1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.571 2009/10/16 12:33:09 nm4 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added patch by Johannes Berg that expands the main option
10 "spamd_servers" if it starts with a dollar sign.
12 TK/02 Write list of recipients to X-Envelope-Sender header when building
13 the mbox-format spool file for content scanning (suggested by Jakob
16 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
17 (http://www.dcc-servers.net/) support via dccifd. Activated by
18 setting EXPERIMENTAL_DCC=yes in Local/Makefile. Check out
19 experimental_spec.txt for more documentation.
21 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
22 by Mark Daniel Reidel <mr@df.eu>.
24 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
25 When building exim an external PCRE library is now needed -
26 PCRE is a system library on the majority of modern systems.
27 See entry on PCRE_LIBS in EDITME file.
29 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
30 conversation. Added nologin parameter to request.
31 Patch contributed by Kirill Miazine
33 TF/01 Do not log submission mode rewrites if they do not change the address.
35 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
37 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
38 log files in place. Contributed by Roberto Lima
40 NM/04 Bugzilla 667: close socket used by dovecot authenticator
42 TF/03 Bugzilla 615: When checking the local_parts router precondition
43 after a local_part_suffix or local_part_prefix option, Exim now
44 does not use the address's named list lookup cache, since this
45 contains cached lookups for the whole local part.
47 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
48 Robert Millan. Documentation is in experimental-spec.txt
50 TF/04 Bugzilla 668: Fix parallel build (make -j).
52 NM/05 Bugzilla 437: Prevent Maildix aux files being created with mode 000
54 NM/05 Bugzilla 598: Improvement to Dovecot authenticator handling.
55 Patch provided by Jan Srzednicki
57 TF/05 Leading white space used to be stripped from $spam_report which
58 wrecked the formatting. Now it is preserved.
60 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
61 that they are available at delivery time.
63 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
65 TF/08 TLS error reporting now respects the incoming_interface and
66 incoming_port log selectors.
68 TF/09 Produce a more useful error message if an SMTP transport's hosts
69 setting expands to an empty string.
71 NM/06 Bugzilla 744: EXPN did not work under TLS.
72 Patch provided by Phil Pennock
74 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
75 Patch provided by Richard Godbee
77 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
78 acl_smtp_notquit, added index entry.
80 NM/09 Bugzilla 787: Potential buffer overflow in string_format
81 Patch provided by Eugene Bujak
83 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to accept()
84 Patch provided by Maxim Dounin
86 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
87 Patch provided by Phil Pennock
89 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
91 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
92 Patch provided by Brad "anomie" Jorsch
94 NM/14 Bugzilla 614: Added timeout setting to transport filter.
95 Patch provided by Dean Brooks
97 TK/05 Add native DKIM support (does not depend on external libraries).
99 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
100 Patch provided by Graeme Fowler
102 NM/16 Bugzilla 851: Documentation example syntax fix.
104 NM/17 Changed NOTICE file to remove references to embedded PCRE.
106 NM/18 Bugzilla 894: Fix issue with very long lines including comments in lsearch
108 NM/19 Bugzilla 745: TLS version reporting
109 Patch provided by Phil Pennock
111 NM/20 Bugzilla 167: bool: condition support
112 Patch provided by Phil Pennock
114 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken clients
115 Patch provided by Phil Pennock
117 NM/22 Bugzilla 508: prepend (not append) Resent-Message-ID and Resent-Date
118 Patch provided by Brad "anomie" Jorsch
124 TK/01 Add preliminary DKIM support. Currently requires a forked version of
125 ALT-N's libdkim that I have put here:
126 http://duncanthrax.net/exim-experimental/
128 Note to Michael Haardt: I had to rename some vars in sieve.c. They
129 were called 'true' and it seems that C99 defines that as a reserved
130 keyword to be used with 'bool' variable types. That means you could
131 not include C99-style headers which use bools without triggering
132 build errors in sieve.c.
134 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
135 as mailq or other aliases. Changed the --help handling significantly
136 to do whats expected. exim_usage() emits usage/help information.
138 SC/01 Added the -bylocaldomain option to eximstats.
140 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
142 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
144 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
150 PH/01 Another patch from the Sieve maintainer.
152 PH/02 When an IPv6 address is converted to a string for single-key lookup
153 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
154 dots are used instead of colons so that keys in lsearch files need not
155 contain colons. This was done some time before quoting was made available
156 in lsearch files. However, iplsearch files do require colons in IPv6 keys
157 (notated using the quote facility) so as to distinguish them from IPv4
158 keys. This meant that lookups for IP addresses in host lists did not work
159 for iplsearch lookups.
161 This has been fixed by arranging for IPv6 addresses to be expressed with
162 colons if the lookup type is iplsearch. This is not incompatible, because
163 previously such lookups could never work.
165 The situation is now rather anomolous, since one *can* have colons in
166 ordinary lsearch keys. However, making the change in all cases is
167 incompatible and would probably break a number of configurations.
169 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
172 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
173 conversion specification without a maximum field width, thereby enabling
174 a rogue spamd server to cause a buffer overflow. While nobody in their
175 right mind would setup Exim to query an untrusted spamd server, an
176 attacker that gains access to a server running spamd could potentially
177 exploit this vulnerability to run arbitrary code as the Exim user.
179 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
180 $primary_hostname instead of what libspf2 thinks the hosts name is.
182 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
183 a directory entry by the name of the lookup key. Previously, if a
184 symlink pointed to a non-existing file or a file in a directory that
185 Exim lacked permissions to read, a lookup for a key matching that
186 symlink would fail. Now it is enough that a matching directory entry
187 exists, symlink or not. (Bugzilla 503.)
189 PH/03 The body_linecount and body_zerocount variables are now exported in the
192 PH/04 Added the $dnslist_matched variable.
194 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
195 This means they are set thereafter only if the connection becomes
198 PH/06 Added the client_condition to authenticators so that some can be skipped
199 by clients under certain conditions.
201 PH/07 The error message for a badly-placed control=no_multiline_responses left
202 "_responses" off the end of the name.
204 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
206 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
207 (without spaces) instead of just copying the configuration text.
209 PH/10 Added the /noupdate option to the ratelimit ACL condition.
211 PH/11 Added $max_received_linelength.
213 PH/12 Added +ignore_defer and +include_defer to host lists.
215 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
216 way in which PCRE > 7.0 is built.
218 PH/14 Implemented queue_only_load_latch.
220 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
221 MAIL command. The effect was to mangle the value on 64-bit systems.
223 PH/16 Another patch from the Sieve maintainer.
225 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
227 PH/18 If a system quota error occurred while trying to create the file for
228 a maildir delivery, the message "Mailbox is full" was not appended to the
229 bounce if the delivery eventually timed out. Change 4.67/27 below applied
230 only to a quota excession during the actual writing of the file.
232 PH/19 It seems that peer DN values may contain newlines (and other non-printing
233 characters?) which causes problems in log lines. The DN values are now
234 passed through string_printing() before being added to log lines.
236 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
237 and InterBase are left for another time.)
239 PH/21 Added message_body_newlines option.
241 PH/22 Guard against possible overflow in moan_check_errorcopy().
243 PH/23 POSIX allows open() to be a macro; guard against that.
245 PH/24 If the recipient of an error message contained an @ in the local part
246 (suitably quoted, of course), incorrect values were put in $domain and
247 $local_part during the evaluation of errors_copy.
253 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
254 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
257 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
258 issue a MAIL command.
260 PH/02 In an ACL statement such as
262 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
264 if a client was not listed at all, or was listed with a value other than
265 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
266 the condition was not true (as it should be), so access was not denied.
267 The bug was that the ! inversion was incorrectly passed on to the second
268 item. This has been fixed.
270 PH/03 Added additional dnslists conditions == and =& which are different from
271 = and & when the dns lookup returns more than one IP address.
273 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
274 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
276 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
277 FSYNC, which compiles an option called disable_fsync that allows for
278 bypassing fsync(). The documentation is heavily laced with warnings.
280 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
282 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
283 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
284 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
285 including adding "make clean"; (3) Added -fPIC when compiling the test
286 dynamically loaded module, to get rid of a warning.
288 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
289 message fails, move_frozen_messages = true and ignore_bounce_errors_after
290 = 0s. The bug is otherwise harmless.
292 PH/07 There was a bug in the dovecot authenticator such that the value of
293 $auth1 could be overwritten, and so not correctly preserved, after a
294 successful authentication. This usually meant that the value preserved by
295 the server_setid option was incorrect.
297 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
299 PH/09 Installed PCRE release 7.0.
301 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
302 run for batched SMTP input. It is now run at the start of every message
303 in the batch. While fixing this I discovered that the process information
304 (output by running exiwhat) was not always getting set for -bs and -bS
305 input. This is fixed, and it now also says "batched" for BSMTP.
307 PH/11 Added control=no_pipelining.
309 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
310 patch, slightly modified), and move the expansion of helo_data till after
311 the connection is made in the smtp transport (so it can use these
314 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
316 PH/14 Added log_selector = +pid.
318 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
320 PH/16 Add ${if forany and ${if forall.
322 PH/17 Added dsn_from option to vary the From: line in DSNs.
324 PH/18 Flush SMTP output before performing a callout, unless control =
325 no_callout_flush is set.
327 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
328 was true (the default) a successful delivery failed to delete the retry
329 item, thus causing premature timeout of the address. The bug is now
332 PH/20 Added hosts_avoid_pipelining to the smtp transport.
334 PH/21 Long custom messages for fakedefer and fakereject are now split up
335 into multiline reponses in the same way that messages for "deny" and
336 other ACL rejections are.
338 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
339 with slight modification.
341 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
342 draft, changing the syntax and factoring some duplicate code".
344 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
345 for deliveries of the second and subsequent messages over the same SMTP
348 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
349 ${reduce, with only minor "tidies".
351 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
353 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
354 expansion side effects.
356 PH/27 When a message times out after an over-quota error from an Exim-imposed
357 quota, the bounce message says "mailbox is full". This message was not
358 being given when it was a system quota that was exceeded. It now should
361 MH/03 Made $recipients available in local_scan(). local_scan() already has
362 better access to the recipient list through recipients_list[], but
363 $recipients can be useful in postmaster-provided expansion strings.
365 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
366 in the case of a MAIL command with additional options following the
367 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
368 were accidentally chopped off.
370 PH/29 SMTP synchronization checks are implemented when a command is read -
371 there is a check that no more input is waiting when there shouldn't be
372 any. However, for some commands, a delay in an ACL can mean that it is
373 some time before the response is written. In this time, more input might
374 arrive, invalidly. So now there are extra checks after an ACL has run for
375 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
376 pipelining has not been advertised.
378 PH/30 MH's patch to allow iscntrl() characters to be list separators.
380 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
381 returned in the SMTP response when smtp_return_error_details was false.
384 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
385 instead of the C I/O that was originally supplied, because problems were
388 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
389 Exim which did not show up earlier: it was assuming that a call to
390 SSL_CTX_set_info_callback() might give an error value. In fact, there is
391 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
392 was a macro that became an assignment, so it seemed to work. This has
393 changed to a proper function call with a void return, hence the compile
394 error. Exim's code has been fixed.
396 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
399 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
401 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
403 PH/37 If a message is not accepted after it has had an id assigned (e.g.
404 because it turns out to be too big or there is a timeout) there is no
405 "Completed" line in the log. When some messages of this type were
406 selected by exigrep, they were listed as "not completed". Others were
407 picked up by some special patterns. I have improved the selection
408 criteria to be more general.
410 PH/38 The host_find_failed option in the manualroute router can now be set
411 to "ignore", to completely ignore a host whose IP address cannot be
412 found. If all hosts are ignored, the behaviour is controlled by the new
413 host_all_ignored option.
415 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
416 homing or because of multiple MX records with /mx) generated more than
417 one IP address, and the following item turned out to be the local host,
418 all the secondary addresses of the first item were incorrectly removed
419 from the list, along with the local host and any following hosts (which
420 is what is supposed to happen).
422 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
423 whoever called Exim into the -H file. In the case of the daemon it was
424 behaving confusingly. When first started, it used values for whoever
425 started the daemon, but after a SIGHUP it used the Exim user (because it
426 calls itself on a restart). I have changed the code so that it now always
429 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
430 message are rejected with the same error (e.g. no authentication or bad
431 sender address), and a DATA command is nevertheless sent (as can happen
432 with PIPELINING or a stupid MUA), the error message that was given to the
433 RCPT commands is included in the rejection of the DATA command. This is
434 intended to be helpful for MUAs that show only the final error to their
437 PH/42 Another patch from the Sieve maintainer.
439 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
440 Eximstats - Fixed some broken HTML links and added missing column headers
442 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
443 columns for Rejects, Temp Rejects, Ham, and Spam rows.
445 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
447 PH/43 Yet another patch from the Sieve maintainer.
449 PH/44 I found a way to check for a TCP/IP connection going away before sending
450 the response to the final '.' that terminates a message, but only in the
451 case where the client has not sent further data following the '.'
452 (unfortunately, this is allowed). However, in many cases there won't be
453 any further data because there won't be any more messages to send. A call
454 to select() can be used: if it shows that the input is "ready", there is
455 either input waiting, or the socket has been closed. An attempt to read
456 the next input character can distinguish the two cases. Previously, Exim
457 would have sent an OK response which the client would never have see.
458 This could lead to message repetition. This fix should cure that, at
459 least in a lot of common cases.
461 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
462 advertised in response to EHLO.
468 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
469 fixed by 4.65/MH/01 (is this a record?) are fixed:
471 (i) An empty string was always treated as zero by the numeric comparison
472 operators. This behaviour has been restored.
474 (ii) It is documented that the numeric comparison operators always treat
475 their arguments as decimal numbers. This was broken in that numbers
476 starting with 0 were being interpreted as octal.
478 While fixing these problems I realized that there was another issue that
479 hadn't been noticed. Values of message_size_limit (both the global option
480 and the transport option) were treated as octal if they started with 0.
481 The documentation was vague. These values are now always treated as
482 decimal, and I will make that clear in the documentation.
488 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
489 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
492 MH/01 Don't check that the operands of numeric comparison operators are
493 integers when their expansion is in "skipping" mode (fixes bug
494 introduced by 4.64-PH/07).
496 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
497 child addresses, Exim now panics and dies. Previously, because the count
498 is held in a short int, deliveries were likely to be lost. As such a
499 large number of recipients for a single message is ridiculous
500 (performance will be very, very poor), I have chosen to impose a limit
501 rather than extend the field.
507 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
508 leftover -K file (the existence of which was triggered by #402).
509 While we were at it, introduced process PID as part of the -K
510 filename. This should rule out race conditions when creating
513 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
514 processing considerably. Previous code took too long for large mails,
515 triggering a timeout which in turn triggers #401.
517 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
518 in the DK code in transports.c. sendfile() is not really portable,
519 hence the _LINUX specificness.
521 TF/01 In the add_headers option to the mail command in an Exim filter,
522 there was a bug that Exim would claim a syntax error in any
523 header after the first one which had an odd number of characters
526 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
527 callout verification, Exim cached a "reject" for the entire domain. This
528 is correct for most verifications, but it is not correct for a recipient
529 verification with use_sender or use_postmaster set, because in that case
530 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
531 case of MAIL FROM:<> rejection from other early rejections (e.g.
532 rejection of HELO). When verifying a recipient using a non-null MAIL
533 address, the cache is ignored if it shows MAIL FROM:<> rejection.
534 Whatever the result of the callout, the value of the domain cache is
535 left unchanged (for any other kind of callout, getting as far as trying
536 RCPT means that the domain itself is ok).
538 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
541 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
542 manifest itself as EPIPE rather than ECONNECT. When tidying away a
543 session, the daemon ignores ECONNECT errors and logs others; it now
544 ignores EPIPE as well.
546 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
547 (quoted-printable decoding).
549 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
550 later the small subsequent patch to fix an introduced bug.
552 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
554 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
556 PH/08 An error is now given if message_size_limit is specified negative.
558 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
559 to be given (somewhat) arbitrary names.
561 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
564 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
565 miscellaneous code fixes
567 PH/10 Added the log_reject_target ACL modifier to specify where to log
570 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
571 hostname. This is wrong, because it relates to the incoming message (and
572 probably the interface on which it is arriving) and not to the outgoing
573 callout (which could be using a different interface). This has been
574 changed to use the value of the helo_data option from the smtp transport
575 instead - this is what is used when a message is actually being sent. If
576 there is no remote transport (possible with a router that sets up host
577 addresses), $smtp_active_hostname is used.
579 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
580 tweaks were necessary in order to get it to work (see also 21 below):
581 (a) The code assumed that strncpy() returns a negative number on buffer
582 overflow, which isn't the case. Replaced with Exim's string_format()
584 (b) There were several signed/unsigned issues. I just did the minimum
585 hacking in of casts. There is scope for a larger refactoring.
586 (c) The code used strcasecmp() which is not a standard C function.
587 Replaced with Exim's strcmpic() function.
588 (d) The code set only $1; it now sets $auth1 as well.
589 (e) A simple test gave the error "authentication client didn't specify
590 service in request". It would seem that Dovecot has changed its
591 interface. Fortunately there's a specification; I followed it and
592 changed what the client sends and it appears to be working now.
594 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
597 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
598 address A is aliased to B and C, where B exists and C does not. Without
599 -v the output is "A verified" because verification stops after a
600 successful redirection if more than one address is generated. However,
601 with -v the child addresses are also verified. Exim was outputting "A
602 failed to verify" and then showing the successful verification for C,
603 with its parentage. It now outputs "B failed to verify", showing B's
604 parentage before showing the successful verification of C.
606 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
607 look up a TXT record in a specific list after matching in a combined
610 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
611 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
612 they consult the DNS. I had assumed they would set it the way they
613 wanted; and indeed my experiments on Linux seem to show that in some
614 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
615 To be on the safe side, however, I have now made the interface to
616 host_find_byname() similar to host_find_bydns(), with an argument
617 containing the DNS resolver options. The host_find_byname() function now
618 sets these options at its start, just as host_find_bydns() does. The smtp
619 transport options dns_qualify_single and dns_search_parents are passed to
620 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
621 of host_find_byname() use the default settings of RES_DEFNAMES
622 (qualify_single) but not RES_DNSRCH (search_parents).
624 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
625 spool_read_header() do less string testing, by means of a preliminary
626 switch on the second character of optional "-foo" lines. (This is
627 overdue, caused by the large number of possibilities that now exist.
628 Originally there were few.) While I was there, I also converted the
629 str(n)cmp tests so they don't re-test the leading "-" and the first
630 character, in the hope this might squeeze out yet more improvement.
632 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
633 flag allowing group syntax was set by the header_syntax check but not
634 turned off, possible causing trouble later; (2) The flag was not being
635 set at all for the header_verify test, causing "group"-style headers to
636 be rejected. I have now set it in this case, and also caused header_
637 verify to ignore an empty address taken from a group. While doing this, I
638 came across some other cases where the code for allowing group syntax
639 while scanning a header line wasn't quite right (mostly, not resetting
640 the flag correctly in the right place). These bugs could have caused
641 trouble for malformed header lines. I hope it is now all correct.
643 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
644 with the "reply" argument non-NULL. The code, however (which originally
645 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
646 but it didn't always do it. This confused somebody who was copying the
647 code for some other use. I have removed all the tests.
649 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
650 feature that was used to support insecure browsers during the U.S. crypto
651 embargo. It requires special client support, and Exim is probably the
652 only MTA that supported it -- and would never use it because real RSA is
653 always available. This code has been removed, because it had the bad
654 effect of slowing Exim down by computing (never used) parameters for the
655 RSA_EXPORT functionality.
657 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
658 authenticator to fail if there's a tab character in the incoming data
659 (there should never be unless someone is messing about, as it's supposed
660 to be base64-encoded). Also added, on Timo's advice, the "secured" option
661 if the connection is using TLS or if the remote IP is the same as the
662 local IP, and the "valid-client-cert option" if a client certificate has
665 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
666 authenticators. This can be used for authorization after authentication
667 succeeds. (In the case of plaintext, it servers for both authentication
670 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
671 if any retry times were supplied.
673 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
674 connection, where there is no client IP address to check. In this
675 situation, the verify now always succeeds.
