophole whereby a rogue DNS
4695 administrator could create reverse DNS records to break through a
4696 wildcarded host restriction in an ACL.
4698 37. If a user filter or a system filter that ran in a subprocess used any of
4699 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4700 the wrong values were passed to the pipe command ($thisaddress had the
4701 value of $0, $0 had the value of $1, etc). This bug was introduced by
4702 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4704 38. Improved the line breaking for long SMTP error messages from ACLs.
4705 Previously, if there was no break point between 40 and 75 characters, Exim
4706 left the rest of the message alone. Two changes have been made: (a) I've
4707 reduced the minimum length to 35 characters; (b) if it can't find a break
4708 point between 35 and 75 characters, it looks ahead and uses the first one
4709 that it finds. This may give the occasional overlong line, but at least the
4710 remaining text gets split now.
4712 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4713 file descriptors might be low, and that setting 1000 would always raise it.
4714 It turns out that in some environments, the limit is already over 1000 and
4715 that lowering it causes trouble. So now Exim takes care not to decrease it.
4717 40. When delivering a message, the value of $return_path is set to $sender_
4718 address at the start of routing (routers may change the value). By an
4719 oversight, this default was not being set up when an address was tested by
4720 -bt or -bv, which affected the outcome if any router or filter referred to
4723 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4724 log only when "message" or "log_message" are set. However, if one of the
4725 conditions was an address verification, or a call to a nested ACL, the
4726 messages generated by the underlying test were being passed through. This
4727 no longer happens. The underlying message is available in $acl_verify_
4728 message for both "message" and "log_message" expansions, so it can be
4729 passed through if needed.
4731 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4732 new expansion $bh_ to give the encoded byte string without charset
4733 translation. Translation happens only if iconv() is available; HAVE_ICONV
4734 indicates this at build time. HEADERS_CHARSET gives the charset to
4735 translate to; headers_charset can change it in the configuration, and
4736 "headers charset" can change it in an individual filter file.
4738 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4739 that creates RFC 2047 encoded "words" labels them as that charset instead
4740 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4741 expansion operator; (ii) when Exim creates a From: line for a local
4742 message; (iii) when a header line is rewritten to include a "phrase" part.
4744 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4745 buggy, causing it to skip the first lines of messages whose message ID
4746 ended in 'D'. This would not have bitten before Exim release 4.14, because
4747 message IDs were unlikely to end in 'D' before then. The effect was to have
4748 incorrect size information for certain domains.
4750 45. #include "config.h" was missing at the start of the crypt16.c module. This
4751 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4754 46. If there was a timeout during a "random" callout check, Exim treated it as
4755 a failure of the random address, and carried on sending RSET and the real
4756 address. If the delay was just some slowness somewhere, the response to the
4757 original RCPT would be taken as a response to RSET and so on, causing
4758 mayhem of various kinds.
4760 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4761 when I implemented it. It didn't allow for the fact that some option values
4762 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4763 the right test for positive values.
4765 48. Domain names in DNS records are case-independent. Exim always looks them up
4766 in lower case. Some resolvers return domain names in exactly the case they
4767 appear in the zone file, that is, they may contain uppercase letters. Not
4768 all resolvers do this - some return always lower case. Exim was treating a
4769 change of case by a resolver as a change of domain, similar to a widening
4770 of a domain abbreviation. This triggered its re-routing code and so it was
4771 trying to route what was effectively the same domain again. This normally
4772 caused routing to fail (because the router wouldn't handle the domain
4773 twice). Now Exim checks for this case specially, and just changes the
4774 casing of the domain that it ultimately uses when it transmits the message
4777 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4780 50. If a filter generated a file delivery with a non-absolute name (possible if
4781 no home directory exists for the router), the forbid_file option was not
4784 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4785 the existing equality matching.
4787 52. Exim was using ints instead of ino_t variables in some places where it was
4788 dealing with inode numbers.
4790 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4791 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4792 called TMPDIR, and if it finds it is different, it changes its value.
4794 54. The smtp_printf() function is now made available to local_scan() so
4795 additional output lines can be written before returning. There is also an
4796 smtp_fflush() function to enable the detection of a dropped connection.
4797 The variables smtp_input and smtp_batched_input are exported to
4800 55. Changed the default runtime configuration: the message "Unknown user"
4801 has been removed from the ACL, and instead placed on the localuser router,
4802 using the cannot_route_message feature. This means that any verification
4803 failures that generate their own messages won't get overridden. Similarly,
4804 the "Unrouteable address" message that was in the ACL for unverifiable
4805 relay addresses has also been removed.
4807 56. Added hosts_avoid_esmtp to the smtp transport.
4809 57. The exicyclog script was not checking for the esoteric option
4810 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4811 will work only if exicyclog is run under the appropriate euid.
4813 58. Following a discussion on the list, the rules by which Exim recognises line
4814 endings on incoming messages have been changed. The -dropcr and drop_cr
4815 options are now no-ops, retained only for backwards compatibility. The
4816 following line terminators are recognized: LF CRLF CR. However, special
4817 processing applies to CR:
4819 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4820 nor a local message in the state where . is a terminator.
4822 (ii) If a bare CR is encountered in a header line, an extra space is added
4823 after the line terminator so as not to end the header. The reasoning
4824 behind this is that bare CRs in header lines are most likely either
4825 to be mistakes, or people trying to play silly games.
4827 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4828 was being incorrectly given as 18 bytes larger than it should have been.
4829 This is a VOB (very old bug).
4831 60. This may never have affected anything current, but just in case it has:
4832 When the local host is found other than at the start of a list of hosts,
4833 the local host, those with the same MX, and any that follow, are discarded.
4834 When the list in question was part of a longer list of hosts, the following
4835 hosts (not currently being processed) were also being discarded. This no
4836 longer happens. I'm not sure if this situation could ever has previously
4839 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4842 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4843 header. This is something that is recommended in a new Internet Draft, and
4844 is something that is documented as being done by Sendmail. There are two
4845 possible values. For messages generated by the autoreply transport, Exim
4848 Auto-Submitted: auto-replied
4850 whereas for all other generated messages (e.g. bounces) it adds
4852 Auto-Submitted: auto-generated
4854 63. The "personal" condition in filters now includes a test for the
4855 Auto-Submitted: header. If it contains the string "auto-" the message it
4856 not considered personal.
4858 64. Added rcpt_include_affixes as a generic transport option.
4860 65. Added queue_only_override (default true).
4862 66. Added the syslog_duplication option.
4864 67. If what should have been the first header line of a message consisted of
4865 a space followed by a colon, Exim was mis-interpreting it as a header line.
4866 It isn't of course - it is syntactically invalid and should therefore be
4867 treated as the start of the message body. The misbehaviour could have
4868 caused a number of strange effects, including loss of data in subsequent
4869 header lines, and spool format errors.
4871 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4872 client host had authenticated. This control can now be exercised by an ACL
4873 for more flexibility.
4875 69. By default, callouts do not happen when testing with -bh. There is now a
4876 variant, -bhc, which does actually run the callout code, including
4877 consulting and updating the callout cache.
4879 70. Added support for saslauthd authentication, courtesy of Alexander
4882 71. If statvfs() failed on the spool or log directories while checking their
4883 size for availability, Exim confusingly gave the error "space shortage".
4884 Furthermore, in debugging mode it crashed with a floating point exception.
4885 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4886 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4887 serious problem, Exim now writes to the main and panic logs when this
4888 happens, with details of the failure. It then refuses to accept the
4889 incoming message, giving the message "spool directory problem" or "log
4890 directory problem" with a 421 code for SMTP messages.
4892 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4893 0, 1, and 2 exist, because some OS complain for execs without them (see
4894 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4895 descriptors. However, the code omitted to check that the open succeeded,
4896 causing mysterious errors if for some reason the permissions on /dev/null
4897 got screwed. Now Exim writes a message to the main and panic logs, and
4898 bombs out if it can't open /dev/null.
4900 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4901 interact so that it is all more flexible. It is supposed to remain
4902 backwards compatible. Also added extra_local_interfaces.
4904 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4905 to bomb out with an assertion failure - to the client this appears as a
4906 connection drop. This problem occurs in the part of the code that was taken
4907 from the Samba project. Fortunately, the assertion is in a very simple
4908 function, so I have fixed this by reproducing the function inline in the
4909 one place where it is called, and arranging for authentication to fail
4910 instead of killing the process with assert().
4912 75. The SPA client code was not working when the server requested OEM rather
4913 than Unicode encoding.
4915 76. Added code to make require_files with a specific uid setting more usable in
4916 the case where statting the file as root fails - usually a non-root-mounted
4917 NFS file system. When this happens and the failure is EACCES, Exim now
4918 forks a subprocess and does the per-uid checking as the relevant uid.
4920 77. Added process_log_path.
4922 78. If log_file_path was not explicitly set, a setting of check_log_space or
4923 check_log_inodes was ignored.
4925 79. If a space check for the spool or log partitions fails, the incident is now
4926 logged. Of course, in the latter case the data may get lost...
4928 80. Added the %p formatting code to string_format() so that it can be used to
4929 print addresses in debug_print(). Adjusted all the address printing in the
4930 debugging in store.c to use %p rather than %d.
4932 81. There was a concern that a line of code in smtp_in.c could overflow a
4933 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4934 initially expressed, the concern was not well-founded, because trailing
4935 spaces are removed early. However, if the trailing spaces were followed by
4936 a NULL, they did not get removed, so the overflow was possible. Two fixes
4939 (a) I re-wrote the offending code in a cleaner fashion.
4940 (b) If an incoming SMTP command contains a NULL character, it is rejected
4943 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4944 runs initgroups(), so that if the Exim user is in any additional groups,
4945 they will be used during message reception.
4951 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob