1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.577 2009/10/19 12:57:33 nm4 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added patch by Johannes Berg that expands the main option
10 "spamd_servers" if it starts with a dollar sign.
12 TK/02 Write list of recipients to X-Envelope-Sender header when building
13 the mbox-format spool file for content scanning (suggested by Jakob
16 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
17 (http://www.dcc-servers.net/) support via dccifd. Activated by
18 setting EXPERIMENTAL_DCC=yes in Local/Makefile. Check out
19 experimental_spec.txt for more documentation.
21 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
22 by Mark Daniel Reidel <mr@df.eu>.
24 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
25 When building exim an external PCRE library is now needed -
26 PCRE is a system library on the majority of modern systems.
27 See entry on PCRE_LIBS in EDITME file.
29 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
30 conversation. Added nologin parameter to request.
31 Patch contributed by Kirill Miazine
33 TF/01 Do not log submission mode rewrites if they do not change the address.
35 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
37 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
38 log files in place. Contributed by Roberto Lima
40 NM/04 Bugzilla 667: close socket used by dovecot authenticator
42 TF/03 Bugzilla 615: When checking the local_parts router precondition
43 after a local_part_suffix or local_part_prefix option, Exim now
44 does not use the address's named list lookup cache, since this
45 contains cached lookups for the whole local part.
47 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
48 Robert Millan. Documentation is in experimental-spec.txt
50 TF/04 Bugzilla 668: Fix parallel build (make -j).
52 NM/05 Bugzilla 437: Prevent Maildix aux files being created with mode 000
54 NM/05 Bugzilla 598: Improvement to Dovecot authenticator handling.
55 Patch provided by Jan Srzednicki
57 TF/05 Leading white space used to be stripped from $spam_report which
58 wrecked the formatting. Now it is preserved.
60 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
61 that they are available at delivery time.
63 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
65 TF/08 TLS error reporting now respects the incoming_interface and
66 incoming_port log selectors.
68 TF/09 Produce a more useful error message if an SMTP transport's hosts
69 setting expands to an empty string.
71 NM/06 Bugzilla 744: EXPN did not work under TLS.
72 Patch provided by Phil Pennock
74 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
75 Patch provided by Richard Godbee
77 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
78 acl_smtp_notquit, added index entry.
80 NM/09 Bugzilla 787: Potential buffer overflow in string_format
81 Patch provided by Eugene Bujak
83 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to accept()
84 Patch provided by Maxim Dounin
86 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
87 Patch provided by Phil Pennock
89 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
91 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
92 Patch provided by Brad "anomie" Jorsch
94 NM/14 Bugzilla 622: Added timeout setting to transport filter.
95 Patch provided by Dean Brooks
97 TK/05 Add native DKIM support (does not depend on external libraries).
99 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
100 Patch provided by Graeme Fowler
102 NM/16 Bugzilla 851: Documentation example syntax fix.
104 NM/17 Changed NOTICE file to remove references to embedded PCRE.
106 NM/18 Bugzilla 894: Fix issue with very long lines including comments in lsearch
108 NM/19 Bugzilla 745: TLS version reporting
109 Patch provided by Phil Pennock
111 NM/20 Bugzilla 167: bool: condition support
112 Patch provided by Phil Pennock
114 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken clients
115 Patch provided by Phil Pennock
117 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date
118 Patch provided by Brad "anomie" Jorsch
120 NM/23 Bugzilla 687: Fix misparses in eximstats
121 Patch provided by Heiko Schlittermann
123 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid
124 Patch provided by Heiko Schlittermann
126 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file
128 NM/26 Bugzilla 799: Documentation correction for ratelimit
134 TK/01 Add preliminary DKIM support. Currently requires a forked version of
135 ALT-N's libdkim that I have put here:
136 http://duncanthrax.net/exim-experimental/
138 Note to Michael Haardt: I had to rename some vars in sieve.c. They
139 were called 'true' and it seems that C99 defines that as a reserved
140 keyword to be used with 'bool' variable types. That means you could
141 not include C99-style headers which use bools without triggering
142 build errors in sieve.c.
144 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
145 as mailq or other aliases. Changed the --help handling significantly
146 to do whats expected. exim_usage() emits usage/help information.
148 SC/01 Added the -bylocaldomain option to eximstats.
150 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
152 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
154 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
160 PH/01 Another patch from the Sieve maintainer.
162 PH/02 When an IPv6 address is converted to a string for single-key lookup
163 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
164 dots are used instead of colons so that keys in lsearch files need not
165 contain colons. This was done some time before quoting was made available
166 in lsearch files. However, iplsearch files do require colons in IPv6 keys
167 (notated using the quote facility) so as to distinguish them from IPv4
168 keys. This meant that lookups for IP addresses in host lists did not work
169 for iplsearch lookups.
171 This has been fixed by arranging for IPv6 addresses to be expressed with
172 colons if the lookup type is iplsearch. This is not incompatible, because
173 previously such lookups could never work.
175 The situation is now rather anomolous, since one *can* have colons in
176 ordinary lsearch keys. However, making the change in all cases is
177 incompatible and would probably break a number of configurations.
179 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
182 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
183 conversion specification without a maximum field width, thereby enabling
184 a rogue spamd server to cause a buffer overflow. While nobody in their
185 right mind would setup Exim to query an untrusted spamd server, an
186 attacker that gains access to a server running spamd could potentially
187 exploit this vulnerability to run arbitrary code as the Exim user.
189 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
190 $primary_hostname instead of what libspf2 thinks the hosts name is.
192 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
193 a directory entry by the name of the lookup key. Previously, if a
194 symlink pointed to a non-existing file or a file in a directory that
195 Exim lacked permissions to read, a lookup for a key matching that
196 symlink would fail. Now it is enough that a matching directory entry
197 exists, symlink or not. (Bugzilla 503.)
199 PH/03 The body_linecount and body_zerocount variables are now exported in the
202 PH/04 Added the $dnslist_matched variable.
204 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
205 This means they are set thereafter only if the connection becomes
208 PH/06 Added the client_condition to authenticators so that some can be skipped
209 by clients under certain conditions.
211 PH/07 The error message for a badly-placed control=no_multiline_responses left
212 "_responses" off the end of the name.
214 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
216 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
217 (without spaces) instead of just copying the configuration text.
219 PH/10 Added the /noupdate option to the ratelimit ACL condition.
221 PH/11 Added $max_received_linelength.
223 PH/12 Added +ignore_defer and +include_defer to host lists.
225 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
226 way in which PCRE > 7.0 is built.
228 PH/14 Implemented queue_only_load_latch.
230 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
231 MAIL command. The effect was to mangle the value on 64-bit systems.
233 PH/16 Another patch from the Sieve maintainer.
235 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
237 PH/18 If a system quota error occurred while trying to create the file for
238 a maildir delivery, the message "Mailbox is full" was not appended to the
239 bounce if the delivery eventually timed out. Change 4.67/27 below applied
240 only to a quota excession during the actual writing of the file.
242 PH/19 It seems that peer DN values may contain newlines (and other non-printing
243 characters?) which causes problems in log lines. The DN values are now
244 passed through string_printing() before being added to log lines.
246 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
247 and InterBase are left for another time.)
249 PH/21 Added message_body_newlines option.
251 PH/22 Guard against possible overflow in moan_check_errorcopy().
253 PH/23 POSIX allows open() to be a macro; guard against that.
255 PH/24 If the recipient of an error message contained an @ in the local part
256 (suitably quoted, of course), incorrect values were put in $domain and
257 $local_part during the evaluation of errors_copy.
263 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
264 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
267 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
268 issue a MAIL command.
270 PH/02 In an ACL statement such as
272 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
274 if a client was not listed at all, or was listed with a value other than
275 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
276 the condition was not true (as it should be), so access was not denied.
277 The bug was that the ! inversion was incorrectly passed on to the second
278 item. This has been fixed.
280 PH/03 Added additional dnslists conditions == and =& which are different from
281 = and & when the dns lookup returns more than one IP address.
283 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
284 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
286 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
287 FSYNC, which compiles an option called disable_fsync that allows for
288 bypassing fsync(). The documentation is heavily laced with warnings.
290 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
292 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
293 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
294 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
295 including adding "make clean"; (3) Added -fPIC when compiling the test
296 dynamically loaded module, to get rid of a warning.
298 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
299 message fails, move_frozen_messages = true and ignore_bounce_errors_after
300 = 0s. The bug is otherwise harmless.
302 PH/07 There was a bug in the dovecot authenticator such that the value of
303 $auth1 could be overwritten, and so not correctly preserved, after a
304 successful authentication. This usually meant that the value preserved by
305 the server_setid option was incorrect.
307 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
309 PH/09 Installed PCRE release 7.0.
311 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
312 run for batched SMTP input. It is now run at the start of every message
313 in the batch. While fixing this I discovered that the process information
314 (output by running exiwhat) was not always getting set for -bs and -bS
315 input. This is fixed, and it now also says "batched" for BSMTP.
317 PH/11 Added control=no_pipelining.
319 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
320 patch, slightly modified), and move the expansion of helo_data till after
321 the connection is made in the smtp transport (so it can use these
324 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
326 PH/14 Added log_selector = +pid.
328 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
330 PH/16 Add ${if forany and ${if forall.
332 PH/17 Added dsn_from option to vary the From: line in DSNs.
334 PH/18 Flush SMTP output before performing a callout, unless control =
335 no_callout_flush is set.
337 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
338 was true (the default) a successful delivery failed to delete the retry
339 item, thus causing premature timeout of the address. The bug is now
342 PH/20 Added hosts_avoid_pipelining to the smtp transport.
344 PH/21 Long custom messages for fakedefer and fakereject are now split up
345 into multiline reponses in the same way that messages for "deny" and
346 other ACL rejections are.
348 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
349 with slight modification.
351 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
352 draft, changing the syntax and factoring some duplicate code".
354 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
355 for deliveries of the second and subsequent messages over the same SMTP
358 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
359 ${reduce, with only minor "tidies".
361 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
363 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
364 expansion side effects.
366 PH/27 When a message times out after an over-quota error from an Exim-imposed
367 quota, the bounce message says "mailbox is full". This message was not
368 being given when it was a system quota that was exceeded. It now should
371 MH/03 Made $recipients available in local_scan(). local_scan() already has
372 better access to the recipient list through recipients_list[], but
373 $recipients can be useful in postmaster-provided expansion strings.
375 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
376 in the case of a MAIL command with additional options following the
377 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
378 were accidentally chopped off.
380 PH/29 SMTP synchronization checks are implemented when a command is read -
381 there is a check that no more input is waiting when there shouldn't be
382 any. However, for some commands, a delay in an ACL can mean that it is
383 some time before the response is written. In this time, more input might
384 arrive, invalidly. So now there are extra checks after an ACL has run for
385 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
386 pipelining has not been advertised.
388 PH/30 MH's patch to allow iscntrl() characters to be list separators.
390 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
391 returned in the SMTP response when smtp_return_error_details was false.
394 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
395 instead of the C I/O that was originally supplied, because problems were
398 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
399 Exim which did not show up earlier: it was assuming that a call to
400 SSL_CTX_set_info_callback() might give an error value. In fact, there is
401 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
402 was a macro that became an assignment, so it seemed to work. This has
403 changed to a proper function call with a void return, hence the compile
404 error. Exim's code has been fixed.
406 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
409 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
411 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
413 PH/37 If a message is not accepted after it has had an id assigned (e.g.
414 because it turns out to be too big or there is a timeout) there is no
415 "Completed" line in the log. When some messages of this type were
416 selected by exigrep, they were listed as "not completed". Others were
417 picked up by some special patterns. I have improved the selection
418 criteria to be more general.
420 PH/38 The host_find_failed option in the manualroute router can now be set
421 to "ignore", to completely ignore a host whose IP address cannot be
422 found. If all hosts are ignored, the behaviour is controlled by the new
423 host_all_ignored option.
425 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
426 homing or because of multiple MX records with /mx) generated more than
427 one IP address, and the following item turned out to be the local host,
428 all the secondary addresses of the first item were incorrectly removed
429 from the list, along with the local host and any following hosts (which
430 is what is supposed to happen).
432 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
433 whoever called Exim into the -H file. In the case of the daemon it was
434 behaving confusingly. When first started, it used values for whoever
435 started the daemon, but after a SIGHUP it used the Exim user (because it
436 calls itself on a restart). I have changed the code so that it now always
439 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
440 message are rejected with the same error (e.g. no authentication or bad
441 sender address), and a DATA command is nevertheless sent (as can happen
442 with PIPELINING or a stupid MUA), the error message that was given to the
443 RCPT commands is included in the rejection of the DATA command. This is
444 intended to be helpful for MUAs that show only the final error to their
447 PH/42 Another patch from the Sieve maintainer.
449 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
450 Eximstats - Fixed some broken HTML links and added missing column headers
452 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
453 columns for Rejects, Temp Rejects, Ham, and Spam rows.
455 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
457 PH/43 Yet another patch from the Sieve maintainer.
459 PH/44 I found a way to check for a TCP/IP connection going away before sending
460 the response to the final '.' that terminates a message, but only in the
461 case where the client has not sent further data following the '.'
462 (unfortunately, this is allowed). However, in many cases there won't be
463 any further data because there won't be any more messages to send. A call
464 to select() can be used: if it shows that the input is "ready", there is
465 either input waiting, or the socket has been closed. An attempt to read
466 the next input character can distinguish the two cases. Previously, Exim
467 would have sent an OK response which the client would never have see.
468 This could lead to message repetition. This fix should cure that, at
469 least in a lot of common cases.
471 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
472 advertised in response to EHLO.
478 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
479 fixed by 4.65/MH/01 (is this a record?) are fixed:
481 (i) An empty string was always treated as zero by the numeric comparison
482 operators. This behaviour has been restored.
484 (ii) It is documented that the numeric comparison operators always treat
485 their arguments as decimal numbers. This was broken in that numbers
486 starting with 0 were being interpreted as octal.
488 While fixing these problems I realized that there was another issue that
489 hadn't been noticed. Values of message_size_limit (both the global option
490 and the transport option) were treated as octal if they started with 0.
491 The documentation was vague. These values are now always treated as
492 decimal, and I will make that clear in the documentation.
498 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
499 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
502 MH/01 Don't check that the operands of numeric comparison operators are
503 integers when their expansion is in "skipping" mode (fixes bug
504 introduced by 4.64-PH/07).
506 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
507 child addresses, Exim now panics and dies. Previously, because the count
508 is held in a short int, deliveries were likely to be lost. As such a
509 large number of recipients for a single message is ridiculous
510 (performance will be very, very poor), I have chosen to impose a limit
511 rather than extend the field.
517 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
518 leftover -K file (the existence of which was triggered by #402).
519 While we were at it, introduced process PID as part of the -K
520 filename. This should rule out race conditions when creating
523 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
524 processing considerably. Previous code took too long for large mails,
525 triggering a timeout which in turn triggers #401.
527 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
528 in the DK code in transports.c. sendfile() is not really portable,
529 hence the _LINUX specificness.
531 TF/01 In the add_headers option to the mail command in an Exim filter,
532 there was a bug that Exim would claim a syntax error in any
533 header after the first one which had an odd number of characters
536 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
537 callout verification, Exim cached a "reject" for the entire domain. This
538 is correct for most verifications, but it is not correct for a recipient
539 verification with use_sender or use_postmaster set, because in that case
540 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
541 case of MAIL FROM:<> rejection from other early rejections (e.g.
542 rejection of HELO). When verifying a recipient using a non-null MAIL
543 address, the cache is ignored if it shows MAIL FROM:<> rejection.
544 Whatever the result of the callout, the value of the domain cache is
545 left unchanged (for any other kind of callout, getting as far as trying
546 RCPT means that the domain itself is ok).
548 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
551 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
552 manifest itself as EPIPE rather than ECONNECT. When tidying away a
553 session, the daemon ignores ECONNECT errors and logs others; it now
554 ignores EPIPE as well.
556 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
557 (quoted-printable decoding).
559 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
560 later the small subsequent patch to fix an introduced bug.
562 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
564 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
566 PH/08 An error is now given if message_size_limit is specified negative.
568 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
569 to be given (somewhat) arbitrary names.
571 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
574 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
575 miscellaneous code fixes
577 PH/10 Added the log_reject_target ACL modifier to specify where to log
580 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
581 hostname. This is wrong, because it relates to the incoming message (and
582 probably the interface on which it is arriving) and not to the outgoing
583 callout (which could be using a different interface). This has been
584 changed to use the value of the helo_data option from the smtp transport
585 instead - this is what is used when a message is actually being sent. If
586 there is no remote transport (possible with a router that sets up host
587 addresses), $smtp_active_hostname is used.
589 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
590 tweaks were necessary in order to get it to work (see also 21 below):
591 (a) The code assumed that strncpy() returns a negative number on buffer
592 overflow, which isn't the case. Replaced with Exim's string_format()
594 (b) There were several signed/unsigned issues. I just did the minimum
595 hacking in of casts. There is scope for a larger refactoring.
596 (c) The code used strcasecmp() which is not a standard C function.
597 Replaced with Exim's strcmpic() function.
598 (d) The code set only $1; it now sets $auth1 as well.
599 (e) A simple test gave the error "authentication client didn't specify
600 service in request". It would seem that Dovecot has changed its
601 interface. Fortunately there's a specification; I followed it and
602 changed what the client sends and it appears to be working now.
604 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
607 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
608 address A is aliased to B and C, where B exists and C does not. Without
609 -v the output is "A verified" because verification stops after a
610 successful redirection if more than one address is generated. However,
611 with -v the child addresses are also verified. Exim was outputting "A
612 failed to verify" and then showing the successful verification for C,
613 with its parentage. It now outputs "B failed to verify", showing B's
614 parentage before showing the successful verification of C.
616 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
617 look up a TXT record in a specific list after matching in a combined
620 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
621 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
622 they consult the DNS. I had assumed they would set it the way they
623 wanted; and indeed my experiments on Linux seem to show that in some
624 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
625 To be on the safe side, however, I have now made the interface to
626 host_find_byname() similar to host_find_bydns(), with an argument
627 containing the DNS resolver options. The host_find_byname() function now
628 sets these options at its start, just as host_find_bydns() does. The smtp
629 transport options dns_qualify_single and dns_search_parents are passed to
630 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
631 of host_find_byname() use the default settings of RES_DEFNAMES
632 (qualify_single) but not RES_DNSRCH (search_parents).
634 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
635 spool_read_header() do less string testing, by means of a preliminary
636 switch on the second character of optional "-foo" lines. (This is
637 overdue, caused by the large number of possibilities that now exist.
638 Originally there were few.) While I was there, I also converted the
639 str(n)cmp tests so they don't re-test the leading "-" and the first
640 character, in the hope this might squeeze out yet more improvement.
642 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
643 flag allowing group syntax was set by the header_syntax check but not
644 turned off, possible causing trouble later; (2) The flag was not being
645 set at all for the header_verify test, causing "group"-style headers to
646 be rejected. I have now set it in this case, and also caused header_
647 verify to ignore an empty address taken from a group. While doing this, I
648 came across some other cases where the code for allowing group syntax
649 while scanning a header line wasn't quite right (mostly, not resetting
650 the flag correctly in the right place). These bugs could have caused
651 trouble for malformed header lines. I hope it is now all correct.
653 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
654 with the "reply" argument non-NULL. The code, however (which originally
655 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
656 but it didn't always do it. This confused somebody who was copying the
657 code for some other use. I have removed all the tests.
659 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
660 feature that was used to support insecure browsers during the U.S. crypto
661 embargo. It requires special client support, and Exim is probably the
662 only MTA that supported it -- and would never use it because real RSA is
663 always available. This code has been removed, because it had the bad
664 effect of slowing Exim down by computing (never used) parameters for the
665 RSA_EXPORT functionality.
667 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
668 authenticator to fail if there's a tab character in the incoming data
669 (there should never be unless someone is messing about, as it's supposed
670 to be base64-encoded). Also added, on Timo's advice, the "secured" option
671 if the connection is using TLS or if the remote IP is the same as the
672 local IP, and the "valid-client-cert option" if a client certificate has
675 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
676 authenticators. This can be used for authorization after authentication
677 succeeds. (In the case of plaintext, it servers for both authentication
680 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
681 if any retry times were supplied.
683 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
684 connection, where there is no client IP address to check. In this
685 situation, the verify now always succeeds.
687 PH/25 Applied John Jetmore's -Mset patch.
689 PH/26 Added -bem to be like -Mset, but loading a message from a file.
691 PH/27 In a string expansion for a processed (not raw) header when multiple
692 headers of the same name were present, leading whitespace was being
693 removed from all of them, but trailing whitespace was being removed only
694 from the last one. Now trailing whitespace is removed from each header
695 before concatenation. Completely empty headers in a concatenation (as
698 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
699 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
701 PH/29 [Removed. This was a change that I later backed out, and forgot to
702 correct the ChangeLog entry (that I had efficiently created) before
703 committing the later change.]
705 PH/30 Exim was sometimes attempting to deliver messages that had suffered
706 address errors (4xx response to RCPT) over the same connection as other
707 messages routed to the same hosts. Such deliveries are always "forced",
708 so retry times are not inspected. This resulted in far too many retries
709 for the affected addresses. The effect occurred only when there were more
710 hosts than the hosts_max_try setting in the smtp transport when it had
711 the 4xx errors. Those hosts that it had tried were not added to the list
712 of hosts for which the message was waiting, so if all were tried, there
713 was no problem. Two fixes have been applied:
715 (i) If there are any address or message errors in an SMTP delivery, none
716 of the hosts (tried or untried) are now added to the list of hosts
717 for which the message is waiting, so the message should not be a
718 candidate for sending over the same connection that was used for a
719 successful delivery of some other message. This seems entirely
720 reasonable: after all the message is NOT "waiting for some host".
721 This is so "obvious" that I'm not sure why it wasn't done
722 previously. Hope I haven't missed anything, but it can't do any
723 harm, as the worst effect is to miss an optimization.
725 (ii) If, despite (i), such a delivery is accidentally attempted, the
726 routing retry time is respected, so at least it doesn't keep
727 hammering the server.
729 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
730 in ${readsocket because some servers need this prod.
732 PH/32 Added some extra debug output when updating a wait-xxx database.
734 PH/33 The hint "could be header name not terminated by colon", which has been
735 given for certain expansion errors for a long time, was not being given
736 for the ${if def:h_colon_omitted{... case.
738 PH/34 The spec says: "With one important exception, whenever a domain list is
739 being scanned, $domain contains the subject domain." There was at least
740 one case where this was not true.
742 PH/35 The error "getsockname() failed: connection reset by peer" was being
743 written to the panic log as well as the main log, but it isn't really
744 panic-worthy as it just means the connection died rather early on. I have
745 removed the panic log writing for the ECONNRESET error when getsockname()
748 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
749 runs only) independently of the message's sender address. This meant
750 that, if the 4xx error was in fact related to the sender, a different
751 message to the same recipient with a different sender could confuse
752 things. In particualar, this can happen when sending to a greylisting
753 server, but other circumstances could also provoke similar problems.
754 I have changed the default so that the retry time for these errors is now
755 based a combination of the sender and recipient addresses. This change
756 can be overridden by setting address_retry_include_sender=false in the
759 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
760 remote server are returned as part of bounce messages. This was not
761 happening for LMTP over a pipe (the lmtp transport), but now it is the
762 same for both kinds of LMTP.
764 PH/38 Despite being documented as not happening, Exim was rewriting addresses
765 in header lines that were in fact CNAMEs. This is no longer the case.
767 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
768 and queue runs started by the daemon processed all messages. This has
769 been fixed so that -R and -S can now usefully be given with -q<time>.
771 PH/40 Import PCRE release 6.7 (fixes some bugs).
773 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
775 PH/42 Give an error if -q is specified more than once.
777 PH/43 Renamed the variables $interface_address and $interface_port as
778 $received_ip_address and $received_port, to make it clear that these
779 values apply to message reception, and not to the outgoing interface when
780 a message is delivered. (The old names remain recognized, of course.)
782 PH/44 There was no timeout on the connect() call when using a Unix domain
783 socket in the ${readsocket expansion. There now is.
785 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
786 be meaningful with "accept".
788 SC/01 Eximstats V1.43
789 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
791 SC/02 Eximstats V1.44
792 Use a glob alias rather than an array ref in the generated
793 parser. This improves both readability and performance.
795 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
796 Collect SpamAssassin and rejection statistics.
797 Don't display local sender or destination tables unless
798 there is data to show.
799 Added average volumes into the top table text output.
801 SC/04 Eximstats V1.46
802 Collect data on the number of addresses (recipients)
803 as well as the number of messages.
805 SC/05 Eximstats V1.47
806 Added 'Message too big' to the list of mail rejection
807 reasons (thanks to Marco Gaiarin).
809 SC/06 Eximstats V1.48
810 Mainlog lines which have GMT offsets and are too short to
811 have a flag are now skipped.
813 SC/07 Eximstats V1.49 (Alain Williams)
814 Added the -emptyok flag.
816 SC/08 Eximstats V1.50
817 Fixes for obtaining the IP address from reject messages.
819 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
820 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
821 whitesspace changes from 4.64-PH/27
823 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
826 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
827 are found, allow negative numbers in numeric criteria)
829 JJ/06 exipick.20061117.2, added new $message_body_missing variable
831 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
832 to match changes made in 4.64-PH/43
834 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
836 PH/47 Put in an explicit test for a DNS lookup of an address record where the
837 "domain" is actually an IP address, and force a failure. This locks out
838 those revolvers/nameservers that support "A-for-A" lookups, in
839 contravention of the specifications.
841 PH/48 When a host name was looked up from an IP address, and the subsequent
842 forward lookup of the name timed out, the host name was left in
843 $sender_host_name, contrary to the specification.
845 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
846 restricted to single-key lookups, Exim was not diagnosing an error if
847 * or *@ was used with a query-style lookup.
849 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
851 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
852 long ago, but noone interested enough thought of it. Let's just say that
853 the "1.1" means that there are some new functions that weren't there at
854 some point in the past.
856 PH/51 Error processing for expansion failure of helo_data from an smtp
857 transport during callout processing was broken.
859 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
860 tested/used via the -bh/-bhc/-bs options.
862 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
863 bug, fixed in subsequent PCRE releases).
865 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
866 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
868 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
874 SC/01 Use a glob alias rather than an array ref in eximstats generated
875 parser. This improves both readability and performance.
877 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
878 Don't display local sender or destination tables in eximstats unless
879 there is data to show.
880 Added average volumes into the eximstats top table text output.
882 SC/03 Collect data on the number of addresses (recipients) as well
883 as the number of messages in eximstats.
885 TF/01 Correct an error in the documentation for the redirect router. Exim
886 does (usually) call initgroups() when daemonizing.
888 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
889 with consistent privilege compared to when running as a daemon.
891 TF/03 Note in the spec that $authenticated_id is not set for local
892 submissions from trusted users.
894 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
895 Thanks to Dean Brooks <dean@iglou.com> for the patch.
897 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
898 by adding some example configuration directives to the default
899 configuration file. A little bit of work is required to uncomment the
900 directives and define how usernames and passwords are checked, but
901 there is now a framework to start from.
903 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
904 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
905 without this. I don't know how relevant this is to other LDAP libraries.
907 PH/02 Add the verb name to the "unknown ACL verb" error.
909 PH/03 Magnus Holmgren's patch for filter_prepend_home.
911 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
913 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
914 directory not expanded when it should be if an expanded home directory
915 was set for the address (which is overridden by the transport).
917 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
920 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
921 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
922 because it is too late at that time, and has no effect.
924 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
925 security issue with \' (bugzilla #107). I could not use the
926 PQescapeStringConn() function, because it needs a PGconn value as one of
929 PH/08 When testing addresses using -bt, indicate those final addresses that
930 are duplicates that would not cause an additional delivery. At least one
931 person was confused, thinking that -bt output corresponded to deliveries.
932 (Suppressing duplicates isn't a good idea as you lose the information
933 about possibly different redirections that led to the duplicates.)
935 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
936 systems where poll() doesn't work, in particular OS X.
938 PH/10 Added more information to debugging output for retry time not reached.
940 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
941 operations in malware.c.
943 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
946 PH/13 If write_rejectlog was set false when logging was sent to syslog with
947 syslog_duplication set false, log lines that would normally be written
948 both the the main log and to the reject log were not written to syslog at
951 PH/14 In the default configuration, change the use of "message" in ACL warn
952 statements to "add_header".
954 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
955 not followed by a command (e.g. "seen endif").
957 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
958 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
961 PH/17 Added extra conditions to the default value of delay_warning_condition
965 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
966 { match{$h_precedence:}{(?i)bulk|list|junk} } \
967 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
970 The Auto-Submitted: and various List- headers are standardised, whereas I
971 don't think Precedence: ever was.
973 PH/18 Refactored debugging code in route_finduser() to show more information,
974 in particular, the error code if getpwnam() issues one.
976 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
977 This is apparently needed in addition to the PH/07 change above to avoid
978 any possible encoding problems.
980 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
981 but not after initializing Perl.
983 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
984 output them only if debugging. By default they are written stderr,
985 apparently, which is not desirable.
987 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
990 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
993 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
995 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
996 authenticated or an ident call has been made. Suppress the default
997 values for $authenticated_id and $authenticated_sender (but permit -oMai
998 and -oMas) when testing with -bh.
1000 PH/24 Re-jigged the order of the tests in the default configuration so that the
1001 tests for valid domains and recipients precede the DNS black list and CSA
1002 tests, on the grounds that those ones are more expensive.
1004 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1005 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1006 This bug exists in every version of Exim that I still have, right back to
1009 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1010 However, an attempt to turn on case-sensitivity in a regex key by
1011 including (?-i) didn't work because the subject string was already
1012 lowercased, and the effects were non-intuitive. It turns out that a
1013 one-line patch can be used to allow (?-i) to work as expected.
1019 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1020 other effects) broke the use of negated acl sub-conditions.
1022 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1025 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1026 "Deny" causes Exim to reject the incoming connection with a 554 error.
1027 Unfortunately, if there is a major crisis, such as a disk failure,
1028 tcp-wrappers gives "deny", whereas what one would like would be some
1029 kind of temporary error. A kludge has been added to help with this.
1030 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1031 554 error is used if errno is still zero or contains ENOENT (which occurs
1032 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1035 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1037 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1038 errors. Otherwise a message that provokes a temporary error (when other
1039 messages do not) can cause a whole host to time out.
1041 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1042 addresses were routed directly to files or pipes from a redirect router.
1043 File deliveries just didn't batch; pipe deliveries might have suffered
1046 PH/06 A failure to get a lock for a hints database would erroneously always say
1047 "Failed to get write lock", even when it was really a read lock.
1049 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1050 of 0600. This has been changed to use the value of the lockfile_mode
1051 option (which defaults to 0600).
1053 PH/08 Applied small patch from the Sieve maintainer.
1055 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1056 folder from quota calculations, a direct delivery into this folder messed
1057 up the contents of the maildirsize file. This was because the regex was
1058 used only to exclude .Trash (or whatever) when the size of the mailbox
1059 was calculated. There was no check that a delivery was happening into an
1060 excluded directory. This bug has been fixed by ignoring all quota
1061 processing for deliveries into excluded directories.
1063 PH/10 Added the maildirfolder_create_regex option to appendfile.
1069 PH/01 The code for finding all the local interface addresses on a FreeBSD
1070 system running IPv6 was broken. This may well have applied to all BSD
1071 systems, as well as to others that have similar system calls. The broken
1072 code found IPv4 interfaces correctly, but gave incorrect values for the
1073 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1074 that it would not match correctly against @[] and not recognize the IPv6
1077 PH/02 The ipliteral router was not recognizing addresses of the form user@
1078 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1080 PH/03 Added disable_ipv6.
1082 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1083 decoded form, because it is most often used to construct To: headers
1084 lines in autoreplies, and the decoded form may well be syntactically
1085 invalid. However, $reply_address has leading white space removed, and all
1086 newlines turned into spaces so that the autoreply transport does not
1089 PH/05 If group was specified without a user on a router, and no group or user
1090 was specified on a transport, the group from the router was ignored.
1092 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1093 for visible compile-time settings that can be used to change these
1094 numbers, for those that want even more. Backwards compatibility with old
1095 spool files has been maintained. However, going back to a previous Exim
1096 release will lost any variables that are in spool files.
1098 PH/07 Two small changes when running in the test harness: increase delay when
1099 passing a TCP/IP connection to a new process, in case the original
1100 process has to generate a bounce, and remove special handling of
1101 127.0.0.2 (sic), which is no longer necessary.
1103 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1104 be the same on different OS.
1106 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1109 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1110 whether --show-vars was specified or not
1112 JJ/02 exipick: Added support for new ACL variable spool format introduced
1115 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1116 syntactically invalid From: or Reply-to: line, and a filter used this to
1117 generate an autoreply, and therefore failed to obtain an address for the
1118 autoreply, Exim could try to deliver to a non-existent relative file
1119 name, causing unrelated and misleading errors. What now happens is that
1120 it logs this as a hard delivery error, but does not attempt to create a
1123 PH/11 The exinext utility has a -C option for testing purposes, but although
1124 the given file was scanned by exinext itself; it wasn't being passed on
1125 when Exim was called.
1127 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1128 an end-of-file indication when reading a command response.
1130 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1131 compiled. In many other places in Exim, IPv6 addresses are always
1132 recognized, so I have changed this. It also means that IPv4 domain
1133 literals of the form [IPV4:n.n.n.n] are now always recognized.
1135 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1136 used if the router is not running as root, for example, when verifying at
1137 ACL time, or when using -bh. The debugging output from this situation was
1138 non-existent - all you got was a failure to exec. I have made two
1141 (a) Failures to set uid/gid, the current directory, or a process leader
1142 in a subprocess such as that created by queryprogram now generate
1143 suitable debugging ouput when -d is set.
1145 (b) The queryprogram router detects when it is not running as root,
1146 outputs suitable debugging information if -d is set, and then runs
1147 the subprocess without attempting to change uid/gid.
1149 PH/15 Minor change to Makefile for building test_host (undocumented testing
1152 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1153 additional section of a DNS packet that returns MX or SRV records.
1154 Instead, it always explicitly searches for A/AAAA records. This avoids
1155 major problems that occur when a DNS server includes only records of one
1156 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1157 fixed another bug: if SRV records were looked up and the corresponding
1158 address records were *not* found in the additional section, the port
1159 values from the SRV records were lost.
1161 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1162 using the correct key (the original address) when searching the retry
1163 rules in order to find which one to use for generating the retry hint.
1165 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1166 adding the default one. Similarly, if it contains a Reply-To: header, the
1167 errors_reply_to option, if set, is not used.
1169 PH/19 When calculating a retry time, Exim used to measure the "time since
1170 failure" by looking at the "first failed" field in the retry record. Now
1171 it does not use this if it is later than than the arrival time of the
1172 message. Instead it uses the arrival time. This makes for better
1173 behaviour in cases where some deliveries succeed, thus re-setting the
1174 "first failed" field. An example is a quota failure for a huge message
1175 when small messages continue to be delivered. Without this change, the
1176 "time since failure" will always be short, possible causing more frequent
1177 delivery attempts for the huge message than are intended.
1178 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1180 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1181 $1, $2, $3) because the numerical variables can be reset during some
1182 expansion items (e.g. "match"), thereby losing the authentication data.
1184 PH/21 Make -bV show the size of off_t variables so that the test suite can
1185 decide whether to run tests for quotas > 2G.
1187 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1188 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1189 filecount value is greater than 2G or if a quota value is greater than 2G
1190 on a system where the size of off_t is not greater than 4, a panic error
1193 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1194 never match. The debug and -bh output now contains an explicit error
1195 message indicating a malformed IPv4 address or mask.
1197 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1198 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1199 PH/23 above applies.
1201 PH/25 Do not write to syslog when running in the test harness. The only
1202 occasion when this arises is a failure to open the main or panic logs
1203 (for which there is an explicit test).
1205 PH/26 Added the /no_tell option to "control=freeze".
1207 PH/27 If a host name lookup failed very early in a connection, for example, if
1208 the IP address matched host_lookup and the reverse lookup yielded a name
1209 that did not have a forward lookup, an error message of the form "no IP
1210 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1211 could be logged. Now it outputs the IP address instead of "NULL".
1213 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1214 allows the sender and the authenticated sender to be set when
1215 submitting a message from within Exim. Since child_open_exim() is
1216 documented for local_scan(), the new function should be too.
1218 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1219 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1220 results in an empty string is now treated as unset.
1222 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1224 PH/31 Added sender_verify_fail logging option.
1226 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1227 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1228 tidied the source and removed it altogether.
1230 PH/33 When a queue run was abandoned because the load average was too high, a
1231 log line was always written; now it is written only if the queue_run log
1232 selector is set. In addition, the log line for abandonment now contains
1233 information about the queue run such as the pid. This is always present
1234 in "start" and "stop" lines but was omitted from the "abandon" line.
1236 PH/34 Omit spaces between a header name and the colon in the error message that
1237 is given when verify = headers_syntax fails (if there are lots of them,
1238 the message gets confusing).
1240 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1241 names, as there are now some PTR records that contain slashes. This check
1242 is only to protect against broken name servers that fall over on strange
1243 characters, so the fact that it applies to all lookups doesn't matter.
1245 PH/36 Now that the new test suite is complete, we can remove some of the
1246 special code in Exim that was needed for the old test suite. For example,
1247 sorting DNS records because real resolvers return them in an arbitrary
1248 order. The new test suite's fake resolver always returns records in the
1251 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1252 bounces) except when queue_only is set, to avoid logging races between
1253 the different processes.
1255 PH/38 Panic-die if .include specifies a non-absolute path.
1257 PH/39 A tweak to the "H" retry rule from its user.
1259 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1260 a label. They prevented compilation on older perls.
1262 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1263 a warning to be raised on newish perls.
1265 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1266 on queue. Changes to match documented behaviour of showing count of
1267 messages matching specified criteria.
1269 PH/40 Changed the default ident timeout from 30s to 5s.
1271 PH/41 Added support for the use of login_cap features, on those BSD systems
1272 that have them, for controlling the resources used by pipe deliveries.
1274 PH/42 The content-scanning code uses fopen() to create files in which to put
1275 message data. Previously it was not paying any attention to the mode of
1276 the files. Exim runs with umask(0) because the rest of the code creates
1277 files with open(), and sets the required mode explicitly. Thus, these
1278 files were ending up world-writeable. This was not a big issue, because,
1279 being within the spool directory, they were not world-accessible. I have
1280 created a function called modefopen, which takes an additional mode
1281 argument. It sets umask(777), creates the file, chmods it to the required
1282 mode, then resets the umask. All the relevant calls to fopen() in the
1283 content scanning code have been changed to use this function.
1285 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1286 to 24 hours. This avoids potential overflow problems when processing G
1287 and H retry rules. I suspect nobody ever tinkers with this value.
1289 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1291 PH/45 When the plaintext authenticator is running as a client, the server's
1292 challenges are checked to ensure they are valid base64 strings. By
1293 default, the authentication attempt is cancelled if an invalid string is
1294 received. Setting client_ignore_invalid_base64 true ignores these errors.
1295 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1296 they are received. Thus, the responses can be made to depend on the
1297 challenges. If an invalid string is ignored, an empty string is placed in
1300 PH/46 Messages that are created by the autoreply transport now contains a
1301 References: header, in accordance with RFCs 2822 and 3834.
1303 PH/47 Added authenticated_sender_force to the smtp transport.
1305 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1307 PH/49 Installed latest patch from the Sieve maintainer.
1309 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1310 was also set, the appendfile transport was unnecessarily scanning a
1311 directory of message files (e.g. for maildir delivery) to find the count
1312 of files (along with the size), even though it did not need this
1313 information. It now does the scan only if it needs to find either the
1314 size of the count of files.
1316 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1318 PH/52 Two bugs concerned with error handling when the smtp transport is
1321 (i) Exim was not creating retry information for temporary errors given
1322 for individual recipients after the DATA command when the smtp transport
1323 was used in LMTP mode. This meant that they could be retried too
1324 frequently, and not timed out correctly.
1326 (ii) Exim was setting the flag that allows error details to be returned
1327 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1328 recipients that were returned after the DATA command.
1330 PH/53 This is related to PH/52, but is more general: for any failing address,
1331 when detailed error information was permitted to be returned to the
1332 sender, but the error was temporary, then after the final timeout, only
1333 "retry timeout exceeded" was returned. Now it returns the full error as
1334 well as "retry timeout exceeded".
1336 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1337 do this, and (what is worse) MTAs that accept it.
1339 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1340 will now be deprecated.
1342 PH/56 New os.c-cygwin from the Cygwin maintainer.
1344 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1345 formats (previously only available in exim formats via -bpr, -bpru,
1346 and -bpra. Now also available in native and exiqgrep formats)
1348 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1349 with very large, slow to parse queues
1351 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1353 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1355 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1356 responses to authentication challenges, though it was showing the
1357 challenges; (ii) I've removed the CR characters from the debug output for
1360 PH/58 Allow for the insertion of a newline as well as a space when a string
1361 is turned into more than one encoded-word during RFC 2047 encoding. The
1362 Sieve code now uses this.
1364 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1365 data_4xx, lost_connection, tls_required.
1367 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1368 message was being sent as an SMTP response.
1370 PH/61 Add -l and -k options to exicyclog.
1372 PH/62 When verifying, if an address was redirected to one new address, so that
1373 verification continued, and the new address failed or deferred after
1374 having set something in $address_data, the value of $address_data was not
1375 passed back to the ACL. This was different to the case when no
1376 redirection occurred. The value is now passed back in both cases.
1378 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1379 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1380 use login_cap.h, so on its own it isn't the distinguishing feature. The
1381 new name refers directly to the setclassresources() function.
1383 PH/65 Added configuration files for NetBSD3.
1385 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1387 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1388 is preferred over IPv4.
1390 PH/68 The bounce_return_message and bounce_return_body options were not being
1391 honoured for bounces generated during the reception of non-SMTP messages.
1392 In particular, this applied to messages rejected by the ACL. This bug has
1393 been fixed. However, if bounce_return_message is true and bounce_return_
1394 body is false, the headers that are returned for a non-SMTP message
1395 include only those that have been read before the error was detected.
1396 (In the case of an ACL rejection, they have all been read.)
1398 PH/69 The HTML version of the specification is now built in a directory called
1399 spec_html instead of spec.html, because the latter looks like a path with
1400 a MIME-type, and this confuses some software.
1402 PH/70 Catch two compiler warnings in sieve.c.
1404 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1405 function verify_get_ident() calls ip_connect() to connect a socket, but
1406 if the "connect()" function timed out, ip_connect() used to close the
1407 socket. However, verify_get_ident() also closes the socket later, and in
1408 between Exim writes to the log, which may get opened at this point. When
1409 the socket was closed in ip_connect(), the log could get the same file
1410 descriptor number as the socket. This naturally causes chaos. The fix is
1411 not to close the socket in ip_connect(); the socket should be closed by
1412 the function that creates it. There was only one place in the code where
1413 this was missing, in the iplookup router, which I don't think anybody now
1414 uses, but I've fixed it anyway.
1416 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1417 well as to direct DNS lookups. Otherwise the handling of names in host
1418 lists is inconsistent and therefore confusing.
1424 PH/01 Two changes to the default runtime configuration:
1426 (1) Move the checks for relay_from_hosts and authenticated clients from
1427 after to before the (commented out) DNS black list checks.
1429 (2) Add control=submission to the relay_from_hosts and authenticated
1430 clients checks, on the grounds that messages accepted by these
1431 statements are most likely to be submissions.
1433 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1435 (1) Generate an error if the third argument for the ${prvs expansion is
1438 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1441 (3) Reset the variables that are obtained from the first argument of
1442 ${prvscheck and used in the second argument before leaving the code,
1443 because their memory is reclaimed, so using them afterwards may do
1446 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1447 one (it's much easier than Tom thought :-).
1449 (5) Because of (4), we can now allow for the use of $prvscheck_result
1450 inside the third argument.
1452 PH/03 For some reason, the default setting of PATH when running a command from
1453 a pipe transport was just "/usr/bin". I have changed it to
1456 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1457 anything to be listed in the output from -bV.
1459 PH/05 When a filter generated an autoreply, the entire To: header line was
1460 quoted in the delivery log line, like this:
1462 => >A.N.Other <ano@some.domain> <original@ddress> ...
1464 This has been changed so that it extracts the operative address. There
1465 may be more than one such address. If so, they are comma-separated, like
1468 => >ano@some.domain,ona@other.domain <original@ddress> ...
1470 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1471 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1472 not being looked up in the rDNS to get a host name, Exim was showing the
1473 IP address twice in Received: lines, even though the IP addresses were
1474 identical. For example:
1476 Received: from [1.2.3.4] (helo=[1.2.3.4])
1478 However, if the real host name was known, it was omitting the HELO data
1479 if it matched the actual IP address. This has been tidied up so that it
1480 doesn't show the same IP address twice.
1482 PH/07 When both +timestamp and +memory debugging was on, the value given by
1483 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1484 called by the debug printing, thereby overwriting the timestamp buffer.
1485 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1487 PH/08 When the original message was included in an autoreply transport, it
1488 always said "this is a copy of the message, including all the headers",
1489 even if body_only or headers_only was set. It now gives an appropriate
1492 PH/09 Applied a patch from the Sieve maintainer which:
1494 o fixes some comments
1495 o adds the (disabled) notify extension core
1496 o adds some debug output for the result of if/elsif tests
1497 o points to the current vacation draft in the documentation
1498 and documents the missing references header update
1502 o fixes a bug in processing the envelope test (when testing
1503 multiple envelope elements, the last element determinted the
1506 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1507 Electronic Mail") by including:
1509 Auto-submitted: auto-generated
1511 in the messages that it generates (bounce messages and others, such as
1512 warnings). In the case of bounce messages for non-SMTP mesages, there was
1513 also a typo: it was using "Auto_submitted" (underscore instead of
1514 hyphen). Since every message generated by Exim is necessarily in response
1515 to another message, thes have all been changed to:
1517 Auto-Submitted: auto-replied
1519 in accordance with these statements in the RFC:
1521 The auto-replied keyword:
1523 - SHOULD be used on messages sent in direct response to another
1524 message by an automatic process,
1526 - MUST NOT be used on manually-generated messages,
1528 - MAY be used on Delivery Status Notifications (DSNs) and Message
1529 Disposition Notifications (MDNs),
1531 - MUST NOT be used on messages generated by automatic or periodic
1532 processes, except for messages which are automatic responses to
1535 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1536 to the default Received: header definition.
1538 PH/12 Added log selector acl_warn_skipped (default on).
1540 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1541 variables because (a) they are in the wrong storage pool and (b) even if
1542 they were copied, it wouldn't work properly because of the caching.
1544 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1545 checking when decoding. Apparently there are clients that generate
1546 overlong encoded strings. Why am I not surprised?
1548 PH/15 If the first argument of "${if match_address" was not empty, but did not
1549 contain an "@" character, Exim crashed. Now it writes a panic log message
1550 and treats the condition as false.
1552 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1554 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1555 extension "envelope-auth". The code is finished and in agreement with
1556 other implementations, but there is no documentation so far and in fact,
1557 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1558 not changing the active code.
1560 Print executed "if" and "elsif" statements when debugging is used. This
1561 helps a great deal to understand what a filter does.
1563 Document more things not specified clearly in RFC3028. I had all this
1564 sorted out, when out of a sudden new issues came to my mind. Oops."
1566 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1569 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1570 canonical form (as documented). However, after a host name lookup from
1571 the IP address, check_host() was doing a simple string comparison with
1572 addresses acquired from the DNS when checking that the found name did
1573 have the original IP as one of its addresses. Since any found IPv6
1574 addresses are likely to be in abbreviated form, the comparison could
1575 fail. Luckily, there already exists a function for doing the comparison
1576 by converting both addresses to binary, so now that is used instead of
1577 the text comparison.
1579 PH/20 There was another similar case to PH/19, when a complete host name was
1580 given in a host list; looking up its IP address could give an abbreviated
1581 form, whereas the current host's name might or might not be abbreviated.
1582 The same fix has been applied.
1588 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1589 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1592 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1593 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1595 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1597 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1598 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1599 Also, for "Auto-Submitted", treat anything other than "no" as
1600 non-personal, in accordance with RFC 3834. (Previously it treated
1601 anything starting "auto-" as non-personal.)
1603 TF/01 The control=submission/name=... option had a problem with syntax
1604 errors if the name included a slash character. The /name= option
1605 now slurps the rest of the string, so it can include any characters
1606 but it must come last in the list of options (after /sender_retain
1609 PH/05 Some modifications to the interface to the fake nameserver for the new
1617 TK/01 Added the "success_on_redirect" address verification option. See
1618 NewStuff for rationale and an example.
1620 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1622 PH/02 Patch to exigrep to allow it to work on syslog lines.
1624 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1625 fread() to skip over the body file's header line, because in Cygwin the
1626 header line is locked and is inaccessible.
1628 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1629 co-exist for some time) to make it clear that it is the Exim ID that is
1630 referenced, not the Message-ID: header line.
1632 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1633 string_format() function, because snprintf() does not exist on all
1636 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1637 ${stat: expansion item.
1639 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1640 protocol synchronization error", to keep the pedants happy.
1642 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1643 well as for IRIX systems, when gcc is being used. See the host.c source
1646 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1648 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1651 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1652 transport and to the smtp transport in LMTP mode.
1654 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1656 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1657 run a filter in a subprocess. This could lead to confusion in subsequent
1658 lookups in the parent process. There should also be a search_tidyup() at
1659 the end of the subprocess.
1661 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1662 only if the host matched helo_try_verify_hosts, which caused the
1663 verification to occur when the EHLO/HELO command was issued. The ACL just
1664 tested the remembered result. Now, if a previous verification attempt has
1665 not happened, "verify = helo" does it there and then.
1667 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1669 TK/03 Fix log output including CR from clamd.
1671 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1672 exist provoked a memory error which could cause a segfault.
1674 PH/15 Installed PCRE 6.2
1676 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1678 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1679 of the problem. Specifically, suggested +O2 rather than +O1 for the
1682 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1684 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1685 record, the port was not correctly logged when the outgoing_port log
1686 selector was set (it logged the transort's default port).
1688 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1689 fallback_hosts, and "hosts" in the smtp transport.
1691 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1692 host errors such as "Connection refused".
1694 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1695 authentication with radiusclient 0.4.9:
1697 - Error returned from rc_read_config was caught wrongly
1698 - Username/password not passed on to radius server due to wrong length.
1700 The presumption is that some radiusclient API changes for 4.51/PH/17
1701 were not taken care of correctly. The code is still untested by me (my
1702 Linux distribution still has 0.3.2 of radiusclient), but it was
1703 contributed by a Radius user.
1705 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1706 expanding the "port" option of the smtp transport.
1708 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1709 while reading a MIME header. Thanks to Tom Hughes for a patch.
1711 PH/24 Include config.h inside local_scan.h so that configuration settings are
1714 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1715 that in an ACL for RCPT (for example), you can examine exactly what was
1718 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1719 commands, but it was not correctly comparing the address with the actual
1720 client host address. Thus, it would show the EHLO address in Received:
1721 header lines when this was not necessary.
1723 PH/27 Added the % operator to ${eval:}.
1725 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1726 it should be ignoring failures (because with -C, for example, it has lost
1727 privilege). It wasn't ignoring creation failures other than "already
1730 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1733 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1734 because an input line was too long, either on its own, or by virtue of
1735 too many continuations, the temporary file was not being removed, and the
1736 return code was incorrect.
1738 PH/31 Missing "BOOL" in function definition in filtertest.c.
1740 PH/32 Applied Sieve patches from the maintainer.
1742 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1744 PH/33 Added "verify = not_blind".
1746 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1747 Local/Makefile (with some defaults set). These are used in built scripts
1748 such as exicyclog, but they have never been used in the exim_install
1749 script (though there are many overriding facilities there). I have
1750 arranged that the exim_install script now takes note of these two
1753 PH/35 Installed configuration files for Dragonfly.
1755 PH/36 When a locally submitted message by a trusted user did not contain a
1756 From: header, and the sender address was obtained from -f or from an SMTP
1757 MAIL command, and the trusted user did not use -F to supply a sender
1758 name, $originator_name was incorrectly used when constructing a From:
1759 header. Furthermore, $originator_name was used for submission mode
1760 messages from external hosts without From: headers in a similar way,
1761 which is clearly wrong.
1763 PH/37 Added control=suppress_local_fixups.
1765 PH/38 When log_selector = +received_sender was set, and the addition of the
1766 sender made the log line's construction buffer exactly full, or one byte
1767 less than full, an overflow happened when the terminating "\n" was
1770 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1771 when the result of a list match is failure because a DNS lookup failed.
1773 PH/40 RM_COMMAND is now used in the building process.
1775 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1776 the "build-* directories that it finds.
1778 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1779 address in a domain literal was a prefix of an interface address.
1781 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1782 when verifying a sender address, unless rewrite_headers is false.
1784 PH/44 Wrote a long comment about why errors_to addresses are verified as
1785 recipients, not senders.
1787 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1788 the ratelimit ACL was added.
1790 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1792 PH/46 Added extra information about PostgreSQL errors to the error string.
1794 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1795 avoiding the need to install special zones in a real server. This is
1796 backwards compatible; if it can't find the fake resolver, it drops back.
1797 Thus, both old and new test suites can be run.
1799 TF/02 Added util/ratelimit.pl
1801 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1802 clock is set back in time.
1804 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1805 Candler <B.Candler@pobox.com>.
1807 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1808 OK for addresses that are the result of redirections.
1810 PH/48 A number of further additions for the benefit of the new test suite,
1811 including a fake gethostbyname() that interfaces to the fake DNS resolver
1814 TF/06 The fix for widen_domains has also been applied to qualify_single and
1815 search_parents which are the other dnslookup options that can cause
1818 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1821 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1823 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1824 service" if it encountered an MX record with an empty target hostname.
1825 The message is now "an MX or SRV record indicated no SMTP service".
1827 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1828 if the DNS of the sending site is misconfigured. This is quite a
1829 common situation. This change restores the behaviour of treating a
1830 helo verification defer as a failure.
1832 PH/51 If self=fail was set on a router, the bounce message did not include the
1833 actual error message.
1839 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1841 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1842 command itself ended in error, the underlying message about the transport
1843 filter timeout was being overwritten with the pipe command error. Now the
1844 underlying error message should be appended to the second error message.
1846 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1848 PH/02 Reduce the amount of output that "make" produces by default. Full output
1849 can still be requested.
1851 PH/03 The warning log line about a condition test deferring for a "warn" verb
1852 was being output only once per connection, rather than after each
1853 occurrence (because it was using the same function as for successful
1854 "warn" verbs). This seems wrong, so I have changed it.
1856 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1857 it should not have, which might have caused a crash in the right
1858 circumstances, but probably never did.
1860 PH/04 Installed a modified version of Tony Finch's patch to make submission
1861 mode fix the return path as well as the Sender: header line, and to
1862 add a /name= option so that you can make the user's friendly name appear
1865 TF/03 Added the control = fakedefer ACL modifier.
1867 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1868 Mark Lowes for thorough testing.
1870 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1872 TK/03 Merged latest SRS patch from Miles Wilton.
1874 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1875 with the definition in sysexits.h (which is #included earlier).
1876 Fortunately, Exim does not actually use EX_OK. The code used to try to
1877 preserve the sysexits.h value, by assumimg that macro definitions were
1878 scanned for macro replacements. I have been disabused of this notion,
1879 so now the code just undefines EX_OK before #including unistd.h.
1881 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1882 in the smtp transport. When a block could not be written in a single
1883 write() function, the timeout was being re-applied to each part-write.
1884 This seems wrong - if the receiver was accepting one byte at a time it
1885 would take for ever. The timeout is now adjusted when this happens. It
1886 doesn't have to be particularly precise.
1888 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1889 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1891 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1892 without a domain if the check to <postmaster@domain> fails.
1894 SC/01 Eximstats: added -xls and the ability to specify output files
1895 (patch written by Frank Heydlauf).
1897 SC/02 Eximstats: use FileHandles for outputing results.
1899 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1901 SC/04 Eximstats: fixed display of large numbers with -nvr option
1903 SC/05 Eximstats: fixed merging of reports with empty tables.
1905 SC/06 Eximstats: added the -include_original_destination flag
1907 SC/07 Eximstats: removed tabs and trailing whitespace.
1909 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1911 TK/06 MBOX spool code: Add real "From " MBOX separator line
1912 so the .eml file is really in mbox format (even though
1913 most programs do not really care). Patch from Alex Miller.
1915 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1916 The latter is generated from $received_to and is only set if the
1917 message has one envelope recipient. SA can use these headers,
1918 obviously out-of-the-box. Patch from Alex Miller.
1920 PH/08 The ${def test on a variable was returning false if the variable's
1921 value was "0", contrary to what the specification has always said!
1922 The result should be true unless the variable is empty.
1924 PH/09 The syntax error of a character other than { following "${if
1925 def:variable_name" (after optional whitespace) was not being diagnosed.
1926 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1927 accidental colon was present, for example, could give incorrect results.
1929 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1930 result is used (not including appendfile, where other changes are about
1933 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1934 This involved changing a lot of size variables from int to off_t. It
1935 should work with maildirs and everything.
1937 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1938 spamd dying while we are connected to it.
1940 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1943 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1944 function for BDB 4.3.
1946 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1948 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1949 the sources for PCRE have been re-arranged and more files are now
1952 PH/15 The code I had for printing potentially long long variables in PH/11
1953 above was not the best (it lost precision). The length of off_t variables
1954 is now inspected at build time, and an appropriate printing format (%ld
1955 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1956 to be "long long int" or "long int". This is needed for the internal
1957 formatting function string_vformat().
1959 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1960 the configuration file to be ":syslog", then the script "guesses" where
1961 the logs files are, rather than using the compiled in default. In our
1962 case the guess is not the same as the compiled default, so the script
1963 suddenly stopped working when I started to use syslog. The patch checks
1964 to see if log_file_path is "". If so, it attempts to read it from exim
1965 with no configuration file to get the compiled in version, before it
1966 falls back to the previous guessing code."
1968 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1969 implementing BATV in an Exim configuration. See NewStuff for the gory
1972 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1973 Makefile that are specific to HP-UX.
1975 PH/18 If the "use_postmaster" option was set for a recipient callout together
1976 with the "random" option, the postmaster address was used as the MAIL
1977 FROM address for the random test, but not for the subsequent recipient
1978 test. It is now used for both.
1980 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1981 patch removes a few documentation additions to RFC 3028, because the
1982 latest draft now contains them. It adds the new en;ascii-case comparator
1983 and a new error check for 8bit text in MIME parts. Comparator and
1984 require names are now matched exactly. I enabled the subaddress
1985 extension, but it is not well tested yet (read: it works for me)."
1987 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1988 rework some of the code of TK/09 above to avoid the hardwired use of
1989 "%lld" and "long long". Replaced the call to snprintf() with a call to
1992 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1993 records point to non-existent hosts", "retry timeout exceeded", and
1994 "retry time not reached for any host after a long failure period".
1996 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1997 experimental DomainKeys support:
1999 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2000 (2) On an error such as an illegally used "control", the wrong name for
2001 the control was given.
2003 These problems did NOT occur unless DomainKeys support was compiled.
2005 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2007 PH/24 Added ${if match_ip condition.
2009 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2010 databases so that it will be absolutely obvious if a crash occurs in the
2011 DB library. This is a regular occurrence (often caused by mis-matched
2014 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2015 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2016 picked up on a user's system that detects such things. There doesn't seem
2017 to be a gcc warning option for this - only an attribute that has to be
2018 put on the function's prototype. It seems that in Fedora Core 4 they have
2019 set this on a number of new functions. No doubt there will be more in due
2022 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2023 specify a transport. However, if an address that was verified by such a
2024 router was the subject of a callout, Exim crashed because it tried to
2025 read the rcpt_include_affixes from the non-existent transport. Now it
2026 just assumes that the setting of that option is false. This bug was
2027 introduced by 4.51/PH/31.
2029 PH/28 Changed -d+all to exclude +memory, because that information is very
2030 rarely of interest, but it makes the output a lot bigger. People tend to
2031 do -d+all out of habit.
2033 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2034 code in os-type was giving problems when libc.so lives in lib64, like on
2037 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2038 aren't the modern standard, and it seems that some systems' include files
2039 don't always have them. Exim was already checking for some of the newer
2040 ones like T_AAAA, and defining it itself. I've added checks for all the
2041 record types that Exim uses.
2043 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2044 not automatically generating a new one, as it is supposed to. This
2045 prevented TLS from working. If the file did exist, but contained invalid
2046 data, a new version was generated, as expected. It was only the case of a
2047 non-existent file that was broken.
2049 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2050 with a change in libdomainkeys > 0.64.
2052 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2053 from DNS. If the selector record carries the flag, it now has
2054 precedence over the domain-wide flag.
2056 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2058 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2059 the use of a port name that isn't defined in /etc/services) occurred, the
2060 message was deferred as in a normal delivery, and thus remained on the
2061 spool, instead of being failed because of the mua_wrapper setting. This
2062 is now fixed, and I tidied up some of the mua_wrapper messages at the
2065 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2066 the messages in a hash of arrays rather than using individual hashes.
2067 This is a bit cleaner and results in dramatic memory savings, albeit
2068 at a slight CPU cost.
2070 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2071 as requested by Marc Sherman.
2073 SC/10 Eximstats: added histograms for user specified patterns as requested
2076 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2078 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2079 fopen() in the content-scanning modules that did not already have it.
2085 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2086 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2088 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2090 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2092 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2093 to test Sieve filters that use "vacation".
2095 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2096 that changes the way the GnuTLS parameters are stored in the cache file.
2097 The new format can be generated externally. For backward compatibility,
2098 if the data in the cache doesn't make sense, Exim assumes it has read an
2099 old-format file, and it generates new data and writes a new file. This
2100 means that you can't go back to an older release without removing the
2103 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2104 work if there are any delivery delays because "one_time" forces the
2105 parent to be marked "delivered", so its unseen clone is never tried
2106 again. For this reason, Exim now forbids the simultaneous setting of
2109 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2110 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2111 entry if you want to know the details. The fix, however, neglected to
2112 consider the case when local delivery batching is involved. The test for
2113 "previously delivered" was not happening when checking to see if an
2114 address could be batched with a previous (undelivered) one; under
2115 certain circumstances this could lead to multiple deliveries to the