677 PH/25 Applied John Jetmore's -Mset patch.
679 PH/26 Added -bem to be like -Mset, but loading a message from a file.
681 PH/27 In a string expansion for a processed (not raw) header when multiple
682 headers of the same name were present, leading whitespace was being
683 removed from all of them, but trailing whitespace was being removed only
684 from the last one. Now trailing whitespace is removed from each header
685 before concatenation. Completely empty headers in a concatenation (as
688 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
689 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
691 PH/29 [Removed. This was a change that I later backed out, and forgot to
692 correct the ChangeLog entry (that I had efficiently created) before
693 committing the later change.]
695 PH/30 Exim was sometimes attempting to deliver messages that had suffered
696 address errors (4xx response to RCPT) over the same connection as other
697 messages routed to the same hosts. Such deliveries are always "forced",
698 so retry times are not inspected. This resulted in far too many retries
699 for the affected addresses. The effect occurred only when there were more
700 hosts than the hosts_max_try setting in the smtp transport when it had
701 the 4xx errors. Those hosts that it had tried were not added to the list
702 of hosts for which the message was waiting, so if all were tried, there
703 was no problem. Two fixes have been applied:
705 (i) If there are any address or message errors in an SMTP delivery, none
706 of the hosts (tried or untried) are now added to the list of hosts
707 for which the message is waiting, so the message should not be a
708 candidate for sending over the same connection that was used for a
709 successful delivery of some other message. This seems entirely
710 reasonable: after all the message is NOT "waiting for some host".
711 This is so "obvious" that I'm not sure why it wasn't done
712 previously. Hope I haven't missed anything, but it can't do any
713 harm, as the worst effect is to miss an optimization.
715 (ii) If, despite (i), such a delivery is accidentally attempted, the
716 routing retry time is respected, so at least it doesn't keep
717 hammering the server.
719 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
720 in ${readsocket because some servers need this prod.
722 PH/32 Added some extra debug output when updating a wait-xxx database.
724 PH/33 The hint "could be header name not terminated by colon", which has been
725 given for certain expansion errors for a long time, was not being given
726 for the ${if def:h_colon_omitted{... case.
728 PH/34 The spec says: "With one important exception, whenever a domain list is
729 being scanned, $domain contains the subject domain." There was at least
730 one case where this was not true.
732 PH/35 The error "getsockname() failed: connection reset by peer" was being
733 written to the panic log as well as the main log, but it isn't really
734 panic-worthy as it just means the connection died rather early on. I have
735 removed the panic log writing for the ECONNRESET error when getsockname()
738 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
739 runs only) independently of the message's sender address. This meant
740 that, if the 4xx error was in fact related to the sender, a different
741 message to the same recipient with a different sender could confuse
742 things. In particualar, this can happen when sending to a greylisting
743 server, but other circumstances could also provoke similar problems.
744 I have changed the default so that the retry time for these errors is now
745 based a combination of the sender and recipient addresses. This change
746 can be overridden by setting address_retry_include_sender=false in the
749 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
750 remote server are returned as part of bounce messages. This was not
751 happening for LMTP over a pipe (the lmtp transport), but now it is the
752 same for both kinds of LMTP.
754 PH/38 Despite being documented as not happening, Exim was rewriting addresses
755 in header lines that were in fact CNAMEs. This is no longer the case.
757 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
758 and queue runs started by the daemon processed all messages. This has
759 been fixed so that -R and -S can now usefully be given with -q<time>.
761 PH/40 Import PCRE release 6.7 (fixes some bugs).
763 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
765 PH/42 Give an error if -q is specified more than once.
767 PH/43 Renamed the variables $interface_address and $interface_port as
768 $received_ip_address and $received_port, to make it clear that these
769 values apply to message reception, and not to the outgoing interface when
770 a message is delivered. (The old names remain recognized, of course.)
772 PH/44 There was no timeout on the connect() call when using a Unix domain
773 socket in the ${readsocket expansion. There now is.
775 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
776 be meaningful with "accept".
778 SC/01 Eximstats V1.43
779 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
781 SC/02 Eximstats V1.44
782 Use a glob alias rather than an array ref in the generated
783 parser. This improves both readability and performance.
785 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
786 Collect SpamAssassin and rejection statistics.
787 Don't display local sender or destination tables unless
788 there is data to show.
789 Added average volumes into the top table text output.
791 SC/04 Eximstats V1.46
792 Collect data on the number of addresses (recipients)
793 as well as the number of messages.
795 SC/05 Eximstats V1.47
796 Added 'Message too big' to the list of mail rejection
797 reasons (thanks to Marco Gaiarin).
799 SC/06 Eximstats V1.48
800 Mainlog lines which have GMT offsets and are too short to
801 have a flag are now skipped.
803 SC/07 Eximstats V1.49 (Alain Williams)
804 Added the -emptyok flag.
806 SC/08 Eximstats V1.50
807 Fixes for obtaining the IP address from reject messages.
809 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
810 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
811 whitesspace changes from 4.64-PH/27
813 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
816 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
817 are found, allow negative numbers in numeric criteria)
819 JJ/06 exipick.20061117.2, added new $message_body_missing variable
821 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
822 to match changes made in 4.64-PH/43
824 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
826 PH/47 Put in an explicit test for a DNS lookup of an address record where the
827 "domain" is actually an IP address, and force a failure. This locks out
828 those revolvers/nameservers that support "A-for-A" lookups, in
829 contravention of the specifications.
831 PH/48 When a host name was looked up from an IP address, and the subsequent
832 forward lookup of the name timed out, the host name was left in
833 $sender_host_name, contrary to the specification.
835 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
836 restricted to single-key lookups, Exim was not diagnosing an error if
837 * or *@ was used with a query-style lookup.
839 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
841 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
842 long ago, but noone interested enough thought of it. Let's just say that
843 the "1.1" means that there are some new functions that weren't there at
844 some point in the past.
846 PH/51 Error processing for expansion failure of helo_data from an smtp
847 transport during callout processing was broken.
849 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
850 tested/used via the -bh/-bhc/-bs options.
852 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
853 bug, fixed in subsequent PCRE releases).
855 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
856 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
858 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
864 SC/01 Use a glob alias rather than an array ref in eximstats generated
865 parser. This improves both readability and performance.
867 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
868 Don't display local sender or destination tables in eximstats unless
869 there is data to show.
870 Added average volumes into the eximstats top table text output.
872 SC/03 Collect data on the number of addresses (recipients) as well
873 as the number of messages in eximstats.
875 TF/01 Correct an error in the documentation for the redirect router. Exim
876 does (usually) call initgroups() when daemonizing.
878 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
879 with consistent privilege compared to when running as a daemon.
881 TF/03 Note in the spec that $authenticated_id is not set for local
882 submissions from trusted users.
884 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
885 Thanks to Dean Brooks <dean@iglou.com> for the patch.
887 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
888 by adding some example configuration directives to the default
889 configuration file. A little bit of work is required to uncomment the
890 directives and define how usernames and passwords are checked, but
891 there is now a framework to start from.
893 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
894 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
895 without this. I don't know how relevant this is to other LDAP libraries.
897 PH/02 Add the verb name to the "unknown ACL verb" error.
899 PH/03 Magnus Holmgren's patch for filter_prepend_home.
901 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
903 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
904 directory not expanded when it should be if an expanded home directory
905 was set for the address (which is overridden by the transport).
907 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
910 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
911 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
912 because it is too late at that time, and has no effect.
914 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
915 security issue with \' (bugzilla #107). I could not use the
916 PQescapeStringConn() function, because it needs a PGconn value as one of
919 PH/08 When testing addresses using -bt, indicate those final addresses that
920 are duplicates that would not cause an additional delivery. At least one
921 person was confused, thinking that -bt output corresponded to deliveries.
922 (Suppressing duplicates isn't a good idea as you lose the information
923 about possibly different redirections that led to the duplicates.)
925 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
926 systems where poll() doesn't work, in particular OS X.
928 PH/10 Added more information to debugging output for retry time not reached.
930 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
931 operations in malware.c.
933 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
936 PH/13 If write_rejectlog was set false when logging was sent to syslog with
937 syslog_duplication set false, log lines that would normally be written
938 both the the main log and to the reject log were not written to syslog at
941 PH/14 In the default configuration, change the use of "message" in ACL warn
942 statements to "add_header".
944 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
945 not followed by a command (e.g. "seen endif").
947 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
948 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
951 PH/17 Added extra conditions to the default value of delay_warning_condition
955 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
956 { match{$h_precedence:}{(?i)bulk|list|junk} } \
957 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
960 The Auto-Submitted: and various List- headers are standardised, whereas I
961 don't think Precedence: ever was.
963 PH/18 Refactored debugging code in route_finduser() to show more information,
964 in particular, the error code if getpwnam() issues one.
966 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
967 This is apparently needed in addition to the PH/07 change above to avoid
968 any possible encoding problems.
970 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
971 but not after initializing Perl.
973 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
974 output them only if debugging. By default they are written stderr,
975 apparently, which is not desirable.
977 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
980 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
983 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
985 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
986 authenticated or an ident call has been made. Suppress the default
987 values for $authenticated_id and $authenticated_sender (but permit -oMai
988 and -oMas) when testing with -bh.
990 PH/24 Re-jigged the order of the tests in the default configuration so that the
991 tests for valid domains and recipients precede the DNS black list and CSA
992 tests, on the grounds that those ones are more expensive.
994 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
995 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
996 This bug exists in every version of Exim that I still have, right back to
999 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1000 However, an attempt to turn on case-sensitivity in a regex key by
1001 including (?-i) didn't work because the subject string was already
1002 lowercased, and the effects were non-intuitive. It turns out that a
1003 one-line patch can be used to allow (?-i) to work as expected.
1009 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1010 other effects) broke the use of negated acl sub-conditions.
1012 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1015 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1016 "Deny" causes Exim to reject the incoming connection with a 554 error.
1017 Unfortunately, if there is a major crisis, such as a disk failure,
1018 tcp-wrappers gives "deny", whereas what one would like would be some
1019 kind of temporary error. A kludge has been added to help with this.
1020 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1021 554 error is used if errno is still zero or contains ENOENT (which occurs
1022 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1025 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1027 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1028 errors. Otherwise a message that provokes a temporary error (when other
1029 messages do not) can cause a whole host to time out.
1031 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1032 addresses were routed directly to files or pipes from a redirect router.
1033 File deliveries just didn't batch; pipe deliveries might have suffered
1036 PH/06 A failure to get a lock for a hints database would erroneously always say
1037 "Failed to get write lock", even when it was really a read lock.
1039 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1040 of 0600. This has been changed to use the value of the lockfile_mode
1041 option (which defaults to 0600).
1043 PH/08 Applied small patch from the Sieve maintainer.
1045 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1046 folder from quota calculations, a direct delivery into this folder messed
1047 up the contents of the maildirsize file. This was because the regex was
1048 used only to exclude .Trash (or whatever) when the size of the mailbox
1049 was calculated. There was no check that a delivery was happening into an
1050 excluded directory. This bug has been fixed by ignoring all quota
1051 processing for deliveries into excluded directories.
1053 PH/10 Added the maildirfolder_create_regex option to appendfile.
1059 PH/01 The code for finding all the local interface addresses on a FreeBSD
1060 system running IPv6 was broken. This may well have applied to all BSD
1061 systems, as well as to others that have similar system calls. The broken
1062 code found IPv4 interfaces correctly, but gave incorrect values for the
1063 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1064 that it would not match correctly against @[] and not recognize the IPv6
1067 PH/02 The ipliteral router was not recognizing addresses of the form user@
1068 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1070 PH/03 Added disable_ipv6.
1072 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1073 decoded form, because it is most often used to construct To: headers
1074 lines in autoreplies, and the decoded form may well be syntactically
1075 invalid. However, $reply_address has leading white space removed, and all
1076 newlines turned into spaces so that the autoreply transport does not
1079 PH/05 If group was specified without a user on a router, and no group or user
1080 was specified on a transport, the group from the router was ignored.
1082 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1083 for visible compile-time settings that can be used to change these
1084 numbers, for those that want even more. Backwards compatibility with old
1085 spool files has been maintained. However, going back to a previous Exim
1086 release will lost any variables that are in spool files.
1088 PH/07 Two small changes when running in the test harness: increase delay when
1089 passing a TCP/IP connection to a new process, in case the original
1090 process has to generate a bounce, and remove special handling of
1091 127.0.0.2 (sic), which is no longer necessary.
1093 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1094 be the same on different OS.
1096 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1099 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1100 whether --show-vars was specified or not
1102 JJ/02 exipick: Added support for new ACL variable spool format introduced
1105 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1106 syntactically invalid From: or Reply-to: line, and a filter used this to
1107 generate an autoreply, and therefore failed to obtain an address for the
1108 autoreply, Exim could try to deliver to a non-existent relative file
1109 name, causing unrelated and misleading errors. What now happens is that
1110 it logs this as a hard delivery error, but does not attempt to create a
1113 PH/11 The exinext utility has a -C option for testing purposes, but although
1114 the given file was scanned by exinext itself; it wasn't being passed on
1115 when Exim was called.
1117 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1118 an end-of-file indication when reading a command response.
1120 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1121 compiled. In many other places in Exim, IPv6 addresses are always
1122 recognized, so I have changed this. It also means that IPv4 domain
1123 literals of the form [IPV4:n.n.n.n] are now always recognized.
1125 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1126 used if the router is not running as root, for example, when verifying at
1127 ACL time, or when using -bh. The debugging output from this situation was
1128 non-existent - all you got was a failure to exec. I have made two
1131 (a) Failures to set uid/gid, the current directory, or a process leader
1132 in a subprocess such as that created by queryprogram now generate
1133 suitable debugging ouput when -d is set.
1135 (b) The queryprogram router detects when it is not running as root,
1136 outputs suitable debugging information if -d is set, and then runs
1137 the subprocess without attempting to change uid/gid.
1139 PH/15 Minor change to Makefile for building test_host (undocumented testing
1142 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1143 additional section of a DNS packet that returns MX or SRV records.
1144 Instead, it always explicitly searches for A/AAAA records. This avoids
1145 major problems that occur when a DNS server includes only records of one
1146 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1147 fixed another bug: if SRV records were looked up and the corresponding
1148 address records were *not* found in the additional section, the port
1149 values from the SRV records were lost.
1151 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1152 using the correct key (the original address) when searching the retry
1153 rules in order to find which one to use for generating the retry hint.
1155 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1156 adding the default one. Similarly, if it contains a Reply-To: header, the
1157 errors_reply_to option, if set, is not used.
1159 PH/19 When calculating a retry time, Exim used to measure the "time since
1160 failure" by looking at the "first failed" field in the retry record. Now
1161 it does not use this if it is later than than the arrival time of the
1162 message. Instead it uses the arrival time. This makes for better
1163 behaviour in cases where some deliveries succeed, thus re-setting the
1164 "first failed" field. An example is a quota failure for a huge message
1165 when small messages continue to be delivered. Without this change, the
1166 "time since failure" will always be short, possible causing more frequent
1167 delivery attempts for the huge message than are intended.
1168 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1170 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1171 $1, $2, $3) because the numerical variables can be reset during some
1172 expansion items (e.g. "match"), thereby losing the authentication data.
1174 PH/21 Make -bV show the size of off_t variables so that the test suite can
1175 decide whether to run tests for quotas > 2G.
1177 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1178 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1179 filecount value is greater than 2G or if a quota value is greater than 2G
1180 on a system where the size of off_t is not greater than 4, a panic error
1183 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1184 never match. The debug and -bh output now contains an explicit error
1185 message indicating a malformed IPv4 address or mask.
1187 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1188 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1189 PH/23 above applies.
1191 PH/25 Do not write to syslog when running in the test harness. The only
1192 occasion when this arises is a failure to open the main or panic logs
1193 (for which there is an explicit test).
1195 PH/26 Added the /no_tell option to "control=freeze".
1197 PH/27 If a host name lookup failed very early in a connection, for example, if
1198 the IP address matched host_lookup and the reverse lookup yielded a name
1199 that did not have a forward lookup, an error message of the form "no IP
1200 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1201 could be logged. Now it outputs the IP address instead of "NULL".
1203 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1204 allows the sender and the authenticated sender to be set when
1205 submitting a message from within Exim. Since child_open_exim() is
1206 documented for local_scan(), the new function should be too.
1208 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1209 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1210 results in an empty string is now treated as unset.
1212 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1214 PH/31 Added sender_verify_fail logging option.
1216 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1217 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1218 tidied the source and removed it altogether.
1220 PH/33 When a queue run was abandoned because the load average was too high, a
1221 log line was always written; now it is written only if the queue_run log
1222 selector is set. In addition, the log line for abandonment now contains
1223 information about the queue run such as the pid. This is always present
1224 in "start" and "stop" lines but was omitted from the "abandon" line.
1226 PH/34 Omit spaces between a header name and the colon in the error message that
1227 is given when verify = headers_syntax fails (if there are lots of them,
1228 the message gets confusing).
1230 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1231 names, as there are now some PTR records that contain slashes. This check
1232 is only to protect against broken name servers that fall over on strange
1233 characters, so the fact that it applies to all lookups doesn't matter.
1235 PH/36 Now that the new test suite is complete, we can remove some of the
1236 special code in Exim that was needed for the old test suite. For example,
1237 sorting DNS records because real resolvers return them in an arbitrary
1238 order. The new test suite's fake resolver always returns records in the
1241 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1242 bounces) except when queue_only is set, to avoid logging races between
1243 the different processes.
1245 PH/38 Panic-die if .include specifies a non-absolute path.
1247 PH/39 A tweak to the "H" retry rule from its user.
1249 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1250 a label. They prevented compilation on older perls.
1252 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1253 a warning to be raised on newish perls.
1255 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1256 on queue. Changes to match documented behaviour of showing count of
1257 messages matching specified criteria.
1259 PH/40 Changed the default ident timeout from 30s to 5s.
1261 PH/41 Added support for the use of login_cap features, on those BSD systems
1262 that have them, for controlling the resources used by pipe deliveries.
1264 PH/42 The content-scanning code uses fopen() to create files in which to put
1265 message data. Previously it was not paying any attention to the mode of
1266 the files. Exim runs with umask(0) because the rest of the code creates
1267 files with open(), and sets the required mode explicitly. Thus, these
1268 files were ending up world-writeable. This was not a big issue, because,
1269 being within the spool directory, they were not world-accessible. I have
1270 created a function called modefopen, which takes an additional mode
1271 argument. It sets umask(777), creates the file, chmods it to the required
1272 mode, then resets the umask. All the relevant calls to fopen() in the
1273 content scanning code have been changed to use this function.
1275 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1276 to 24 hours. This avoids potential overflow problems when processing G
1277 and H retry rules. I suspect nobody ever tinkers with this value.
1279 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1281 PH/45 When the plaintext authenticator is running as a client, the server's
1282 challenges are checked to ensure they are valid base64 strings. By
1283 default, the authentication attempt is cancelled if an invalid string is
1284 received. Setting client_ignore_invalid_base64 true ignores these errors.
1285 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1286 they are received. Thus, the responses can be made to depend on the
1287 challenges. If an invalid string is ignored, an empty string is placed in
1290 PH/46 Messages that are created by the autoreply transport now contains a
1291 References: header, in accordance with RFCs 2822 and 3834.
1293 PH/47 Added authenticated_sender_force to the smtp transport.
1295 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1297 PH/49 Installed latest patch from the Sieve maintainer.
1299 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1300 was also set, the appendfile transport was unnecessarily scanning a
1301 directory of message files (e.g. for maildir delivery) to find the count
1302 of files (along with the size), even though it did not need this
1303 information. It now does the scan only if it needs to find either the
1304 size of the count of files.
1306 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1308 PH/52 Two bugs concerned with error handling when the smtp transport is
1311 (i) Exim was not creating retry information for temporary errors given
1312 for individual recipients after the DATA command when the smtp transport
1313 was used in LMTP mode. This meant that they could be retried too
1314 frequently, and not timed out correctly.
1316 (ii) Exim was setting the flag that allows error details to be returned
1317 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1318 recipients that were returned after the DATA command.
1320 PH/53 This is related to PH/52, but is more general: for any failing address,
1321 when detailed error information was permitted to be returned to the
1322 sender, but the error was temporary, then after the final timeout, only
1323 "retry timeout exceeded" was returned. Now it returns the full error as
1324 well as "retry timeout exceeded".
1326 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1327 do this, and (what is worse) MTAs that accept it.
1329 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1330 will now be deprecated.
1332 PH/56 New os.c-cygwin from the Cygwin maintainer.
1334 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1335 formats (previously only available in exim formats via -bpr, -bpru,
1336 and -bpra. Now also available in native and exiqgrep formats)
1338 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1339 with very large, slow to parse queues
1341 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1343 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1345 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1346 responses to authentication challenges, though it was showing the
1347 challenges; (ii) I've removed the CR characters from the debug output for
1350 PH/58 Allow for the insertion of a newline as well as a space when a string
1351 is turned into more than one encoded-word during RFC 2047 encoding. The
1352 Sieve code now uses this.
1354 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1355 data_4xx, lost_connection, tls_required.
1357 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1358 message was being sent as an SMTP response.
1360 PH/61 Add -l and -k options to exicyclog.
1362 PH/62 When verifying, if an address was redirected to one new address, so that
1363 verification continued, and the new address failed or deferred after
1364 having set something in $address_data, the value of $address_data was not
1365 passed back to the ACL. This was different to the case when no
1366 redirection occurred. The value is now passed back in both cases.
1368 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1369 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1370 use login_cap.h, so on its own it isn't the distinguishing feature. The
1371 new name refers directly to the setclassresources() function.
1373 PH/65 Added configuration files for NetBSD3.
1375 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1377 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1378 is preferred over IPv4.
1380 PH/68 The bounce_return_message and bounce_return_body options were not being
1381 honoured for bounces generated during the reception of non-SMTP messages.
1382 In particular, this applied to messages rejected by the ACL. This bug has
1383 been fixed. However, if bounce_return_message is true and bounce_return_
1384 body is false, the headers that are returned for a non-SMTP message
1385 include only those that have been read before the error was detected.
1386 (In the case of an ACL rejection, they have all been read.)
1388 PH/69 The HTML version of the specification is now built in a directory called
1389 spec_html instead of spec.html, because the latter looks like a path with
1390 a MIME-type, and this confuses some software.
1392 PH/70 Catch two compiler warnings in sieve.c.
1394 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1395 function verify_get_ident() calls ip_connect() to connect a socket, but
1396 if the "connect()" function timed out, ip_connect() used to close the
1397 socket. However, verify_get_ident() also closes the socket later, and in
1398 between Exim writes to the log, which may get opened at this point. When
1399 the socket was closed in ip_connect(), the log could get the same file
1400 descriptor number as the socket. This naturally causes chaos. The fix is
1401 not to close the socket in ip_connect(); the socket should be closed by
1402 the function that creates it. There was only one place in the code where
1403 this was missing, in the iplookup router, which I don't think anybody now
1404 uses, but I've fixed it anyway.
1406 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1407 well as to direct DNS lookups. Otherwise the handling of names in host
1408 lists is inconsistent and therefore confusing.
1414 PH/01 Two changes to the default runtime configuration:
1416 (1) Move the checks for relay_from_hosts and authenticated clients from
1417 after to before the (commented out) DNS black list checks.
1419 (2) Add control=submission to the relay_from_hosts and authenticated
1420 clients checks, on the grounds that messages accepted by these
1421 statements are most likely to be submissions.
1423 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1425 (1) Generate an error if the third argument for the ${prvs expansion is
1428 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1431 (3) Reset the variables that are obtained from the first argument of
1432 ${prvscheck and used in the second argument before leaving the code,
1433 because their memory is reclaimed, so using them afterwards may do
1436 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1437 one (it's much easier than Tom thought :-).
1439 (5) Because of (4), we can now allow for the use of $prvscheck_result
1440 inside the third argument.
1442 PH/03 For some reason, the default setting of PATH when running a command from
1443 a pipe transport was just "/usr/bin". I have changed it to
1446 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1447 anything to be listed in the output from -bV.
1449 PH/05 When a filter generated an autoreply, the entire To: header line was
1450 quoted in the delivery log line, like this:
1452 => >A.N.Other <ano@some.domain> <original@ddress> ...
1454 This has been changed so that it extracts the operative address. There
1455 may be more than one such address. If so, they are comma-separated, like
1458 => >ano@some.domain,ona@other.domain <original@ddress> ...
1460 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1461 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1462 not being looked up in the rDNS to get a host name, Exim was showing the
1463 IP address twice in Received: lines, even though the IP addresses were
1464 identical. For example:
1466 Received: from [1.2.3.4] (helo=[1.2.3.4])
1468 However, if the real host name was known, it was omitting the HELO data
1469 if it matched the actual IP address. This has been tidied up so that it
1470 doesn't show the same IP address twice.
1472 PH/07 When both +timestamp and +memory debugging was on, the value given by
1473 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1474 called by the debug printing, thereby overwriting the timestamp buffer.
1475 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1477 PH/08 When the original message was included in an autoreply transport, it
1478 always said "this is a copy of the message, including all the headers",
1479 even if body_only or headers_only was set. It now gives an appropriate
1482 PH/09 Applied a patch from the Sieve maintainer which:
1484 o fixes some comments
1485 o adds the (disabled) notify extension core
1486 o adds some debug output for the result of if/elsif tests
1487 o points to the current vacation draft in the documentation
1488 and documents the missing references header update
1492 o fixes a bug in processing the envelope test (when testing
1493 multiple envelope elements, the last element determinted the
1496 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1497 Electronic Mail") by including:
1499 Auto-submitted: auto-generated
1501 in the messages that it generates (bounce messages and others, such as
1502 warnings). In the case of bounce messages for non-SMTP mesages, there was
1503 also a typo: it was using "Auto_submitted" (underscore instead of
1504 hyphen). Since every message generated by Exim is necessarily in response
1505 to another message, thes have all been changed to:
1507 Auto-Submitted: auto-replied
1509 in accordance with these statements in the RFC:
1511 The auto-replied keyword:
1513 - SHOULD be used on messages sent in direct response to another
1514 message by an automatic process,
1516 - MUST NOT be used on manually-generated messages,
1518 - MAY be used on Delivery Status Notifications (DSNs) and Message
1519 Disposition Notifications (MDNs),
1521 - MUST NOT be used on messages generated by automatic or periodic
1522 processes, except for messages which are automatic responses to
1525 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1526 to the default Received: header definition.
1528 PH/12 Added log selector acl_warn_skipped (default on).
1530 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1531 variables because (a) they are in the wrong storage pool and (b) even if
1532 they were copied, it wouldn't work properly because of the caching.
1534 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1535 checking when decoding. Apparently there are clients that generate
1536 overlong encoded strings. Why am I not surprised?
1538 PH/15 If the first argument of "${if match_address" was not empty, but did not
1539 contain an "@" character, Exim crashed. Now it writes a panic log message
1540 and treats the condition as false.
1542 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1544 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1545 extension "envelope-auth". The code is finished and in agreement with
1546 other implementations, but there is no documentation so far and in fact,
1547 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1548 not changing the active code.
1550 Print executed "if" and "elsif" statements when debugging is used. This
1551 helps a great deal to understand what a filter does.
1553 Document more things not specified clearly in RFC3028. I had all this
1554 sorted out, when out of a sudden new issues came to my mind. Oops."
1556 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1559 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1560 canonical form (as documented). However, after a host name lookup from
1561 the IP address, check_host() was doing a simple string comparison with
1562 addresses acquired from the DNS when checking that the found name did
1563 have the original IP as one of its addresses. Since any found IPv6
1564 addresses are likely to be in abbreviated form, the comparison could
1565 fail. Luckily, there already exists a function for doing the comparison
1566 by converting both addresses to binary, so now that is used instead of
1567 the text comparison.
1569 PH/20 There was another similar case to PH/19, when a complete host name was
1570 given in a host list; looking up its IP address could give an abbreviated
1571 form, whereas the current host's name might or might not be abbreviated.
1572 The same fix has been applied.
1578 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1579 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1582 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1583 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1585 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1587 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1588 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1589 Also, for "Auto-Submitted", treat anything other than "no" as
1590 non-personal, in accordance with RFC 3834. (Previously it treated
1591 anything starting "auto-" as non-personal.)
1593 TF/01 The control=submission/name=... option had a problem with syntax
1594 errors if the name included a slash character. The /name= option
1595 now slurps the rest of the string, so it can include any characters
1596 but it must come last in the list of options (after /sender_retain
1599 PH/05 Some modifications to the interface to the fake nameserver for the new
1607 TK/01 Added the "success_on_redirect" address verification option. See
1608 NewStuff for rationale and an example.
1610 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1612 PH/02 Patch to exigrep to allow it to work on syslog lines.
1614 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1615 fread() to skip over the body file's header line, because in Cygwin the
1616 header line is locked and is inaccessible.
1618 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1619 co-exist for some time) to make it clear that it is the Exim ID that is
1620 referenced, not the Message-ID: header line.
1622 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1623 string_format() function, because snprintf() does not exist on all
1626 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1627 ${stat: expansion item.
1629 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1630 protocol synchronization error", to keep the pedants happy.
1632 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1633 well as for IRIX systems, when gcc is being used. See the host.c source
1636 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1638 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1641 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1642 transport and to the smtp transport in LMTP mode.
1644 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1646 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1647 run a filter in a subprocess. This could lead to confusion in subsequent
1648 lookups in the parent process. There should also be a search_tidyup() at
1649 the end of the subprocess.
1651 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1652 only if the host matched helo_try_verify_hosts, which caused the
1653 verification to occur when the EHLO/HELO command was issued. The ACL just
1654 tested the remembered result. Now, if a previous verification attempt has
1655 not happened, "verify = helo" does it there and then.
1657 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1659 TK/03 Fix log output including CR from clamd.
1661 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1662 exist provoked a memory error which could cause a segfault.
1664 PH/15 Installed PCRE 6.2
1666 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1668 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1669 of the problem. Specifically, suggested +O2 rather than +O1 for the
1672 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1674 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1675 record, the port was not correctly logged when the outgoing_port log
1676 selector was set (it logged the transort's default port).
1678 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1679 fallback_hosts, and "hosts" in the smtp transport.
1681 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1682 host errors such as "Connection refused".
1684 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1685 authentication with radiusclient 0.4.9:
1687 - Error returned from rc_read_config was caught wrongly
1688 - Username/password not passed on to radius server due to wrong length.
1690 The presumption is that some radiusclient API changes for 4.51/PH/17
1691 were not taken care of correctly. The code is still untested by me (my
1692 Linux distribution still has 0.3.2 of radiusclient), but it was
1693 contributed by a Radius user.
1695 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1696 expanding the "port" option of the smtp transport.
1698 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1699 while reading a MIME header. Thanks to Tom Hughes for a patch.
1701 PH/24 Include config.h inside local_scan.h so that configuration settings are
1704 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1705 that in an ACL for RCPT (for example), you can examine exactly what was
1708 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1709 commands, but it was not correctly comparing the address with the actual
1710 client host address. Thus, it would show the EHLO address in Received:
1711 header lines when this was not necessary.
1713 PH/27 Added the % operator to ${eval:}.
1715 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1716 it should be ignoring failures (because with -C, for example, it has lost
1717 privilege). It wasn't ignoring creation failures other than "already
1720 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1723 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1724 because an input line was too long, either on its own, or by virtue of
1725 too many continuations, the temporary file was not being removed, and the
1726 return code was incorrect.
1728 PH/31 Missing "BOOL" in function definition in filtertest.c.
1730 PH/32 Applied Sieve patches from the maintainer.
1732 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1734 PH/33 Added "verify = not_blind".
1736 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1737 Local/Makefile (with some defaults set). These are used in built scripts
1738 such as exicyclog, but they have never been used in the exim_install
1739 script (though there are many overriding facilities there). I have
1740 arranged that the exim_install script now takes note of these two
1743 PH/35 Installed configuration files for Dragonfly.
1745 PH/36 When a locally submitted message by a trusted user did not contain a
1746 From: header, and the sender address was obtained from -f or from an SMTP
1747 MAIL command, and the trusted user did not use -F to supply a sender
1748 name, $originator_name was incorrectly used when constructing a From:
1749 header. Furthermore, $originator_name was used for submission mode
1750 messages from external hosts without From: headers in a similar way,
1751 which is clearly wrong.
1753 PH/37 Added control=suppress_local_fixups.
1755 PH/38 When log_selector = +received_sender was set, and the addition of the
1756 sender made the log line's construction buffer exactly full, or one byte
1757 less than full, an overflow happened when the terminating "\n" was
1760 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1761 when the result of a list match is failure because a DNS lookup failed.
1763 PH/40 RM_COMMAND is now used in the building process.
1765 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1766 the "build-* directories that it finds.
1768 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1769 address in a domain literal was a prefix of an interface address.
1771 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1772 when verifying a sender address, unless rewrite_headers is false.
1774 PH/44 Wrote a long comment about why errors_to addresses are verified as
1775 recipients, not senders.
1777 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1778 the ratelimit ACL was added.
1780 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1782 PH/46 Added extra information about PostgreSQL errors to the error string.
1784 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1785 avoiding the need to install special zones in a real server. This is
1786 backwards compatible; if it can't find the fake resolver, it drops back.
1787 Thus, both old and new test suites can be run.
1789 TF/02 Added util/ratelimit.pl
1791 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1792 clock is set back in time.
1794 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1795 Candler <B.Candler@pobox.com>.
1797 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1798 OK for addresses that are the result of redirections.
1800 PH/48 A number of further additions for the benefit of the new test suite,
1801 including a fake gethostbyname() that interfaces to the fake DNS resolver
1804 TF/06 The fix for widen_domains has also been applied to qualify_single and
1805 search_parents which are the other dnslookup options that can cause
1808 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1811 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1813 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1814 service" if it encountered an MX record with an empty target hostname.
1815 The message is now "an MX or SRV record indicated no SMTP service".
1817 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1818 if the DNS of the sending site is misconfigured. This is quite a
1819 common situation. This change restores the behaviour of treating a
1820 helo verification defer as a failure.
1822 PH/51 If self=fail was set on a router, the bounce message did not include the
1823 actual error message.
1829 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1831 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1832 command itself ended in error, the underlying message about the transport
1833 filter timeout was being overwritten with the pipe command error. Now the
1834 underlying error message should be appended to the second error message.
1836 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1838 PH/02 Reduce the amount of output that "make" produces by default. Full output
1839 can still be requested.
1841 PH/03 The warning log line about a condition test deferring for a "warn" verb
1842 was being output only once per connection, rather than after each
1843 occurrence (because it was using the same function as for successful
1844 "warn" verbs). This seems wrong, so I have changed it.
1846 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1847 it should not have, which might have caused a crash in the right
1848 circumstances, but probably never did.
1850 PH/04 Installed a modified version of Tony Finch's patch to make submission
1851 mode fix the return path as well as the Sender: header line, and to
1852 add a /name= option so that you can make the user's friendly name appear
1855 TF/03 Added the control = fakedefer ACL modifier.
1857 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1858 Mark Lowes for thorough testing.
1860 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1862 TK/03 Merged latest SRS patch from Miles Wilton.
1864 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1865 with the definition in sysexits.h (which is #included earlier).
1866 Fortunately, Exim does not actually use EX_OK. The code used to try to
1867 preserve the sysexits.h value, by assumimg that macro definitions were
1868 scanned for macro replacements. I have been disabused of this notion,
1869 so now the code just undefines EX_OK before #including unistd.h.
1871 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1872 in the smtp transport. When a block could not be written in a single
1873 write() function, the timeout was being re-applied to each part-write.
1874 This seems wrong - if the receiver was accepting one byte at a time it
1875 would take for ever. The timeout is now adjusted when this happens. It
1876 doesn't have to be particularly precise.
1878 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1879 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1881 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1882 without a domain if the check to <postmaster@domain> fails.
1884 SC/01 Eximstats: added -xls and the ability to specify output files
1885 (patch written by Frank Heydlauf).
1887 SC/02 Eximstats: use FileHandles for outputing results.
1889 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1891 SC/04 Eximstats: fixed display of large numbers with -nvr option
1893 SC/05 Eximstats: fixed merging of reports with empty tables.
1895 SC/06 Eximstats: added the -include_original_destination flag
1897 SC/07 Eximstats: removed tabs and trailing whitespace.
1899 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1901 TK/06 MBOX spool code: Add real "From " MBOX separator line
1902 so the .eml file is really in mbox format (even though
1903 most programs do not really care). Patch from Alex Miller.
1905 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1906 The latter is generated from $received_to and is only set if the
1907 message has one envelope recipient. SA can use these headers,
1908 obviously out-of-the-box. Patch from Alex Miller.
1910 PH/08 The ${def test on a variable was returning false if the variable's
1911 value was "0", contrary to what the specification has always said!
1912 The result should be true unless the variable is empty.
1914 PH/09 The syntax error of a character other than { following "${if
1915 def:variable_name" (after optional whitespace) was not being diagnosed.
1916 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1917 accidental colon was present, for example, could give incorrect results.
1919 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1920 result is used (not including appendfile, where other changes are about
1923 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1924 This involved changing a lot of size variables from int to off_t. It
1925 should work with maildirs and everything.
1927 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1928 spamd dying while we are connected to it.
1930 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1933 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1934 function for BDB 4.3.
1936 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1938 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1939 the sources for PCRE have been re-arranged and more files are now
1942 PH/15 The code I had for printing potentially long long variables in PH/11
1943 above was not the best (it lost precision). The length of off_t variables
1944 is now inspected at build time, and an appropriate printing format (%ld
1945 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1946 to be "long long int" or "long int". This is needed for the internal
1947 formatting function string_vformat().
1949 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1950 the configuration file to be ":syslog", then the script "guesses" where
1951 the logs files are, rather than using the compiled in default. In our
1952 case the guess is not the same as the compiled default, so the script
1953 suddenly stopped working when I started to use syslog. The patch checks
1954 to see if log_file_path is "". If so, it attempts to read it from exim
1955 with no configuration file to get the compiled in version, before it
1956 falls back to the previous guessing code."
1958 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1959 implementing BATV in an Exim configuration. See NewStuff for the gory
1962 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1963 Makefile that are specific to HP-UX.
1965 PH/18 If the "use_postmaster" option was set for a recipient callout together
1966 with the "random" option, the postmaster address was used as the MAIL
1967 FROM address for the random test, but not for the subsequent recipient
1968 test. It is now used for both.
1970 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1971 patch removes a few documentation additions to RFC 3028, because the
1972 latest draft now contains them. It adds the new en;ascii-case comparator
1973 and a new error check for 8bit text in MIME parts. Comparator and
1974 require names are now matched exactly. I enabled the subaddress
1975 extension, but it is not well tested yet (read: it works for me)."
1977 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1978 rework some of the code of TK/09 above to avoid the hardwired use of
1979 "%lld" and "long long". Replaced the call to snprintf() with a call to
1982 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1983 records point to non-existent hosts", "retry timeout exceeded", and
1984 "retry time not reached for any host after a long failure period".
1986 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1987 experimental DomainKeys support:
1989 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1990 (2) On an error such as an illegally used "control", the wrong name for
1991 the control was given.
1993 These problems did NOT occur unless DomainKeys support was compiled.
1995 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1997 PH/24 Added ${if match_ip condition.
1999 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2000 databases so that it will be absolutely obvious if a crash occurs in the
2001 DB library. This is a regular occurrence (often caused by mis-matched
2004 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2005 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2006 picked up on a user's system that detects such things. There doesn't seem
2007 to be a gcc warning option for this - only an attribute that has to be
2008 put on the function's prototype. It seems that in Fedora Core 4 they have
2009 set this on a number of new functions. No doubt there will be more in due
2012 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2013 specify a transport. However, if an address that was verified by such a
2014 router was the subject of a callout, Exim crashed because it tried to
2015 read the rcpt_include_affixes from the non-existent transport. Now it
2016 just assumes that the setting of that option is false. This bug was
2017 introduced by 4.51/PH/31.
2019 PH/28 Changed -d+all to exclude +memory, because that information is very
2020 rarely of interest, but it makes the output a lot bigger. People tend to
2021 do -d+all out of habit.
2023 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2024 code in os-type was giving problems when libc.so lives in lib64, like on
2027 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2028 aren't the modern standard, and it seems that some systems' include files
2029 don't always have them. Exim was already checking for some of the newer
2030 ones like T_AAAA, and defining it itself. I've added checks for all the
2031 record types that Exim uses.
2033 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2034 not automatically generating a new one, as it is supposed to. This
2035 prevented TLS from working. If the file did exist, but contained invalid
2036 data, a new version was generated, as expected. It was only the case of a
2037 non-existent file that was broken.
2039 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2040 with a change in libdomainkeys > 0.64.
2042 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2043 from DNS. If the selector record carries the flag, it now has
2044 precedence over the domain-wide flag.
2046 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2048 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2049 the use of a port name that isn't defined in /etc/services) occurred, the
2050 message was deferred as in a normal delivery, and thus remained on the
2051 spool, instead of being failed because of the mua_wrapper setting. This
2052 is now fixed, and I tidied up some of the mua_wrapper messages at the
2055 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2056 the messages in a hash of arrays rather than using individual hashes.
2057 This is a bit cleaner and results in dramatic memory savings, albeit
2058 at a slight CPU cost.
2060 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2061 as requested by Marc Sherman.
2063 SC/10 Eximstats: added histograms for user specified patterns as requested
2066 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2068 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2069 fopen() in the content-scanning modules that did not already have it.
2075 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2076 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2078 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2080 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2082 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2083 to test Sieve filters that use "vacation".
2085 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2086 that changes the way the GnuTLS parameters are stored in the cache file.
2087 The new format can be generated externally. For backward compatibility,
2088 if the data in the cache doesn't make sense, Exim assumes it has read an
2089 old-format file, and it generates new data and writes a new file. This
2090 means that you can't go back to an older release without removing the
2093 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2094 work if there are any delivery delays because "one_time" forces the
2095 parent to be marked "delivered", so its unseen clone is never tried
2096 again. For this reason, Exim now forbids the simultaneous setting of
2099 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2100 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2101 entry if you want to know the details. The fix, however, neglected to
2102 consider the case when local delivery batching is involved. The test for
2103 "previously delivered" was not happening when checking to see if an
2104 address could be batched with a previous (undelivered) one; under
2105 certain circumstances this could lead to multiple deliveries to the same
2108 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2109 in its include files, and this causes problems building Exim.
2111 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2112 header_syntax) but Exim was just ignoring anything given after a slash.
2113 In particular, this caused confusion with an attempt to use "verify =
2114 reverse_host_lookup/defer_ok". An error is now given when options are
2115 supplied for verify items that do not have them. (Maybe reverse_host_
2116 lookup should have a defer_ok option, but that's a different point.)
2118 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2119 defined by RFC 821) to 2048, because there were problems with some AUTH
2120 commands, and RFC 1869 says the size should be increased for extended
2121 SMTP commands that take arguments.
2123 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2126 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2127 "unknown" error; now it says that the functionality isn't in the binary.
2129 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2130 an address' error message when a string expansion fails (syntax or
2131 whatever). Otherwise the password may appear in the log. Following change
2132 PH/42 below, there is no longer a chance of it appearing in a bounce
2135 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2137 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2138 of its addresses was ever tried. (Bugzilla bug #2.)
2140 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2141 the result incorrectly in the debug output. (It correctly added a newline
2142 to what was transported.)
2144 TF/01 Added $received_time.
2146 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2147 commented out examples of how to interface to a virus scanner and to
2148 SpamAssassin. Also added commented examples of av_scanner and
2149 spamd_address settings.
2151 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2152 and controls are allowed in which ACLs. There were a couple of minor
2153 errors. Some of the entries in the conditions table (which is a table of
2154 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2155 negation of where the condition IS allowed.
2157 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2159 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2160 header file does not have a version number, so I've had to invent a new
2161 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2162 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2163 radiusclient), but it was contributed by a Radius user.
2165 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2166 files or directories, for OpenSSL.
2168 PH/19 When an Exim process that is running as root has to create an Exim log
2169 file, it does so in a subprocess that runs as exim:exim so as to get the
2170 ownership right at creation (otherwise, other Exim processes might see
2171 the file with the wrong ownership). There was no test for failure of this
2172 fork() call, which would lead to the process getting stuck as it waited
2173 for a non-existent subprocess. Forks do occasionally fail when resources
2174 run out. I reviewed all the other calls to fork(); they all seem to check
2177 PH/20 When checking for unexpected SMTP input at connect time (before writing
2178 the banner), Exim was not dealing correctly with a non-positive return
2179 from the read() function. If the client had disconnected by this time,
2180 the result was a log entry for a synchronization error with an empty
2181 string after "input=" when read() returned zero. If read() returned -1
2182 (an event I could not check), uninitialized data bytes were printed.
2183 There were reports of junk text (parts of files, etc) appearing after
2186 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2188 PH/22 Added support for macro redefinition, and (re)definition in between
2189 driver and ACL definitions.
2191 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2192 forgetting to use the resulting value; it was using the unexpanded value.
2194 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2195 hadn't been configured. The fix is from Juergen Kreileder, who
2196 understands it better than I do:
2198 "Here's what I see happening with three configured cyrus_sasl
2199 authenticators configured (plain, login, cram-md5):
2201 On startup auth_cyrus_sasl_init() gets called for each of these.
2202 This means three calls to sasl_listmech() without a specified mech_list.
2203 => SASL tests which mechs of all available mechs actually work
2204 => three warnings about OTP not working
2205 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2207 With the patch, sasl_listmech() also gets called three times. But now
2208 SASL's mech_list option is set to the server_mech specified in the the
2209 authenticator. Or in other words, the answer from sasl_listmech()
2210 gets limited to just the mech you're testing for (which is different
2212 => the return list contains just 'plain' or 'login', 'cram-md5' or
2213 nothing depending on the value of ob->server_mech.
2215 I've just tested the patch: Authentication still works fine,
2216 unavailable mechs specified in the exim configuration are still
2217 caught, and the auth.log warnings about OTP are gone."
2219 PH/25 When debugging is enabled, the contents of the command line are added
2220 to the debugging output, even when log_selector=+arguments is not
2223 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2224 answer is "GNU", and only if the return is "GNU/something" is the answer
2227 PH/27 $acl_verify_message is now set immediately after the failure of a
2228 verification in an ACL, and so is available in subsequent modifiers. In
2229 particular, the message can be preserved by coding like this:
2231 warn !verify = sender
2232 set acl_m0 = $acl_verify_message
2234 Previously, $acl_verify_message was set only while expanding "message"
2235 and "log_message" when a very denied access.
2237 PH/28 Modified OS/os.c-Linux with
2239 -#ifndef OS_LOAD_AVERAGE
2240 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2242 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2243 nomenclature these days.)
2245 PH/29 Installed patch from the Sieve maintainer that adds the options
2246 sieve_useraddress and sieve_subaddress to the redirect router.
2248 PH/30 In these circumstances:
2249 . Two addresses routed to the same list of hosts;
2250 . First host does not offer TLS;
2251 . First host accepts first address;
2252 . First host gives temporary error to second address;
2253 . Second host offers TLS and a TLS session is established;
2254 . Second host accepts second address.
2255 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2256 and peerdn, if requested) that were in fact used only for the second
2259 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2260 attention to any local part prefix or suffix that was matched by the
2261 router that accepted the address. It now behaves in the same way as it
2262 does for delivery: the affixes are removed from the local part unless
2263 rcpt_include_affixes is set on the transport.
2265 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2266 timeout during the DATA phase of an incoming message.
2268 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2269 applied a patch sanctioned by the Sieve maintainer.
2271 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2272 the uid or gid is negative. A case of a negative gid caused this to be
2273 noticed. The fix allows for either to be negative.
2275 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2276 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2279 PH/36 The change PH/12 above was broken. Fixed it.
2281 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2282 the grounds that routing the same address twice would always produce the
2283 same answer. This might have been true once, but it is certainly no
2284 longer true now. Routing a child address may depend on the previous
2285 routing that produced that child. Some complicated redirection strategies
2286 went wrong when messages had multiple recipients, and made Exim's
2287 behaviour dependent on the order in which the addresses were given.
2289 I have moved the duplicate checking until after the routing is complete.
2290 Exim scans the addresses that are assigned to local and remote
2291 transports, and removes any duplicates. This means that more work will be
2292 done, as duplicates will always all be routed, but duplicates are
2293 presumably rare, so I don't expect this is of any significance.
2295 For deliveries to pipes, files, and autoreplies, the duplicate checking
2296 still happens during the routing process, since they are not going to be
2299 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2300 It corrects a timeout issue with spamd. This is Ian's comment: "The
2301 background is that sometimes spamd either never reads data from a
2302 connection it has accepted, or it never writes response data. The exiscan
2303 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2304 blindly assumes that writes won't block so it may never time out."
2306 PH/39 Allow G after quota size as well as K and M.
2308 PH/40 The value set for $authenticated_id in an authenticator may not contain
2309 binary zeroes or newlines because the value is written to log lines and
2310 to spool files. There was no check on this. Now the value is run through
2311 the string_printing() function so that such characters are converted to
2312 printable escape sequences.
2314 PH/41 $message_linecount is a new variable that contains the total number of
2315 lines in the message. Compare $body_linecount, which is the count for the
2318 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2319 bounce and delay warning messages, except in certain special cases, which
2322 (a) An SMTP error message from a remote host;
2323 (b) A message specified in a :fail: redirection;
2324 (c) A message specified in a "fail" command in a system filter;
2325 (d) A message specified in a FAIL return from the queryprogram router;
2326 (e) A message specified by the cannot_route_message router option.
2328 In these cases only, Exim does include the error details in bounce and
2329 warning messages. There are also a few cases where bland messages such
2330 as "unrouteable address" or "local delivery error" are given.
2332 PH/43 $value is now also set for the "else" part of a ${run expansion.
2334 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2335 being worked on, but at least Exim now implements the latest version to
2338 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2339 process to complete was treated as a delivery failure, a timeout while
2340 writing the message to the pipe was logged, but erroneously treated as a
2341 successful delivery. Such timeouts include transport filter timeouts. For
2342 consistency with the overall process timeout, these timeouts are now
2343 treated as errors, giving rise to delivery failures by default. However,
2344 there is now a new Boolean option for the pipe transport called
2345 timeout_defer, which, if set TRUE, converts the failures into defers for
2346 both kinds of timeout. A transport filter timeout is now identified in
2349 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2350 systems where "make" and "gmake" are different, calling "gmake" at top
2351 level broke things. I've arranged for the value of $(MAKE) to be passed
2352 from the Makefile to this script so that it can call the same version of
2356 A note about Exim versions 4.44 and 4.50
2357 ----------------------------------------
2359 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2360 changes of various kinds. As a consequence, a big documentation update was
2361 needed. This delayed the release for rather longer than seemed good, especially
2362 in the light of a couple of (minor) security issues. Therefore, the changes
2363 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2364 release. So 4.44 and 4.50 are in effect two different branches that both start
2367 I have left the 4.50 change log unchanged; it contains all the changes since
2368 4.43. The change log for 4.44 is below; many of its items are identical to
2369 those for 4.50. This seems to be the most sensible way to preserve the
2370 historical information.
2376 1. Minor wording change to the doc/README.SIEVE file.
2378 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2379 computation of the current number of files was incorrect.
2381 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2382 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2385 4. Give more explanation in the error message when the command for a transport
2386 filter fails to execute.
2388 5. There are several places where Exim runs a non-Exim command in a
2389 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2390 was being done only for the command run by the queryprogram router. It is
2391 now done for all such subprocesses. The other cases are: ${run, transport
2392 filters, and the commands run by the lmtp and pipe transports.
2394 6. Added CONFIGURE_GROUP build-time option.
2396 7. Some older OS have a limit of 256 on the maximum number of file
2397 descriptors. Exim was using setrlimit() to set 1000 as a large value
2398 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2399 systems. I've change it so that if it can't get 1000, it tries for 256.
2401 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2402 was an oversight, and furthermore, ever since the addition of extra
2403 controls (e.g. 4.43/32), the checks on when to allow different forms of
2404 "control" were broken. There should now be diagnostics for all cases when a
2405 control that does not make sense is encountered.
2407 9. Added the /retain_sender option to "control=submission".
2409 10. $recipients is now available in the predata ACL (oversight).
2411 11. Tidy the search cache before the fork to do a delivery from a message
2412 received from the command line. Otherwise the child will trigger a lookup
2413 failure and thereby defer the delivery if it tries to use (for example) a
2414 cached ldap connection that the parent has called unbind on.
2416 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2417 of $address_data from the recipient verification was clobbered by the
2418 sender verification.
2420 13. The value of address_data from a sender verification is now available in
2421 $sender_address_data in subsequent conditions in the ACL statement.
2423 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2425 15. Added a new option "connect=<time>" to callout options, to set a different
2428 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2429 was its contents. (It was OK if the option was not defined at all.)
2431 17. A "Completed" log line is now written for messages that are removed from
2432 the spool by the -Mrm option.
2434 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2435 information about exactly what failed.
2437 19. Added -dd to debug only the daemon process.
2439 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2440 handles timeouts, both on the server side and network timeouts. Renamed the
2441 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2443 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2444 It is now set to "smtps".
2446 22. $host_address is now set to the target address during the checking of
2447 ignore_target_hosts.
2449 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2450 being passed; this would have caused $sender_host_name to have been used if
2451 matching the list had actually called for a host name (not very likely,
2452 since this list is usually IP addresses). A host name is now passed as
2455 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2456 code that specifies a non-restarting handler (typically sigaction() in
2457 modern systems) in an attempt to fix a rare and obscure crash bug.
2459 25. Narrowed the window for a race in the daemon that could cause it to ignore
2460 SIGCHLD signals. This is not a major problem, because they are used only to
2461 wake it up if nothing else does.
2463 26. A malformed maildirsize file could cause Exim to calculate negative values
2464 for the mailbox size or file count. Odd effects could occur as a result.
2465 The maildirsize information is now recalculated if the size or filecount
2468 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2469 support for a long time. Removed HAVE_SYS_VFS_H.
2471 28. Installed the latest version of exipick from John Jetmore.
2473 29. In an address list, if the pattern was not a regular expression, an empty
2474 subject address (from a bounce message) matched only if the pattern was an
2475 empty string. Non-empty patterns were not even tested. This was the wrong
2476 because it is perfectly reasonable to use an empty address as part of a
2477 database query. An empty address is now tested by patterns that are
2478 lookups. However, all the other forms of pattern expect the subject to
2479 contain a local part and a domain, and therefore, for them, an empty
2480 address still always fails if the pattern is not itself empty.
2482 30. Exim went into a mad DNS loop when attempting to do a callout where the
2483 host was specified on an smtp transport, and looking it up yielded more
2484 than one IP address.
2486 31. Re-factored the code for checking spool and log partition space into a
2487 function that finds that data and another that does the check. The former
2488 is then used to implement four new variables: $spool_space, $log_space,
2489 $spool_inodes, and $log_inodes.
2491 32. The RFC2047 encoding function was originally intended for short strings
2492 such as real names; it was not keeping to the 75-character limit for
2493 encoded words that the RFC imposes. It now respects the limit, and
2494 generates multiple encoded words if necessary. To be on the safe side, I
2495 have increased the buffer size for the ${rfc2047: expansion operator from
2498 33. It is now permitted to omit both strings after an "if" condition; if the
2499 condition is true, the result is "true". As before, when the second string
2500 is omitted, a false condition yields an empty string. This makes it less
2501 cumbersome to write custom ACL and router conditions.
2503 34. Failure to deliver a bounce message always caused it to be frozen, even if
2504 there was an errors_to setting on the router. The errors_to setting is now
2507 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2508 canonical form (fully expanded) before being placed in
2509 $sender_host_address.
2511 36. The table in the code that translates DNS record types into text (T_A to
2512 "A" for instance) was missing entries for NS and CNAME. It is just possible
2513 that this could have caused confusion if both these types were looked up
2514 for the same domain, because the text type is used as part of Exim's
2515 per-process caching. But the chance of anyone hitting this buglet seems
2518 37. The dnsdb lookup has been extended in a number of ways.
2520 (1) There is a new type, "zns", which walks up the domain tree until it
2521 finds some nameserver records. It should be used with care.
2523 (2) There is a new type, "mxh", which is like "mx" except that it returns
2524 just the host names, not the priorities.
2526 (3) It is now possible to give a list of domains (or IP addresses) to be
2527 looked up. The behaviour when one of the lookups defers can be
2528 controlled by a keyword.
2530 (4) It is now possible to specify the separator character for use when
2531 multiple records are returned.
2533 38. The dnslists ACL condition has been extended: it is now possible to supply
2534 a list of IP addresses and/or domains to be looked up in a particular DNS
2537 39. Added log_selector=+queue_time_overall.
2539 40. When running the queue in the test harness, wait just a tad after forking a
2540 delivery process, to get repeatability of debugging output.
2542 41. Include certificate and key file names in error message when GnuTLS fails
2543 to set them up, because the GnuTLS error message doesn't include the name
2544 of the failing file when there is a problem reading it.
2546 42. Allow both -bf and -bF in the same test run.
2548 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2550 44. The "Exiscan patch" is now merged into the mainline Exim source.
2552 45. Sometimes the final signoff response after QUIT could fail to get
2553 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2554 < 0 before doing a fflush(). This bug looks as though it goes back to the
2555 introduction of TLS in release 3.20, but "sometimes" must have been rare
2556 because the tests only now provoked it.
2558 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2559 (this can affect the format of dates).
2561 47. exim_tidydb, when checking for the continued existence of a message for
2562 which it has found a message-specific retry record, was not finding
2563 messages that were in split spool directories. Consequently, it was
2564 deleting retry records that should have stayed in existence.
2566 48. Steve fixed some bugs in eximstats.
2568 49. The SPA authentication driver was not abandoning authentication and moving
2569 on to the next authenticator when an expansion was forced to fail,
2570 contradicting the general specification for all authenticators. Instead it
2571 was generating a temporary error. It now behaves as specified.
2573 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2574 (the order specifies the preference for clients). The order is now AES256,
2575 AES128, 3DES, ARCFOUR128.
2577 51. Small patch to Sieve code - explicitly set From: when generating an
2580 52. Exim crashed if a remote delivery caused a very long error message to be
2581 recorded - for instance if somebody sent an entire SpamAssassin report back
2582 as a large number of 550 error lines. This bug was coincidentally fixed by
2583 increasing the size of one of Exim's internal buffers (big_buffer) that
2584 happened as part of the Exiscan merge. However, to be on the safe side, I
2585 have made the code more robust (and fixed the comments that describe what
2588 53. Now that there can be additional text after "Completed" in log lines (if
2589 the queue_time_overall log selector is set), a one-byte patch to exigrep
2590 was needed to allow it to recognize "Completed" as not the last thing in
2593 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2594 patch that reportedly fixes this has been added. I am not expert enough to
2595 create a test for it. This is what the patch creator wrote:
2597 "I found a little strange behaviour of ldap code when working with
2598 Windows 2003 AD Domain, where users was placed in more than one
2599 Organization Units. When I tried to give exim partial DN, the exit code
2600 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2601 But simultaneously result of request was absolutely normal ldap result,
2602 so I produce this patch..."
2604 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2605 REFERENCE, so I have modified the code to exclude the patch when that macro
2608 55. Some experimental protocols are using DNS PTR records for new purposes. The
2609 keys for these records are domain names, not reversed IP addresses. The
2610 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2611 leaves it alone. Component reversal etc. now happens only for IP addresses.
2614 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2616 57. Double the size of the debug message buffer (to 2048) so that more of very
2617 long debug lines gets shown.
2619 58. The exicyclog utility now does better if the number of log files to keep
2620 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2622 59. Two changes related to the smtp_active_hostname option:
2624 (1) $smtp_active_hostname is now available as a variable.
2625 (2) The default for smtp_banner uses $smtp_active_hostname instead
2626 of $primary_hostname.
2628 60. The host_aton() function is supposed to be passed a string that is known
2629 to be a valid IP address. However, in the case of IPv6 addresses, it was
2630 not checking this. This is a hostage to fortune. Exim now panics and dies
2631 if the condition is not met. A case was found where this could be provoked
2632 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2633 components; fortuitously, this particular loophole had already been fixed
2634 by change 4.50/55 above.
2636 If there are any other similar loopholes, the new check in host_aton()
2637 itself should stop them being exploited. The report I received stated that
2638 data on the command line could provoke the exploit when Exim was running as
2639 exim, but did not say which command line option was involved. All I could
2640 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2641 running as the user.
2644 61. There was a buffer overflow vulnerability in the SPA authentication code
2645 (which came originally from the Samba project). I have added a test to the
2646 spa_base64_to_bits() function which I hope fixes it.
2649 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2650 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2652 63. The daemon start-up calls getloadavg() while still root for those OS that
2653 need the first call to be done as root, but it missed one case: when
2654 deliver_queue_load_max is set with deliver_drop_privilege. This is
2655 necessary for the benefit of the queue runner, because there is no re-exec
2656 when deliver_drop_privilege is set.
2658 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2659 This has been fixed.
2661 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2662 was in use, was not putting the data itself into the right store pool;
2663 consequently, it could be overwritten for a subsequent message in the same
2664 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2667 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2669 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2670 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2671 the function were treating the return as a boolean value, which happened to
2672 work because 0=false and not-0=true, but is not correct code.
2674 68. The host_aton() function was not handling scoped IPv6 addresses (those
2675 with, for example, "%eth0" on the end) correctly.
2677 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2678 negated items (that is, ~something) in unsigned ints. Some compilers
2679 apparently mutter when there is no cast.
2681 70. If an address verification called from an ACL failed, and did not produce a
2682 user-specific message (i.e. there was only a "system" message), nothing was
2683 put in $acl_verify_message. In this situation, it now puts the system
2686 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2687 session; change 4.31/43 added the unwanted input to the log line - except
2688 that it did not do this in the start of session case. It now does.
2690 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2691 This is wrong and can cause the other end to generate a synchronization
2692 error if it is another Exim or anything else that does the synchronization
2693 check. A QUIT command is no longer sent after a timeout.
2695 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2696 during host lookups.
2698 74. The defer_ok option of callout verification was not working if it was used
2699 when verifying addresses in header lines, that is, for this case:
2701 verify = header_sender/callout=defer_ok
2703 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2704 those file descriptors could be used for SMTP connections. If anything
2705 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2706 could be sent to the SMTP client, causing chaos. The daemon now opens
2707 stdin, stdout, and stderr to /dev/null when it puts itself into the
2710 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2711 log by default. The user can override this with suitable Perl magic.
2713 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2714 the log message when discard triggers, was not working for the DATA ACL or
2715 for the non-SMTP ACL.
2717 78. Error message wording change in sieve.c.
2719 79. If smtp_accept_max_per_host was set, the number of connections could be
2720 restricted to fewer than expected, because the daemon was trying to set up
2721 a new connection before checking whether the processes handling previous
2722 connections had finished. The check for completed processes is now done
2723 earlier. On busy systems, this bug wouldn't be noticed because something
2724 else would have woken the daemon, and it would have reaped the completed
2727 80. If a message was submitted locally by a user whose login name contained one
2728 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2729 It caused a spool format error. I have fixed the spool reading code. A
2730 related problem was that the "from" clause in the Received: line became
2731 illegal because of the space(s). It is now covered by ${quote_local_part.
2733 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2736 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2738 83. Patch from Sieve maintainer to fix unterminated string problem in
2739 "vacation" handling.
2741 84. Some minor changes to the Linux configuration files to help with other
2742 OS variants using glibc.
2744 85. One more patch for Sieve to update vacation handling to latest spec.
2747 ----------------------------------------------------
2748 See the note above about the 4.44 and 4.50 releases.
2749 ----------------------------------------------------
2755 1. Change 4.43/35 introduced a bug that caused file counts to be
2756 incorrectly computed when quota_filecount was set in an appendfile
2759 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2760 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2763 3. Give more explanation in the error message when the command for a transport
2764 filter fails to execute.
2766 4. There are several places where Exim runs a non-Exim command in a
2767 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2768 was being done only for the command run by the queryprogram router. It is
2769 now done for all such subprocesses. The other cases are: ${run, transport
2770 filters, and the commands run by the lmtp and pipe transports.
2772 5. Some older OS have a limit of 256 on the maximum number of file
2773 descriptors. Exim was using setrlimit() to set 1000 as a large value
2774 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2775 systems. I've change it so that if it can't get 1000, it tries for 256.
2777 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2778 was an oversight, and furthermore, ever since the addition of extra
2779 controls (e.g. 4.43/32), the checks on when to allow different forms of
2780 "control" were broken. There should now be diagnostics for all cases when a
2781 control that does not make sense is encountered.
2783 7. $recipients is now available in the predata ACL (oversight).
2785 8. Tidy the search cache before the fork to do a delivery from a message
2786 received from the command line. Otherwise the child will trigger a lookup
2787 failure and thereby defer the delivery if it tries to use (for example) a
2788 cached ldap connection that the parent has called unbind on.
2790 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2791 of $address_data from the recipient verification was clobbered by the
2792 sender verification.
2794 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2795 was its contents. (It was OK if the option was not defined at all.)
2797 11. A "Completed" log line is now written for messages that are removed from
2798 the spool by the -Mrm option.
2800 12. $host_address is now set to the target address during the checking of
2801 ignore_target_hosts.
2803 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2804 being passed; this would have caused $sender_host_name to have been used if
2805 matching the list had actually called for a host name (not very likely,
2806 since this list is usually IP addresses). A host name is now passed as
2809 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2810 code that specifies a non-restarting handler (typically sigaction() in
2811 modern systems) in an attempt to fix a rare and obscure crash bug.
2813 15. Narrowed the window for a race in the daemon that could cause it to ignore
2814 SIGCHLD signals. This is not a major problem, because they are used only to
2815 wake it up if nothing else does.
2817 16. A malformed maildirsize file could cause Exim to calculate negative values
2818 for the mailbox size or file count. Odd effects could occur as a result.
2819 The maildirsize information is now recalculated if the size or filecount
2822 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2823 support for a long time. Removed HAVE_SYS_VFS_H.
2825 18. Updated exipick to current release from John Jetmore.
2827 19. Allow an empty sender to be matched against a lookup in an address list.
2828 Previously the only cases considered were a regular expression, or an
2831 20. Exim went into a mad DNS lookup loop when doing a callout where the
2832 host was specified on the transport, if the DNS lookup yielded more than
2835 21. The RFC2047 encoding function was originally intended for short strings
2836 such as real names; it was not keeping to the 75-character limit for
2837 encoded words that the RFC imposes. It now respects the limit, and
2838 generates multiple encoded words if necessary. To be on the safe side, I
2839 have increased the buffer size for the ${rfc2047: expansion operator from
2842 22. Failure to deliver a bounce message always caused it to be frozen, even if
2843 there was an errors_to setting on the router. The errors_to setting is now
2846 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2847 canonical form (fully expanded) before being placed in
2848 $sender_host_address.
2850 24. Updated eximstats to version 1.33
2852 25. Include certificate and key file names in error message when GnuTLS fails
2853 to set them up, because the GnuTLS error message doesn't include the name
2854 of the failing file when there is a problem reading it.
2856 26. Expand error message when OpenSSL has problems setting up cert/key files.
2859 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2860 (this can affect the format of dates).
2862 28. exim_tidydb, when checking for the continued existence of a message for
2863 which it has found a message-specific retry record, was not finding
2864 messages that were in split spool directories. Consequently, it was
2865 deleting retry records that should have stayed in existence.
2867 29. eximstats updated to version 1.35
2868 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2869 1.35 - bugfix such that pie charts by volume are generated correctly
2871 30. The SPA authentication driver was not abandoning authentication and moving
2872 on to the next authenticator when an expansion was forced to fail,
2873 contradicting the general specification for all authenticators. Instead it
2874 was generating a temporary error. It now behaves as specified.
2876 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2877 (the order specifies the preference for clients). The order is now AES256,
2878 AES128, 3DES, ARCFOUR128.
2880 31. Small patch to Sieve code - explicitly set From: when generating an
2883 32. Exim crashed if a remote delivery caused a very long error message to be
2884 recorded - for instance if somebody sent an entire SpamAssassin report back
2885 as a large number of 550 error lines. This bug was coincidentally fixed by
2886 increasing the size of one of Exim's internal buffers (big_buffer) that
2887 happened as part of the Exiscan merge. However, to be on the safe side, I
2888 have made the code more robust (and fixed the comments that describe what
2891 33. Some experimental protocols are using DNS PTR records for new purposes. The
2892 keys for these records are domain names, not reversed IP addresses. The
2893 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2894 leaves it alone. Component reversal etc. now happens only for IP addresses.
2897 34. The host_aton() function is supposed to be passed a string that is known
2898 to be a valid IP address. However, in the case of IPv6 addresses, it was
2899 not checking this. This is a hostage to fortune. Exim now panics and dies
2900 if the condition is not met. A case was found where this could be provoked
2901 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2902 components; fortuitously, this particular loophole had already been fixed
2903 by change 4.50/55 or 4.44/33 above.
2905 If there are any other similar loopholes, the new check in host_aton()
2906 itself should stop them being exploited. The report I received stated that
2907 data on the command line could provoke the exploit when Exim was running as
2908 exim, but did not say which command line option was involved. All I could
2909 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2910 running as the user.
2913 35. There was a buffer overflow vulnerability in the SPA authentication code
2914 (which came originally from the Samba project). I have added a test to the
2915 spa_base64_to_bits() function which I hope fixes it.
2918 36. The daemon start-up calls getloadavg() while still root for those OS that
2919 need the first call to be done as root, but it missed one case: when
2920 deliver_queue_load_max is set with deliver_drop_privilege. This is
2921 necessary for the benefit of the queue runner, because there is no re-exec
2922 when deliver_drop_privilege is set.
2924 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2925 was in use, was not putting the data itself into the right store pool;
2926 consequently, it could be overwritten for a subsequent message in the same
2927 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2930 38. Sometimes the final signoff response after QUIT could fail to get
2931 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2932 < 0 before doing a fflush(). This bug looks as though it goes back to the
2933 introduction of TLS in release 3.20, but "sometimes" must have been rare
2934 because the tests only now provoked it.
2940 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2941 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2942 Now it returns an integer. A number of calls were still expecting a T/F
2943 return. Fortuitously, in all cases, the tests worked in OK situations,
2944 which is the norm. However, things would have gone wrong on any write
2945 failures on the smtp file descriptor. This function is used when sending
2946 messages over SMTP and also when doing verify callouts.
2948 2. When Exim is called to do synchronous delivery of a locally submitted
2949 message (the -odf or -odi options), it no longer closes stderr before doing
2952 3. Implemented the mua_wrapper option.
2954 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2956 5. Implemented the functions header_remove(), header_testname(),
2957 header_add_at_position(), and receive_remove_recipient(), and exported them
2960 6. If an ACL "warn" statement specified the addition of headers, Exim already
2961 inserted X-ACL-Warn: at the start if there was no header name. However, it
2962 was not making this test for the second and subsequent header lines if
2963 there were newlines in the string. This meant that an invalid header could
2964 be inserted if Exim was badly configured.
2966 7. Allow an ACL "warn" statement to add header lines at the start or after all
2967 the Received: headers, as well as at the end.
2969 8. Added the rcpt_4xx retry error code.
2971 9. Added postmaster_mailfrom=xxx to callout verification option.
2973 10. Added mailfrom=xxxx to the callout verification option, for verify=
2976 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2977 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2979 12. Inserted some casts to stop certain compilers warning when using pointer
2980 differences as field lengths or precisions in printf-type calls (mostly
2981 affecting debugging statements).
2983 13. Added optional readline() support for -be (dynamically loaded).
2985 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2986 same clock tick as a message's arrival, so that its received time was the
2987 same as the "first fail" time on the retry record, and that message
2988 remained on the queue past the ultimate address timeout, every queue runner
2989 would try a delivery (because it was past the ultimate address timeout) but
2990 after another failure, the ultimate address timeout, which should have then
2991 bounced the address, did not kick in. This was a "< instead of <=" error;
2992 in most cases the first failure would have been in the next clock tick
2993 after the received time, and all would be well.
2995 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2996 being recognized when the domain list was tested by the match_domain
2997 condition in an expansion string.
2999 16. Added the ${str2b64: operator.
3001 17. Exim was always calling setrlimit() to set a large limit for the number of
3002 processes, without checking whether the existing limit was already
3003 adequate. (It did check for the limit on file descriptors.) Furthermore,
3004 errors from getrlimit() and setrlimit() were being ignored. Now they are
3005 logged to the main and panic logs, but Exim does carry on, to try to do its
3006 job under whatever limits there are.
3008 18. Imported PCRE 5.0.
3010 19. Trivial typo in log message " temporarily refused connection" (the leading
3013 20. If the log selector return_path_on_delivery was set and an address was
3014 redirected to /dev/null, the delivery process crashed because it assumed
3015 that a return path would always be set for a "successful" delivery. In this
3016 case, the whole delivery is bypassed as an optimization, and therefore no
3019 21. Internal re-arrangement: the function for sending a challenge and reading
3020 a response while authentication was assuming a zero-terminated challenge
3021 string. It's now changed to take a pointer and a length, to allow for
3022 binary data in such strings.
3024 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3026 23. Exim was not respecting finduser_retries when seeking the login of the
3027 uid under which it was called; it was always trying 10 times. (The default
3028 setting of finduser_retries is zero.) Also, it was sleeping after the final
3029 failure, which is pointless.
3031 24. Implemented tls_on_connect_ports.
3033 25. Implemented acl_smtp_predata.
3035 26. If the domain in control=submission is set empty, Exim assumes that the
3036 authenticated id is a complete email address when it generates From: or
3037 Sender: header lines.
3039 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3040 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3041 chown and chgrp in /bin and hostname in /usr/bin.
3043 28. Exim was keeping the "process log" file open after each use, just as it
3044 does for the main log. This opens the possibility of it remaining open for
3045 long periods when the USR1 signal hits a daemon. Occasional processlog
3046 errors were reported, that could have been caused by this. Anyway, it seems
3047 much more sensible not to leave this file open at all, so that is what now
3050 29. The long-running daemon process does not normally write to the log once it
3051 has entered its main loop, and it closes the log before doing so. This is
3052 so that log files can straightforwardly be renamed and moved. However,
3053 there are a couple of unusual error situations where the daemon does write
3054 log entries, and I had neglected to close the log afterwards.
3056 30. The text of an SMTP error response that was received during a remote
3057 delivery was being truncated at 512 bytes. This is too short for some of
3058 the long messages that one sometimes sees. I've increased the limit to
3061 31. It is now possible to make retry rules that apply only when a message has a
3062 specific sender, in particular, an empty sender.
3064 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3065 it possible to be selective about when SMTP synchronization is enforced.
3067 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3069 32. Implemented hosts_connection_nolog.
3071 33. Added an ACL for QUIT.
3073 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3076 35. Added mailbox_size and mailbox_filecount to appendfile.
3078 36. Added control = no_multiline_responses to ACLs.
3080 37. There was a bug in the logic of the code that waits for the clock to tick
3081 in the case where the clock went backwards by a substantial amount such
3082 that the microsecond fraction of "now" was more than the microsecond
3083 fraction of "then" (but the whole seconds number was less).
3085 38. Added support for the libradius Radius client library this is found on
3086 FreeBSD (previously only the radiusclient library was supported).
3092 1. When certain lookups returned multiple values in the form name=value, the
3093 quoting of the values was not always being done properly. Specifically:
3094 (a) If the value started with a double quote, but contained no whitespace,
3096 (b) If the value contained whitespace other than a space character (i.e.
3097 tabs or newlines or carriage returns) it was not quoted.
3098 This fix has been applied to the mysql and pgsql lookups by writing a
3099 separate quoting function and calling it from the lookup code. The fix
3100 should probably also be applied to nisplus, ibase and oracle lookups, but
3101 since I cannot test any of those, I have not disturbed their existing code.
3103 2. A hit in the callout cache for a specific address caused a log line with no
3104 reason for rejecting RCPT. Now it says "Previous (cached) callout
3105 verification failure".
3107 3. There was an off-by-one bug in the queryprogram router. An over-long
3108 return line was truncated at 256 instead of 255 characters, thereby
3109 overflowing its buffer with the terminating zero. As well as fixing this, I
3110 have increased the buffer size to 1024 (and made a note to document this).
3112 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3113 when Exim is waiting for an SMTP response from a remote server, Exim
3114 restarts its select() call on the socket, thereby resetting its timeout.
3115 This is not a problem when such interrupts are rare. Somebody set up a cron
3116 job to run exiwhat every 2 minutes, which is less than the normal select()
3117 timeout (5 or 10 minutes). This meant that the select() timeout never
3118 kicked in because it was always reset. I have fixed this by comparing the
3119 time when an interrupt arrives with the time at the start of the first call
3120 to select(). If more time than the timeout has elapsed, the interrupt is
3121 treated as a timeout.
3123 5. Some internal re-factoring in preparation for the addition of Sieve
3124 extensions (by MH). In particular, the "personal" test is moved to a
3125 separate function, and given an option for scanning Cc: and Bcc: (which is
3126 not set for Exim filters).
3128 6. When Exim created an email address using the login of the caller as the
3129 local part (e.g. when creating a From: or Sender: header line), it was not
3130 quoting the local part when it contained special characters such as @.
3132 7. Installed new OpenBSD configuration files.
3134 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3135 try to make them clearer.
3137 9. Callout options, other than the timeout value, were being ignored when
3138 verifying sender addresses in header lines. For example, when using
3140 verify = header_sender/callout=no_cache
3142 the cache was (incorrectly) being used.
3144 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3145 only the Cygwin environment.
3147 11. When return_path_on_delivery was set as a log selector, if different remote
3148 addresses in the same message used different return paths and parallel
3149 remote delivery occurred, the wrong values would sometimes be logged.
3150 (Whenever a remote delivery process finished, the return path value from
3151 the most recently started remote delivery process was logged.)
3153 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3154 lines when AUTH and/or TLS are in use. This is the "received protocol"
3155 field. Exim used to use "asmtp" for authenticated SMTP, without any
3156 indication (in the protocol name) for TLS use. Now it follows the RFC and
3157 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3158 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3159 names appear in log lines as well as in Received: header lines.
3161 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3162 extensions, and comparison tests, and to fix some bugs.
3164 14. Changes to the "personal" filter test:
3166 (1) The test was buggy in that it was just doing the equivalent of
3167 "contains" tests on header lines. For example, if a user's address was
3168 anne@some.where, the "personal" test would incorrectly be true for
3170 To: susanne@some.where
3172 This test is now done by extracting each address from the header in turn,
3173 and checking the entire address. Other tests that are part of "personal"
3174 are now done using regular expressions (for example, to check local parts
3175 of addresses in From: header lines).
3177 (2) The list of non-personal local parts in From: addresses has been
3178 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3179 taken from the Sieve specification recommendations.
3181 (3) If the message contains any header line starting with "List-" it is
3182 treated as non-personal.
3184 (4) The test for "circular" in the Subject: header line has been removed
3185 because it now seems ill-conceived.
3187 15. Minor typos in src/EDITME comments corrected.
3189 16. Installed latest exipick from John Jetmore.
3191 17. If headers_add on a router specified a text string that was too long for
3192 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3193 of string_sprintf() is now avoided.
3195 18. $message_body_size was not set (it was always zero) when running the DATA
3196 ACL and the local_scan() function.
3198 19. For the "mail" command in an Exim filter, no default was being set for
3199 the once_repeat time, causing a random time value to be used if "once" was
3200 specified. (If the value happened to be <= 0, no repeat happened.) The
3201 default is now 0s, meaning "never repeat". The "vacation" command was OK
3202 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3203 (I found it when inspecting the code).
3205 20. There is now an overall timeout for performing a callout verification. It
3206 defaults to 4 times the callout timeout, which applies to individual SMTP
3207 commands during the callout. The overall timeout applies when there is more
3208 than one host that can be tried. The timeout is checked before trying the
3209 next host. This prevents very long delays if there are a large number of
3210 hosts and all are timing out (e.g. when the network connections are timing
3211 out). The value of the overall timeout can be changed by specifying an
3212 additional sub-option for "callout", called "maxwait". For example:
3214 verify = sender/callout=5s,maxwait=20s
3216 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3217 to the end before writing, but this should make it even safer).
3219 22. Exim was forgetting that it had advertised PIPELINING for the second and
3220 subsequent messages on an SMTP connection. It was also not resetting its
3221 memory on STARTTLS and an internal HELO.
3223 23. When Exim logs an SMTP synchronization error within a session, it now
3224 records whether PIPELINING has been advertised or not.
3226 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3227 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3228 rather than long int.
3230 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3232 26. Added the never_mail option to autoreply.
3238 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3239 crash if the getsockname() call failed; this can happen if a connection is
3240 closed very soon after it is established. The problem was simply in the
3241 order in which certain operations were done, causing Exim to try to write
3242 to the SMTP stream before it had set up the file descriptor. The bug has
3243 been fixed by making things happen in the correct order.
3249 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3250 before the connection was closed, thus losing the rejection response.
3252 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3253 some early Solaris releases, but causes trouble in current releases where
3254 socklen_t is defined.
3256 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3259 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3262 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3263 that releasing the top of it at the end releases what was used for sub-
3264 expansions (unless the block got too big). However, discard this block if
3265 the first thing is a variable or header, so that we can use its block when
3266 it is dynamic (useful for very large $message_headers, for example).
3268 6. Lookups now cache *every* query, not just the most recent. A new, separate
3269 store pool is used for this. It can be recovered when all lookup caches are
3270 flushed. Lookups now release memory at the end of their result strings.
3271 This has involved some general refactoring of the lookup sources.
3273 7. Some code has been added to the store_xxx() functions to reduce the amount
3274 of flapping under certain conditions.
3276 8. log_incoming_interface used to affect only the <= reception log lines. Now
3277 it causes the local interface and port to be added to several more SMTP log
3278 lines, for example "SMTP connection from", and rejection lines.
3280 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3282 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3284 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3285 could be overwritten at the end of the current message (or the start of a
3286 new message if it was set in a HELO ACL). The value is now preserved for
3287 the duration of the SMTP connection.
3289 12. If a transport had a headers_rewrite setting, and a matching header line
3290 contained an unqualified address, that address was qualified, even if it
3291 did not match any rewriting rules. The underlying bug was that the values
3292 of the flags that permit the existence of unqualified sender and recipient
3293 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3294 non-local messages, and by -bnq for local messages) were not being
3295 preserved with the message after it was received.
3297 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3298 "next input=" as part of the text comprising the host identity instead of
3299 the correct text. The code was using the same buffer for two different
3300 strings. However, depending on which order the printing function evaluated
3301 its arguments, the bug did not always show up. Under Linux, for example, my
3302 test suite worked just fine.
3304 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3305 doesn't work with some older versions of Perl. It has been changed to "my",
3306 which in any case is probably the better facility to use.
3308 15. A really picky compiler found some instances of statements for creating
3309 error messages that either had too many or two few arguments for the format
3312 16. The size of the buffer for calls to the DNS resolver has been increased
3313 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3314 for addresses that have a lot of PTR records. This alleviates a problem; it
3315 does not fully solve it.
3317 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3318 buffer now truncates the list and logs the incident, which is the same
3319 action as happens when Exim is looking up a host name and its aliases.
3320 Previously in this situation something unpredictable would happen;
3321 sometimes it was "internal error: store_reset failed".
3323 18. If a server dropped the connection unexpectedly when an Exim client was
3324 using GnuTLS and trying to read a response, the client delivery process
3325 crashed while trying to generate an error log message.
3327 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3328 string, for example:
3330 warn message = H1: something\nH2: something
3332 the text was added as a single header line from Exim's point of view
3333 though it ended up OK in the delivered message. However, searching for the
3334 second and subsequent header lines using $h_h2: did not work. This has been
3335 fixed. Similarly, if a system filter added multiple headers in this way,
3336 the routers could not see them.
3338 20. Expanded the error message when iplsearch is called with an invalid key to
3339 suggest using net-iplsearch in a host list.
3341 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3342 ACLs are no longer actually imposed (and a message to that effect is
3345 22. If a "gecos" field in a passwd entry contained escaped characters, in
3346 particular, if it contained a \" sequence, Exim got it wrong when building
3347 a From: or a Sender: header from that name. A second bug also caused
3348 incorrect handling when an unquoted " was present following a character
3349 that needed quoting.
3351 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3352 was not being matched caselessly.
3354 24. Arranged for all hyphens in the exim.8 source to be escaped with
3357 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3358 properly. Recipient callout cache records were still being keyed to include
3359 the sender, even when use_sender was set false. This led to far more
3360 callouts that were necessary. The sender is no longer included in the key
3361 when use_sender is false.
3363 26. Added "control = submission" modifier to ACLs.
3365 27. Added the ${base62d: operator to decode base 62 numbers.
3367 28. dnsdb lookups can now access SRV records.
3369 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3370 the configuration file.
3372 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3373 (-v) mode. This makes the output for a verbose queue run more intelligible.
3375 31. Added a use_postmaster feature to recipient callouts.
3377 32. Added the $body_zerocount variable, containing the number of binary zero
3378 bytes in the message body.
3380 33. The time of last modification of the "new" subdirectory is now used as the
3381 "mailbox time last read" when there is a quota error for a maildir
3384 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3386 35. Added +ignore_unknown as a special item in host lists.
3388 36. Code for decoding IPv6 addresses in host lists is now included, even if
3389 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3390 address was recognized as an IP address, but was then not correctly decoded
3391 into binary, causing unexpected and incorrect effects when compared with
3398 1. Very minor rewording of debugging text in manualroute to say "list of
3399 hosts" instead of "hostlist".
3401 2. If verify=header_syntax was set, and a header line with an unqualified
3402 address (no domain) and a large number of spaces between the end of the
3403 name and the colon was received, the reception process suffered a buffer
3404 overflow, and (when I tested it) crashed. This was caused by some obsolete
3405 code that should have been removed. The fix is to remove it!
3407 3. When running in the test harness, delay a bit after writing a bounce
3408 message to get a bit more predictability in the log output.
3410 4. Added a call to search_tidyup() just before forking a reception process. In
3411 theory, someone could use a lookup in the expansion of smtp_accept_max_
3412 per_host which, without the tidyup, could leave open a database connection.
3414 5. Added the variables $recipient_data and $sender_data which get set from a
3415 lookup success in an ACL "recipients" or "senders" condition, or a router
3416 "senders" option, similar to $domain_data and $local_part_data.
3418 6. Moved the writing of debug_print from before to after the "senders" test
3421 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3422 problems for message scanning, either using a data ACL, or using
3423 local_scan() because the Received: header was not generated till after they
3424 were called (in order to set the time as the time of reception completion).
3425 I have revised the way this works. The header is now generated after the
3426 body is received, but before the ACL or local_scan() are called. After they
3427 are run, the timestamp in the header is updated.
3433 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3434 before starting a queue runner without re-exec. This happened only when
3435 deliver_drop_privilege was set or when the Exim user was set to root. The
3436 effect of the bug was that timeouts during subsequent deliveries caused
3437 crashes instead of being properly handled. The handler is now left at its
3438 default (and expected) setting.
3440 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3441 message, again when deliver_drop_privilege is set or Exim is run as root.
3442 The bug described in (1) was not present in this case, but the tidying up
3443 of the other signals was missing. I have made the two cases consistent.
3445 3. The ignore_target_hosts setting on a manualroute router was being ignored
3446 for hosts that were looked up using the /MX notation.
3448 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3451 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3452 operated on the sender address. After changing the $sender_address to <>
3453 for the sender address verify, Exim was re-instated it as the original
3454 (before rewriting) address, but remembering that it had rewritten it, so it
3455 wasn't rewriting it again. This bug also had the effect of breaking the
3456 sender address verification caching when the sender address was rewritten.
3458 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3459 This has been changed so that if the ip literal address matches
3460 ignore_target_hosts, the router declines.
3462 7. Added expansion conditions match_domain, match_address, and match_local_
3463 part (NOT match_host).
3465 8. The placeholder for the Received: header didn't have a length field set.
3467 9. Added code to Exim itself and to exim_lock to test for a specific race
3468 condition that could lead to file corruption when using MBX delivery. The
3469 issue is with the lockfile that is created in /tmp. If this file is removed
3470 after a process has opened it but before that process has acquired a lock,
3471 there is the potential for a second process to recreate the file and also
3472 acquire a lock. This could lead to two Exim processes writing to the file
3473 at the same time. The added code performs the same test as UW imapd; it
3474 checks after acquiring the lock that its file descriptor still refers to
3475 the same named file.
3477 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3478 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3481 11. Added the smtp_active_hostname option. If used, this will typically be made
3482 to depend on the incoming interface address. Because $interface_address is
3483 not set up until the daemon has forked a reception process, error responses
3484 that can happen earlier (such as "too many connections") no longer contain
3487 12. If an expansion in a condition on a "warn" statement fails because a lookup
3488 defers, the "warn" statement is abandoned, and the next ACL statement is
3489 processed. Previously this caused the whole ACL to be aborted.
3491 13. Added the iplsearch lookup type.
3493 14. Added ident_timeout as a log selector.
3495 15. Added tls_certificate_verified as a log selector.
3497 16. Added a global option tls_require_ciphers (compare the smtp transport
3498 option of the same name). This controls incoming TLS connections.
3500 17. I finally figured out how to make tls_require_ciphers do a similar thing
3501 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3502 before starting the TLS session.
3504 18. Tabs are now shown as \t in -bP output.
3506 19. If the log selector return_path_on_delivery was set, Exim crashed when
3507 bouncing a message because it had too many Received: header lines.
3509 20. If two routers both had headers_remove settings, and the first one included
3510 a superfluous trailing colon, the final name in the first list and the
3511 first name in the second list were incorrectly joined into one item (with a
3512 colon in the middle).
3518 1. Added -C and -D options to the exinext utility, mainly to make it easier
3519 to include in the automated testing, but these could be helpful when
3520 multiple configurations are in use.
3522 2. The exinext utility was not formatting the output nicely when there was
3523 an alternate port involved in the retry record key, nor when there was a
3524 message id as well (for retries that were specific to a specific message
3525 and a specific host). It was also confused by IPv6 addresses, because of
3526 the additional colons they contain. I have fixed the IPv4 problem, and
3527 patched it up to do a reasonable job for IPv6.
3529 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3530 delivery, the log line now contains "pipelined" if PIPELINING was used.
3532 4. An SMTP transport process used to panic and die if the bind() call to set
3533 an explicit outgoing interface failed. This has been changed; it is now
3534 treated in the same way as a connect() failure.
3536 5. A reference to $sender_host_name in the part of a conditional expansion
3537 that was being skipped was still causing a DNS lookup. This no longer
3540 6. The def: expansion condition was not recognizing references to header lines
3541 that used bh_ and bheader_.
3543 7. Added the _cache feature to named lists.
3545 8. The code for checking quota_filecount in the appendfile transport was
3546 allowing one more file than it should have been.
3548 9. For compatibility with Sendmail, the command line option
3556 and sets the incoming protocol and host name (for trusted callers). The
3557 host name and its colon can be omitted when only the protocol is to be set.
3558 Note the Exim already has two private options, -pd and -ps, that refer to
3559 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3560 "s", but I don't think that's a major issue.
3562 10. A number of refactoring changes to the code, none of which should affect
3565 (a) The number of logging options was getting close to filling up the
3566 32-bit word that was used as a bit map. I have split them into two classes:
3567 those that are passed in the argument to log_write(), and those that are
3568 only ever tested independently outside of that function. These are now in
3569 separate 32-bit words, so there is plenty of room for expansion again.
3570 There is no change in the user interface or the logging behaviour.
3572 (b) When building, for example, log lines, the code previously used a
3573 macro that called string_cat() twice, in order to add two strings. This is
3574 not really sufficiently general. Furthermore, there was one instance where
3575 it was actually wrong because one of the argument was used twice, and in
3576 one call a function was used. (As it happened, calling the function twice
3577 did not affect the overall behaviour.) The macro has been replaced by a
3578 function that can join an arbitrary number of extra strings onto a growing
3581 (c) The code for expansion conditions now uses a table and a binary chop
3582 instead of a serial search (which was left over from when there were very
3583 few conditions). Also, it now recognizes conditions like "pam" even when
3584 the relevant support is not compiled in: a suitably worded error message is
3585 given if an attempt is made to use such a condition.
3587 11. Added ${time_interval:xxxxx}.
3589 12. A bug was causing one of the ddress fields not to be passed back correctly
3590 from remote delivery subprocesses. The field in question was not being
3591 subsequently used, so this caused to problems in practice.
3593 13. Added new log selectors queue_time and deliver_time.
3595 14. Might have fixed a bug in maildirsizefile handling that threw up
3596 "unexpected character" debug warnings, and recalculated the data
3597 unnecessarily. In any case, I expanded the warning message to give more
3600 15. Added the message "Restricted characters in address" to the statements in
3601 the default ACL that block characters like @ and % in local parts.
3603 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3604 Three changes have been made:
3606 (a) There was a serious bug; a negative response to MAIL caused the whole
3607 recipient domain to be cached as invalid, thereby blocking all messages
3608 to all local parts at the same domain, from all senders. This bug has
3609 been fixed. The domain is no longer cached after a negative response to
3610 MAIL if the sender used is not empty.
3612 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3615 (c) A new callout option, "use_sender" has been added for people who want
3616 the modified behaviour.
3622 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3625 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3626 indeed breaks things for older releases.
3628 3. Added additional logging to the case where there is a problem reading data
3629 from a filter that is running in a subprocess using a pipe, in order to
3630 try to track down a specific problem.
3632 4. Testing facility fudge: when running in the test harness and attempting
3633 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3634 getting "No route to host". Convert this to a timeout.
3636 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3639 6. Some OS don't have socklen_t but use size_t instead. This affects the
3640 fifth argument of getsockopt() amongst other things. This is now
3641 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3642 can be set for individual OS. I have set it for SunOS5, OSF1, and
3643 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3644 some earlier ones do not.
3646 7. Change 4.30/15 was not doing the test caselessly.
3648 8. The standard form for an IPv6 address literal was being rejected by address
3649 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3650 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3651 this, as well as the form without the "IPv6" on the front (but only when
3652 address literals are enabled, of course).
3654 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3656 10. Exim crashed if a message with an empty sender address specified by -f
3657 encountered a router with an errors_to setting. This could be provoked only
3658 by a command such as
3662 where an empty string was supplied; "<>" did not hit this bug.
3664 11. Installed PCRE release 4.5.
3666 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3667 remained set. It is now erased.
3669 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3670 times from message ids (which are base 36 rather than the normal 62).
3672 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3673 were being counted as actual protocol errors, and logged if the log
3674 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3675 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3676 and DATA following a set of rejected RCPTs do not count as protocol errors.
3677 In other words, Exim assumes they were pipelined, though this may not
3678 actually be the case. Of course, in all cases the client gets an
3679 appropriate error code.
3681 15. If a lookup fails in an ACL condition, a message about the failure may
3682 be available; it is used if testing the ACL cannot continue, because most
3683 such messages specify what the cause of the deferral is. However, some
3684 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3685 that caused an old message to be retained and used if a later statement
3686 caused a defer, replacing the real cause of the deferral.
3688 16. If an IP address had so many PTR records that the DNS lookup buffer
3689 was not large enough to hold them, Exim could crash while trying to process
3690 the truncated data. It now detects and logs this case.
3692 17. Further to 4.21/58, another change has been made: if (and only if) the
3693 first line of a message (the first header line) ends with CRLF, a bare LF
3694 in a subsequent header line has a space inserted after it, so as not to
3695 terminate the header.
3697 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3698 unnecessarily, used atoi() instead of strtol(), and didn't check the
3699 termination when getting file sizes from file names by regex.
3701 19. Completely re-implemented the support for maildirsize files, in the light
3702 of a number of problems with the previous contributed implementation
3703 (4.30/29). In particular:
3705 . If the quota is zero, the maildirsize file is maintained, but no quota is
3708 . If the maildir directory does not exist, it is created before any attempt
3709 to write a maildirsize file.
3711 . The quota value in the file is just a cache; if the quota is changed in
3712 the transport, the new value overrides.
3714 . A regular expression is available for excluding directories from the
3717 20. The autoreply transport checks the characters in options that define the
3718 message's headers; it allows continued headers, but it was checking with
3719 isspace() after an embedded newline instead of explicitly looking for a
3722 21. If all the "regular" hosts to which an address was routed had passed their
3723 expiry times, and had not reached their retry times, the address was
3724 bounced, even if fallback hosts were defined. Now Exim should go on to try
3727 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3728 equivalent code in the SMTP transport. Some hosts send humungous responses
3729 to HELO/EHLO, more than 1024 it seems.
3731 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3732 gives compiler warnings in some environments. I've now done it "properly",
3735 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3736 (because of problems with the built-in one) was declared to return uschar *
3737 instead of char *, causing compiler failure.
3739 25. Fixed a file descriptor leak when processing alias/forward files.
3741 26. Fixed a minor format string issue in dbfn.c.
3743 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3745 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3746 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3749 29. When a sender address is verified, it is cached, to save repeating the test
3750 when there is more than one recipient in a message. However, when the
3751 verification involves a callout, it is possible for different callout
3752 options to be set for different recipients. It is too complicated to keep
3753 track of this in the cache, so now Exim always runs a verification when a
3754 callout is required, relying on the callout cache for the optimization.
3755 The overhead is duplication of the address routing, but this should not be
3758 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3759 to be verified with callout=postmaster, and the main callout worked but the
3760 postmaster check failed, the verification correctly failed. However, if a
3761 subsequent RCPT command asked for sender verification *without* the
3762 postmaster check, incorrect caching caused this verification also to fail,
3765 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3766 it was not caching the DNS options (qualify_single, search_parents) that
3767 were used when the lookup failed. A subsequent lookup with different
3768 options therefore always gave the same answer, though there were cases
3769 where it should not have. (Example: a "domains = !$mx_any" option on a
3770 dnslookup router: the "domains" option is always processed without any
3771 widening, but the router might have qualify_single set.) Now Exim uses the
3772 cached value only when the same options are set.
3774 32. Added John Jetmore's "exipick" utility to the distribution.
3776 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3777 than a timeout (e.g. a certificate is required, and is not provided), an
3778 Exim server now closes the connection immediately. Previously it waited for
3779 the client to close - but if the client is SSL, it seems that they each
3780 wait for each other, leading to a delay before one of them times out.
3782 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3783 maintained 0.8.x compatibility because I don't think many are using it, and
3784 it is clearly obsolete.
3786 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3789 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3790 client certificate was expired. A simple patch fixes this, though I don't
3791 understand the full logic of why the verify callback is called multiple
3794 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3795 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3796 which causes problems with some clients (such as the Certicom SSL Plus
3797 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3798 disables the coutermeasure allowing Eudora to connect."
3800 38. Exim was not checking that a write() to a log file succeeded. This could
3801 lead to Bad Things if a log got too big, in particular if it hit a file
3802 size limit. Exim now panics and dies if it cannot write to a log file, just
3803 as it does if it cannot open a log file.
3805 39. Modified OS/Makefile-Linux so that it now contains
3807 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3809 The two -D definitions ensure that Exim is compiled with large file
3810 support, which makes it possible to handle log files that are bigger than
3813 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3814 instance) a domain was checked against a named list that involved a lookup,
3815 causing $domain_data to be set, then another domain was checked against the
3816 same list, then the first domain was re-checked, the value of $domain_data
3817 after the final check could be wrong. In particular, if the second check
3818 failed, it could be set empty. This bug probably also applied to
3821 41. The strip_trailing_dot option was not being applied to the address given
3822 with the -f command-line option.
3824 42. The code for reading a message's header from the spool was incrementing
3825 $received_count, but never initializing it. This meant that the value was
3826 incorrect (doubled) while delivering a message in the same process in which
3827 it was received. In the most common configuration of Exim, this never
3828 happens - a fresh exec is done - but it can happen when
3829 deliver_drop_privilege is set.
3831 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3832 it now includes up to 150 characters of the unexpected data in the log
3835 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3836 and building data strings. The size of both of these buffers was 10 000
3837 bytes - far larger than anybody would *ever* want, thought I. Needless to
3838 say, somebody hit the limit. I have increased the maximum line length to
3839 20 000 and the maximum data length of concatenated lines to 100 000. I have
3840 also fixed two bugs, because there was no checking on these buffers. Tsk,
3841 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3842 buffer is too small.
3844 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3845 lsearch lookups. Now it does.
3847 46. When parsing a route_list item in a manualroute router, a fixed-length
3848 buffer was used for the list of hosts. I made this 1024 bytes long,
3849 thinking that nobody would ever have a list of hosts that long. Wrong.
3850 Somebody had a whole pile of complicated expansion conditions, and the
3851 string was silently truncated, leading to an expansion error. It turns out
3852 that it is easier to change to an unlimited length (owing to other changes
3853 that have happened since this code was originally written) than to build
3854 structure for giving a limitation error. The length of the item that
3855 expands into the list of hosts is now unlimited.
3857 47. The lsearch lookup could not handle data where the length of text line was
3858 more than 4095 characters. Such lines were truncated, leading to shortened
3859 data being returned. It should now handle lines of any length.
3861 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3862 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3865 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3866 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3867 no longer happens in comment lines. A list of replacements is now placed
3868 at the head of all of the source files, except those whose only change is
3869 to replace PERL_COMMAND in the very first #! line.
3871 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3872 messages on the queue, with a bottom-up merge sort, using code contributed
3873 by Michael Haardt. This should make operations like -bp somewhat faster on
3874 large queues. It won't affect queue runners, except when queue_run_in_order
3877 51. Installed eximstats 1.31 in the distribution.
3879 52. Added support for SRV lookups to the dnslookup router.
3881 53. If an ACL referred to $message_body or $message_body_end, the value was not
3882 reset for any messages that followed in the same SMTP session.
3884 54. The store-handling optimization for building very long strings was not
3885 differentiating between the different store pools. I don't think this
3886 actually made any difference in practice, but I've tidied it.
3888 55. While running the routers to verify a sender address, $sender_address
3889 was still set to the sender address. This is wrong, because when routing to
3890 send a bounce to the sender, it would be empty. Therefore, I have changed
3891 it so that, while verifying a sender address, $sender_address is set to <>.
3892 (There is no change to what happens when verifying a recipient address.)
3894 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3895 target A or AAAA records (if not already returned) without resetting the
3896 qualify_single or search_parents options of the DNS resolver. These are
3897 inappropriate in this case because the targets of MX and SRV records must
3898 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3899 target that, when qualified, matched something in the local domain. These
3900 two options are now turned off when doing these lookups.
3902 57. It seems that at least some releases of Reiserfs (which does not have the
3903 concept of a fixed number of inodes) returns zero and not -1 for the
3904 number of available inodes. This interacted badly with check_spool_inodes,
3905 which assumed that -1 was the "no such thing" setting. What I have done is
3906 to check that the total number of inodes is greater than zero before doing
3907 the test of how many are available.
3909 58. When a "warn" ACL statement has a log_message modifier, the message is
3910 remembered, and not repeated. This is to avoid a lot of repetition when a
3911 message has many recipients that cause the same warning to be written.
3912 Howewer, Exim was preserving the list of already written lines for an
3913 entire SMTP session, which doesn't seem right. The memory is now reset if a
3914 new message is started.
3916 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3917 debugging output unless log_rewrite was also set.
3919 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3920 of (int)(handle) when we know that handle contains (void *)(-1).
3922 61. The Exim daemon panic-logs an error return when it closes the incoming
3923 connection. However "connection reset by peer" seems to be common, and
3924 isn't really an error worthy of noting specially, so that particular error
3927 62. When Exim is trying to find all the local interfaces, it used to panic and
3928 die if the ioctl to get the interface flags failed. However, it seems that
3929 on at least one OS (Solaris 9) it is possible to have an interface that is
3930 included in the list of interfaces, but for which you get a failure error
3931 for this call. This happens when the interface is not "plumbed" into a
3932 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3933 failure of the "get flags" call assumes that the interface is down.
3935 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3936 makes life easier for people who are doing arithmetic on fields extracted
3937 from dates, where you often get leading zeros that should not be
3938 interpreted as octal.
3940 64. Added qualify_domain to the redirect router, to override the global
3943 65. If a pathologically long header line contained very many addresses (the
3944 report of this problem mentioned 10 000) and each of them was rewritten,
3945 Exim could use up a very large amount of memory. (It kept on making new
3946 copies of the header line as it rewrote, and never released the old ones.)
3947 At the expense of a bit more processing, the header rewriting function has
3948 been changed so that it no longer eats memory in this way.
3950 66. The generation of the Received: header has been moved from the time that a
3951 message starts to be received, to the time that it finishes. The timestamp
3952 in the Received: header should now be very close to that of the <= log
3953 line. There are two side-effects of this change:
3955 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3956 logged header lines no longer include the local Received: line, because
3957 it has not yet been created. The same applies to a copy of the message
3958 that is returned to a non-SMTP sender when a message is rejected.
3960 (b) When a filter file is tested using -bf, no additional Received: header
3961 is added to the test message. After some thought, I decided that this
3964 This change does not affect the value of $received_for. It is still set
3965 after address rewriting, but before local_scan() is called.
3967 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3969 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3970 gave an unhelpful panic error message, and a defer error. I have managed to
3971 change this behaviour so that it now rejects any supplied certificate,
3972 which seems right, as the list of acceptable certificates is empty.
3974 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3975 gave an unhelpful defer error. I have not managed to make this reject any
3976 supplied certificates, but the error message it gives is "no certificate
3977 supplied", which is not helpful.
3979 70. exigrep's output now also includes lines that are not associated with any
3980 message, but which match the given pattern. Implemented by a patch from
3981 Martin Sluka, which also tidied up the Perl a bit.
3983 71. Recipient callout verification, like sender verification, was using <> in
3984 the MAIL FROM command. This isn't really the right thing, since the actual
3985 sender may affect whether the remote host accepts the recipient or not. I
3986 have changed it to use the actual sender in the callout; this means that
3987 the cache record is now keyed on a recipient/sender pair, not just the
3988 recipient address. There doesn't seem to be a real danger of callout loops,
3989 since a callout by the remote host to check the sender would use <>.
3990 [SEE ABOVE: changed after hitting problems.]
3992 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3993 temporary errors. However, in the case of such a code being given after
3994 the end of a data transmission (i.e. after ".") Exim was failing to write
3995 a retry record for the message. (Yes, there was some broken host that was
3996 actually sending 8xx at this point.)
3998 73. An unknown lookup type in a host list could cause Exim to panic-die when
3999 the list was checked. (An example that provoked this was putting <; in the
4000 middle of a list instead of at the start.) If this happened during a DATA
4001 ACL check, a -D file could be left lying around. This kind of configuration
4002 error no longer causes Exim to die; instead it causes a defer errror. The
4003 incident is still logged to the main and panic logs.
4005 74. Buglet left over from Exim 3 conversion. The message "too many messages
4006 in one connection" was written to the rejectlog but not the mainlog, except
4007 when address rewriting (yes!) was being logged.
4009 75. Added write_rejectlog option.
4011 76. When a system filter was run not as root (that is, when system_filter_user
4012 was set), the values of the $n variables were not being returned to the
4013 main process; thus, they were not subsequently available in the $sn
4016 77. Added +return_path_on_delivery log selector.
4018 78. A connection timeout was being treated differently from recipients deferred
4019 when testing hosts_max_try with a message that was older than the host's
4020 retry timeout. (The host should not be counted, thus allowing all hosts to
4021 be tried at least once before bouncing.) This may have been the cause of an
4022 occasionally reported bug whereby a message would remain on the queue
4023 longer than the retry timeout, but would be bounced if a delivery was
4024 forced. I say "may" because I never totally pinned down the problem;
4025 setting up timeout/retry tests is difficult. See also the next item.
4027 79. The ultimate address timeout was not being applied to errors that involved
4028 a combination of host plus message (for example, a timeout on a MAIL
4029 command). When an address resolved to a number of possible hosts, and they
4030 were not all tried for each delivery (e.g. because of hosts_max_try), a
4031 message could remain on the queue longer than the retry timeout.
4033 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4036 81. Fixed an obscure SMTP outgoing bug which required at least the following
4037 conditions: (a) there was another message waiting for the same server;
4038 (b) the server returned 5xx to all RCPT commands in the first message so
4039 that the message was not completed; (c) the server dropped the connection
4040 or gave a negative response to the RSET that Exim sends to abort the
4041 transaction. The observed case was a dropped connection after DATA that had
4042 been sent in pipelining mode. That is, the server had advertised PIPELINING
4043 but was not implementing it correctly. The effect of the bug was incorrect
4044 behaviour, such as trying another host, and this could lead to a crash.
4050 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4051 and daemon.c were passed as pointers to ints; they should have been
4052 pointers to socklen_t variables (which are typically unsigned ints).
4054 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4057 3. Fixed a really odd bug that affected only the testing scheme; patching a
4058 certain fixed string in the binary changed the value of another string that
4059 happened to be identical to the end of the original first string.
4061 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4062 name", it returns that address as the IP address. On some operating
4063 systems (e.g. Solaris), it also passes back the IP address string as the
4064 "host name". However, on others (e.g. Linux), it passes back an empty
4065 string. Exim wasn't checking for this, and was changing the host name to an
4066 empty string, assuming it had been canonicized.
4068 5. Although rare, it is permitted to have more than one PTR record for a given
4069 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4070 all the names associated with an address, because they do in Solaris.
4071 However, it seems that they do not in Linux for data that comes from the
4072 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4073 I found this out when I moved to a new Linux workstation and tried to run
4074 the Exim test suite.
4076 To get round this problem I have changed the code so that it now does its
4077 own call to the DNS to look up PTR records when searching for a host name.
4078 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4079 addresses that are only in /etc/hosts are still found.
4081 This behaviour is, however, controlled by an option called host_lookup_
4082 order, which defaults to "bydns:byaddr". If people want to use the other
4083 order, or indeed, just use one or the other means of lookup, they can
4084 specify it in this variable.
4086 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4087 some operating systems, this comes back from gethostbyaddr() as an empty
4088 string, and this is what Exim used to test for. However, it seems that in
4089 other systems, "." is yielded. Exim now tests for this case too.
4091 7. The values of check_spool_space and check_log_space are now held internally
4092 as a number of kilobytes instead of an absolute number of bytes. If a
4093 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4094 kilobyte. This means that much larger values can be stored.
4096 8. Exim monitor: an attempt to get the action menu when not actually pointing
4097 at a message produces an empty menu entitled "No message selected". This
4098 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4099 no entries in it ("Shell widget menu has zero width and/or height"). So I
4100 have added a single, blank menu entry in this case.
4102 9. Added ${quote_local_part.
4104 10. MIME decoding is now applied to the contents of Subject: header lines when
4107 11. Now that a reference to $sender_host_address automatically causes a reverse
4108 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4109 host lookup before query-style lookups in lists that might use this
4110 variable. This has therefore been abolished, and the "net-" prefix is no
4111 longer necessary for query-style lookups.
4113 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4114 have been a typo for LFLAGS, so it has been changed.
4116 13. The install script calls Exim with "-C /dev/null" in order to find the
4117 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4118 to be output. Howeve, since Exim outputs its version number before the
4119 error, it didn't break the script. It just looked ugly. I fixed this by
4120 always allowing "-C /dev/null" if the caller is root.
4122 14. Ignore overlarge ACL variable number when reading spool file - insurance
4123 against a later release with more variables having written the file.
4125 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4126 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4127 this, as well as the form without the "IPv6" on the front.
4129 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4130 OS/Makefile-Darwin file.
4132 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4133 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4135 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4136 variable. However, some OS use ints for inodes. I've added cast to long int
4137 to get rid of the compiler warning.
4139 19. I had forgotten to lock out "/../" in configuration file names when
4140 ALT_CONFIG_PREFIX was set.
4142 20. Routers used for verification do not need to specify transports. However,
4143 if such a router generated a host list, and callout was configured, Exim
4144 crashed, because it could not find a port number from the (non-existent)
4145 transport. It now assumes port 25 in this circumstance.
4147 21. Added the -t option to exigrep.
4149 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4150 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4151 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4152 them is set, LOOKUP_LSEARCH is forced.
4154 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4156 24. Added sender and host information to the "rejected by local_scan()" log
4157 line; previously there was no indication of these.
4159 25. Added .include_if_exists.
4161 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4162 Linux. It turns out that not all file systems support this. Apparently some
4163 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4164 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4165 not supported on the file descriptor, is now ignored when Exim is trying to
4166 sync a directory. This applies only to Linux.
4168 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4170 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4171 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4174 29. Added support for maildirsize files from supplied patch (modified a bit).
4176 30. The use of :fail: followed by an empty string could lead Exim to respond to
4177 sender verification failures with (e.g.):
4179 550 Verification failed for <xxx>
4180 550 Sender verify failed
4182 where the first response line was missing the '-' that indicates it is not
4183 the final line of the response.
4185 31. The loop for finding the name of the user that called Exim had a hardwired
4186 limit of 10; it now uses the value of finduser_retries, which is used for
4187 all other user lookups.
4189 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4192 33. Exim was neglecting to zero errno before one call of strtol() when
4193 expanding a string and expecting an integer value. On some systems this
4194 resulted in spurious "integer overflow" errors. Also, it was casting the
4195 result into an int without checking.
4197 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4198 did not work. The code looks as if it has *never* worked, though it appears
4199 to have been documented since at least releast 1.62. I have made it work.
4201 35. The "timeout_DNS" error in retry rules, also documented since at least
4202 1.62, also never worked. As it isn't clear exactly what this means, and
4203 clearly it isn't a major issue, I have abolished the feature by treating it
4204 as "timeout", and writing a warning to the main and panic logs.
4206 36. The display of retry rules for -brt wasn't always showing the error code
4209 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4210 timeout_connect_A, timeout_connect_MX.
4212 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4213 to the empty sender.
4215 39. The daemon was not analysing the content of -oX till after it had closed
4216 stderr and disconnected from the controlling terminal. This meant that any
4217 syntax errors were only noted on the panic log, and the return code from
4218 the command was 0. By re-arranging the code a little, I've made the
4219 decoding happen first, so such errors now appear on stderr, and the return
4220 code is 1. However, the actual setting up of the sockets still happens in
4221 the disconnected process, so errors there are still only recorded on the
4224 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4225 connections (as happens on some IP stacks) was logged at start up time as
4226 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4227 it from "IPv6 and IPv4", which means that two separate sockets are being
4230 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4231 says whether AF_INET or AF_INET6 was passed as an argument.
4233 42. Exiwhat output was messed up when time zones were included in log
4236 43. Exiwhat now gives more information about the daemon's listening ports,
4237 and whether -tls-on-connect was used.
4239 44. The "port" option of the smtp transport is now expanded.
4241 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4242 silently ignored. Now an error message is written to the main and panic
4245 46. There's a new ACL modifier called "logwrite" which writes to a log file
4246 as soon as it is encountered.
4248 47. Added $local_user_uid and $local_user_gid at routing time.
4250 48. Exim crashed when trying to verify a sender address that was being
4253 49. Exim was recognizing only a space character after ".include". It now also
4254 recognizes a tab character.
4256 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4257 extracting the relevant information from the specification. The man page no
4258 longer contains scrambled data for the -d option, and I've added a section
4259 at the front about calling Exim under different names.
4261 51. Added "extra_headers" argument to the "mail" command in filter files.
4263 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4266 53. Installed eximstats 1.29.
4268 54. Added transport_filter_timeout as a generic transport option.
4270 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4271 Cc: header lines. This was required by RFC 822, but it not required by RFC
4274 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4275 incoming messages that did not have them. Now it does so only if the
4276 message originates locally, that is, if there is no associated remote host
4277 address. When Resent- header lines are present, this applies to the Resent-
4278 lines rather than the non-Resent- lines.
4280 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4281 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4283 58. Messages for configuration errors now include the name of the main
4284 configuration file - useful now that there may be more than one file in a
4285 list (.included file names were always shown).
4287 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4288 for those rare installations that do not start the daemon as root or run it
4289 setuid root. I've cut out the call to initgroups() if the daemon is not
4292 60. The Exim user and group can now be bound into the binary as text strings
4293 that are looked up at the start of Exim's processing.
4295 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4297 62. Added $mailstore_basename variable.
4299 63. Installed patch to sieve.c from Michael Haardt.
4301 64. When Exim failed to open the panic log after failing to open the main log,
4302 the original message it was trying to log was written to stderr and debug
4303 output, but if they were not available (the usual case in production), it
4304 was lost. Now it is written to syslog before the two lines that record the
4305 failures to open the logs.
4307 65. Users' Exim filters run in subprocesses under the user's uid. It is
4308 possible for a "deliver" command or an alias in a "personal" command to
4309 provoke an address rewrite. If logging of address rewriting is configured,
4310 this fails because the process is not running as root or exim. There may be
4311 a better way of dealing with this, but for the moment (because 4.30 needs
4312 to be released), I have disabled address rewrite logging when running a
4313 filter in a non-root, non-exim process.
4319 1. The buildconfig auxiliary program wasn't quoting the value set for
4320 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4321 not defined. This bug was masked in 4.22 by the effect that was fixed in
4324 2. Some messages that were rejected after a message id was allocated were
4325 shown as "incomplete" by exigrep. It no longer does this for messages that
4326 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4328 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4329 have allow_domain_literals set, the ID did not get logged in the <= line.
4330 Domain literals are now always recognized in Message-ID: header lines.
4332 4. The first argument for a ${extract expansion item is the key name or field
4333 number. Leading and trailing spaces in this item were not being ignored,
4334 causing some misleading effects.
4336 5. When deliver_drop_privilege was set, single queue runner processes started
4337 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4338 same command in the process it spins off) were not dropping privilege.
4340 6. When the daemon running as "exim" started a queue runner, it always
4341 re-executed Exim in the spun-off process. This is a waste of effort when
4342 deliver_drop_privilege is set. The new process now just calls the
4343 queue-runner function directly.
4349 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4352 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4353 ignored. Though the use of -oP was forcing the writing of a pid file, it
4354 was always written to the default place.
4356 3. If the message "no IP address found for host xxxx" is generated during
4357 incoming verification, it is now followed by identification of the incoming
4358 connection (so you can more easily find what provoked it).
4360 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4362 5. Added some features to "harden" Exim a bit more against certain attacks:
4364 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4365 be put in Local/Makefile. This is like the never_users runtime option,
4366 but it cannot be overridden. The default setting is "root".
4368 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4369 prefix string with which any file named in a -C command line option
4372 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4373 is retained for -C and -D only if the caller of Exim is root. Without
4374 it, the exim user may also use -C and -D and retain privilege.
4376 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4377 command line option is disabled.
4379 6. Macro names set by the -D option must start with an upper case letter, just
4380 like macro names defined in the configuration file.
4382 7. Added "dereference=" facility to LDAP.
4384 8. Two instances of the typo "uknown" in the source files are fixed.
4386 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4387 the Configure-Makefile script screwed up while processing it.
4389 10. Incorporated PCRE 4.4.
4391 11. The SMTP synchronization check was not operating right at the start of an
4392 SMTP session. For example, it could not catch a HELO sent before the client
4393 waited for the greeting. There is now a check for outstanding input at the
4394 point when the greeting is written. Because of the duplex, asynchronous
4395 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4396 way, but not yet received, when the check is performed.
4398 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4399 on TCP/IP sockets, because this apparently causes some broken clients to
4402 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4403 unchanged) from the Cygwin maintainer.
4405 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4406 was supported instead of testing for mbx. Effectively a typo.
4408 15. The spa authenticator server code was not checking that the input it
4409 received was valid base64.
4411 16. The debug output line for the "set" modifier in ACLs was not showing the
4412 name of the variable that was being set.
4414 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4416 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4417 lookup on its own. Something else was needed to trigger the lookup. For
4418 example, a match in host_lookup or the need for a host name in a host list.
4419 Now, if $sender_host_name is referenced and the host name has not yet been
4420 looked up, a lookup is performed. If the lookup fails, the variable remains
4421 empty, and $host_lookup_failed is set to "1".
4423 19. Added "eqi" as a case-independent comparison operator.
4425 20. The saslauthd authentication condition could segfault if neither service
4426 nor realm was specified.
4428 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4429 error message that was logged was misleading, and incoming SMTP
4430 connections were dropped. The message is now more accurate, and temporary
4431 errors are given to SMTP connections.
4433 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4434 (except RSET) with a 421 error, until QUIT is received. However, it was
4435 failing to send a response to QUIT.
4437 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4438 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4439 ACL is now run after the helo_try_verify_hosts code.
4441 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4442 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4443 case-independent, so other case variants are also recognized). Apparently
4444 some systems use these upper case variants.
4446 25. If more than two messages were waiting for the same host, and a transport
4447 filter was specified for the transport, Exim sent two messages over the
4448 same TCP/IP connection, and then failed with "socket operation on non-
4449 socket" when it tried to send the third.
4451 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4453 27. The extern definition of crypt16() in expand.c was not being excluded when
4454 the OS had its own crypt16() function.
4456 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4457 as a preferred synonym for return_size_limit, both as an option and as an
4460 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4462 30. Changed the default configuration ACL to relax the local part checking rule
4463 for addresses that are not in any local domains. For these addresses,
4464 slashes and pipe symbols are allowed within local parts, but the sequence
4465 /../ is explicitly forbidden.
4467 31. SPA server authentication was not clearing the challenge buffer before
4470 32. log_message in a "warn" ACL statement was writing to the reject log as
4471 well as to the main log, which contradicts the documentation and doesn't
4472 seem right (because no rejection is happening). So I have stopped it.
4474 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4475 However, I am unable to do any testing of this.
4477 34. Fixed an infelicity in the appendfile transport. When checking directories
4478 for a mailbox, to see if any needed to be created, it was accidentally
4479 using path names with one or more superfluous leading slashes; tracing
4480 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4482 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4483 contents are added to the log line that is written for every discarded
4484 recipient. (Previously a log_message setting was ignored.)
4486 36. The ${quote: operator now quotes the string if it is empty.
4488 37. The install script runs exim in order to find its version number. If for
4489 some reason other than non-existence or emptiness, which it checks, it
4490 could not run './exim', it was installing it with an empty version number,
4491 i.e. as "exim-". This error state is now caught, and the installation is
4494 38. An argument was missing from the function that creates an error message
4495 when Exim fails to connect to the socket for saslauthd authentication.
4496 This could cause Exim to crash, or give a corrupted message.
4498 39. Added isip, isip4, and isip6 to ${if conditions.
4500 40. The ACL variables $acl_xx are now saved with the message, and can be
4501 accessed later in routers, transports, and filters.
4503 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4504 strings in the file are not string-expanded.
4506 42. If a MAIL command specified a SIZE value that was too large to fit into an
4507 int variable, the check against message_size_limit failed. Such values are
4508 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4509 day this will have to be increased, but I don't think I want to be around
4510 when emails are that large.
4517 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4518 iconv() is not standard in FreeBSD.
4520 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4521 IPv6 enabled. The observed symptom was a segmentation fault on return from
4522 the function os_common_find_running_interfaces() in src/os.c.
4524 3. In the check_special_case() function in daemon.c I had used "errno" as an
4525 argument name, which causes warnings on some systems. This was basically a
4526 typo, since it was named "eno" in the comments!
4528 4. The code that waits for the clock to tick (at a resolution of some fraction
4529 of a second) so as to ensure message-id uniqueness was always waiting for
4530 at least one whole tick, when it could have waited for less. [This is
4531 almost certainly not relevant at current processor speeds, where it is
4532 unlikely to ever wait at all. But we try to future-proof.]
4534 5. The function that sleeps for a time interval that includes fractions of a
4535 second contained a race. It did not block SIGALRM between setting the
4536 timer, and suspending (a couple of lines later). If the interval was short
4537 and the sigsuspend() was delayed until after it had expired, the suspension
4538 never ended. On busy systems this could lead to processes getting stuck for
4541 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4542 process, before it forks any delivery processes. The open lookup caching
4543 mechanism meant that the open file or database connection was passed into
4544 the delivery process. The problem was that delivery processes always tidy
4545 up cached lookup data. This could cause a problem for the next delivery
4546 process started by the queue runner, because the external queue runner
4547 process does not know about the closure. So the next delivery process
4548 still has data in the lookup cache. In the case of a file lookup, there was
4549 no problem because closing a file descriptor in a subprocess doesn't affect
4550 the parent. However, if the lookup was caching a connection to a database,
4551 the connection was closed, and the second delivery process was likely to
4552 see errors such as "PGSQL: query failed: server closed the connection
4553 unexpectedly". The problem has been fixed by closing all cached lookups
4554 in a queue runner before running a delivery process.
4556 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4557 seem to have the "const" qualifier which it has on other OS. I've
4560 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4561 *of the same type* (client or server) with the same public name that an
4562 error should be diagnosed.
4564 9. When Exim looked up a host name for an IP address, but failed to find the
4565 original IP address when looking up the host name (a safety check), it
4566 output the message "<ip address> does not match any IP for NULL", which was
4567 confusing, to say the least. The bug was that the host name should have
4568 appeared instead of "NULL".
4570 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4571 user that is built into the binary, and the -C or -D options is used, root
4572 privilege is dropped before the configuration file is read. In addition,
4573 logging is switched to stderr instead of the normal log files. If the
4574 configuration then re-defines the Exim user, the unprivileged environment
4575 is probably not what is expected, so Exim logs a panic warning message (but
4578 However, if deliver_drop_privilege is set, the unprivileged state may well
4579 be exactly what is intended, so the warning has been cut out in that case,
4580 and Exim is allowed to try to write to its normal log files.
4586 1. smtp_return_error_details was not giving details for temporary sender
4587 or receiver verification errors.
4589 2. Diagnose a configuration error if two authenticators have the same public
4592 3. Exim used not to create the message log file for a message until the first
4593 delivery attempt. This could be confusing when incoming messages were held
4594 for policy or load reasons. The message log file is now created at the time
4595 the message is received, and an initial "Received" line is written to it.
4597 4. The automatically generated man page for command line options had a minor
4598 bug that caused no ill effects; however, a more serious problem was that
4599 the procedure for building the man page automatically didn't always
4600 operate. Consequently, release 4.20 contains an out-of-date version. This
4601 shouldn't happen again.
4603 5. When building Exim with embedded Perl support, the script that builds the
4604 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4605 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4607 6. The freeze_tell option was not being used for messages that were frozen on
4608 arrival, either by an ACL or by local_scan().
4610 7. Added the smtp_incomplete_transaction log selector.
4612 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4613 was accepting AUTH without a new EHLO.
4615 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4616 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4619 10. Logging of TCP/IP connections (when configured) now happens in the main
4620 daemon process instead of the child process, so that the TCP/IP connection
4621 count is more accurate (but it can never be perfect).
4623 11. The use of "drop" in a nested ACL was not being handled correctly in the
4624 outer ACL. Now, if condition failure induced by the nested "drop" causes
4625 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4626 or "require"), the connection is dropped.
4628 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4629 that yield "discard" can now be used with an "accept" or a "discard" verb,
4630 but an error is generated for any others (because I can't see a useful way
4631 to define what should happen).
4633 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4634 are now processed in the same way as lines in the Exim configuration file.
4635 In particular, continuation lines are supported.
4637 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4639 15. Added -ti meaning -t -i.
4641 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4642 domains, and warn by logging if others are found.
4644 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4645 returned from GIFCONF when Exim is trying to find the list of interfaces on
4646 a host. The code in os.c has been modified to copy each ifreq to an aligned
4647 structure in all cases.
4649 Also, in some cases, the returned ifreq's were being copied to a 'struct
4650 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4651 means the last couple of bytes of an IPv6 address could be chopped if the
4652 ifreq contained only a normal sockaddr (14 bytes storage).
4654 18. Named domain lists were not supported in the hosts_treat_as_local option.
4655 An entry such as +xxxx was not recognized, and was treated as a literal
4658 19. Ensure that header lines added by a DATA ACL are included in the reject log
4659 if the ACL subsequently rejects the message.
4661 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4662 MD5 (which is deprecated).
4664 21. When testing a filter file using -bf, Exim was writing a message when it
4665 took the sender from a "From " line in the message, but it was not doing so
4666 when it took $return_path from a Return-Path: header line. It now does.
4668 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4669 with a valid header line field name (a series of printing characters
4670 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4672 23. Changed "disc" in the source to "disk" to conform to the documentation and
4673 the book and for uniformity.
4675 24. Ignore Sendmail's -Ooption=value command line item.
4677 25. When execve() failed while trying to run a command in a pipe transport,
4678 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4679 could be confused with a return value of 69 from the command itself. This
4680 has been changed to 127, the value the shell returns if it is asked to run
4681 a non-existent command. The wording for the related log line suggests a
4682 non-existent command as the problem.
4684 26. If received_header_text expands to an empty string, do not add a Received:
4685 header line to the message. (Well, it adds a token one on the spool, but
4686 marks it "old" so that it doesn't get used or transmitted.)
4688 27. Installed eximstats 1.28 (addition of -nt option).
4690 28. There was no check for failure on the call to getsockname() in the daemon
4691 code. This can fail if there is a shortage of resources on the system, with
4692 ENOMEM, for example. A temporary error is now given on failure.
4694 29. Contrary to the C standard, it seems that in some environments, the
4695 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4696 program. Exim now does this explicitly; it affects the formatting of
4697 timestamps using strftime().
4699 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4700 complaints. I've now initialized all the variables, to avoid this.
4702 32. Header lines added by a system filter were not being "seen" during
4703 transport-time rewrites.
4705 33. The info_callback() function passed to OpenSSL is set up with type void
4706 (*)(SSL *, int, int), as described somewhere. However, when calling the
4707 function (actually a macro) that sets it up, the type void(*)() is
4708 expected. I've put in a cast to prevent warnings from picky compilers.
4710 34. If a DNS black list lookup found a CNAME record, but there were no A
4711 records associated with the domain it pointed at, Exim crashed.
4713 35. If a DNS black list lookup returned more than one A record, Exim ignored
4714 all but the first. It now scans all returned addresses if a particular IP
4715 value is being sought. In this situation, the contents of the
4716 $dnslist_value variable are a list of all the addresses, separated by a
4719 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4720 to accept a host name and all its aliases if the forward lookup for any of
4721 them yielded the IP address of the incoming connection. Now it accepts only
4722 those names whose forward lookup yields the correct IP address. Any other
4723 names are discarded. This closes a loophole whereby a rogue DNS
4724 administrator could create reverse DNS records to break through a
4725 wildcarded host restriction in an ACL.
4727 37. If a user filter or a system filter that ran in a subprocess used any of
4728 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4729 the wrong values were passed to the pipe command ($thisaddress had the
4730 value of $0, $0 had the value of $1, etc). This bug was introduced by
4731 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4733 38. Improved the line breaking for long SMTP error messages from ACLs.
4734 Previously, if there was no break point between 40 and 75 characters, Exim
4735 left the rest of the message alone. Two changes have been made: (a) I've
4736 reduced the minimum length to 35 characters; (b) if it can't find a break
4737 point between 35 and 75 characters, it looks ahead and uses the first one
4738 that it finds. This may give the occasional overlong line, but at least the
4739 remaining text gets split now.
4741 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4742 file descriptors might be low, and that setting 1000 would always raise it.
4743 It turns out that in some environments, the limit is already over 1000 and
4744 that lowering it causes trouble. So now Exim takes care not to decrease it.
4746 40. When delivering a message, the value of $return_path is set to $sender_
4747 address at the start of routing (routers may change the value). By an
4748 oversight, this default was not being set up when an address was tested by
4749 -bt or -bv, which affected the outcome if any router or filter referred to
4752 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4753 log only when "message" or "log_message" are set. However, if one of the
4754 conditions was an address verification, or a call to a nested ACL, the
4755 messages generated by the underlying test were being passed through. This
4756 no longer happens. The underlying message is available in $acl_verify_
4757 message for both "message" and "log_message" expansions, so it can be
4758 passed through if needed.
4760 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4761 new expansion $bh_ to give the encoded byte string without charset
4762 translation. Translation happens only if iconv() is available; HAVE_ICONV
4763 indicates this at build time. HEADERS_CHARSET gives the charset to
4764 translate to; headers_charset can change it in the configuration, and
4765 "headers charset" can change it in an individual filter file.
4767 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4768 that creates RFC 2047 encoded "words" labels them as that charset instead
4769 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4770 expansion operator; (ii) when Exim creates a From: line for a local
4771 message; (iii) when a header line is rewritten to include a "phrase" part.
4773 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4774 buggy, causing it to skip the first lines of messages whose message ID
4775 ended in 'D'. This would not have bitten before Exim release 4.14, because
4776 message IDs were unlikely to end in 'D' before then. The effect was to have
4777 incorrect size information for certain domains.
4779 45. #include "config.h" was missing at the start of the crypt16.c module. This
4780 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4783 46. If there was a timeout during a "random" callout check, Exim treated it as
4784 a failure of the random address, and carried on sending RSET and the real
4785 address. If the delay was just some slowness somewhere, the response to the
4786 original RCPT would be taken as a response to RSET and so on, causing
4787 mayhem of various kinds.
4789 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4790 when I implemented it. It didn't allow for the fact that some option values
4791 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4792 the right test for positive values.
4794 48. Domain names in DNS records are case-independent. Exim always looks them up
4795 in lower case. Some resolvers return domain names in exactly the case they
4796 appear in the zone file, that is, they may contain uppercase letters. Not
4797 all resolvers do this - some return always lower case. Exim was treating a
4798 change of case by a resolver as a change of domain, similar to a widening
4799 of a domain abbreviation. This triggered its re-routing code and so it was
4800 trying to route what was effectively the same domain again. This normally
4801 caused routing to fail (because the router wouldn't handle the domain
4802 twice). Now Exim checks for this case specially, and just changes the
4803 casing of the domain that it ultimately uses when it transmits the message
4806 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4809 50. If a filter generated a file delivery with a non-absolute name (possible if
4810 no home directory exists for the router), the forbid_file option was not
4813 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4814 the existing equality matching.
4816 52. Exim was using ints instead of ino_t variables in some places where it was
4817 dealing with inode numbers.
4819 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4820 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4821 called TMPDIR, and if it finds it is different, it changes its value.
4823 54. The smtp_printf() function is now made available to local_scan() so
4824 additional output lines can be written before returning. There is also an
4825 smtp_fflush() function to enable the detection of a dropped connection.
4826 The variables smtp_input and smtp_batched_input are exported to
4829 55. Changed the default runtime configuration: the message "Unknown user"
4830 has been removed from the ACL, and instead placed on the localuser router,
4831 using the cannot_route_message feature. This means that any verification
4832 failures that generate their own messages won't get overridden. Similarly,
4833 the "Unrouteable address" message that was in the ACL for unverifiable
4834 relay addresses has also been removed.
4836 56. Added hosts_avoid_esmtp to the smtp transport.
4838 57. The exicyclog script was not checking for the esoteric option
4839 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4840 will work only if exicyclog is run under the appropriate euid.
4842 58. Following a discussion on the list, the rules by which Exim recognises line
4843 endings on incoming messages have been changed. The -dropcr and drop_cr
4844 options are now no-ops, retained only for backwards compatibility. The
4845 following line terminators are recognized: LF CRLF CR. However, special
4846 processing applies to CR:
4848 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4849 nor a local message in the state where . is a terminator.
4851 (ii) If a bare CR is encountered in a header line, an extra space is added
4852 after the line terminator so as not to end the header. The reasoning
4853 behind this is that bare CRs in header lines are most likely either
4854 to be mistakes, or people trying to play silly games.
4856 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4857 was being incorrectly given as 18 bytes larger than it should have been.
4858 This is a VOB (very old bug).
4860 60. This may never have affected anything current, but just in case it has:
4861 When the local host is found other than at the start of a list of hosts,
4862 the local host, those with the same MX, and any that follow, are discarded.
4863 When the list in question was part of a longer list of hosts, the following
4864 hosts (not currently being processed) were also being discarded. This no
4865 longer happens. I'm not sure if this situation could ever has previously
4868 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4871 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4872 header. This is something that is recommended in a new Internet Draft, and
4873 is something that is documented as being done by Sendmail. There are two
4874 possible values. For messages generated by the autoreply transport, Exim
4877 Auto-Submitted: auto-replied
4879 whereas for all other generated messages (e.g. bounces) it adds
4881 Auto-Submitted: auto-generated
4883 63. The "personal" condition in filters now includes a test for the
4884 Auto-Submitted: header. If it contains the string "auto-" the message it
4885 not considered personal.
4887 64. Added rcpt_include_affixes as a generic transport option.
4889 65. Added queue_only_override (default true).
4891 66. Added the syslog_duplication option.
4893 67. If what should have been the first header line of a message consisted of
4894 a space followed by a colon, Exim was mis-interpreting it as a header line.
4895 It isn't of course - it is syntactically invalid and should therefore be
4896 treated as the start of the message body. The misbehaviour could have
4897 caused a number of strange effects, including loss of data in subsequent
4898 header lines, and spool format errors.
4900 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4901 client host had authenticated. This control can now be exercised by an ACL
4902 for more flexibility.
4904 69. By default, callouts do not happen when testing with -bh. There is now a
4905 variant, -bhc, which does actually run the callout code, including
4906 consulting and updating the callout cache.
4908 70. Added support for saslauthd authentication, courtesy of Alexander
4911 71. If statvfs() failed on the spool or log directories while checking their
4912 size for availability, Exim confusingly gave the error "space shortage".
4913 Furthermore, in debugging mode it crashed with a floating point exception.
4914 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4915 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4916 serious problem, Exim now writes to the main and panic logs when this
4917 happens, with details of the failure. It then refuses to accept the
4918 incoming message, giving the message "spool directory problem" or "log
4919 directory problem" with a 421 code for SMTP messages.
4921 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4922 0, 1, and 2 exist, because some OS complain for execs without them (see
4923 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4924 descriptors. However, the code omitted to check that the open succeeded,
4925 causing mysterious errors if for some reason the permissions on /dev/null
4926 got screwed. Now Exim writes a message to the main and panic logs, and
4927 bombs out if it can't open /dev/null.
4929 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4930 interact so that it is all more flexible. It is supposed to remain
4931 backwards compatible. Also added extra_local_interfaces.
4933 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4934 to bomb out with an assertion failure - to the client this appears as a
4935 connection drop. This problem occurs in the part of the code that was taken
4936 from the Samba project. Fortunately, the assertion is in a very simple
4937 function, so I have fixed this by reproducing the function inline in the
4938 one place where it is called, and arranging for authentication to fail
4939 instead of killing the process with assert().
4941 75. The SPA client code was not working when the server requested OEM rather
4942 than Unicode encoding.
4944 76. Added code to make require_files with a specific uid setting more usable in
4945 the case where statting the file as root fails - usually a non-root-mounted
4946 NFS file system. When this happens and the failure is EACCES, Exim now
4947 forks a subprocess and does the per-uid checking as the relevant uid.
4949 77. Added process_log_path.
4951 78. If log_file_path was not explicitly set, a setting of check_log_space or
4952 check_log_inodes was ignored.
4954 79. If a space check for the spool or log partitions fails, the incident is now
4955 logged. Of course, in the latter case the data may get lost...
4957 80. Added the %p formatting code to string_format() so that it can be used to
4958 print addresses in debug_print(). Adjusted all the address printing in the
4959 debugging in store.c to use %p rather than %d.
4961 81. There was a concern that a line of code in smtp_in.c could overflow a
4962 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4963 initially expressed, the concern was not well-founded, because trailing
4964 spaces are removed early. However, if the trailing spaces were followed by
4965 a NULL, they did not get removed, so the overflow was possible. Two fixes
4968 (a) I re-wrote the offending code in a cleaner fashion.
4969 (b) If an incoming SMTP command contains a NULL character, it is rejected
4972 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4973 runs initgroups(), so that if the Exim user is in any additional groups,
4974 they will be used during message reception.
4980 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob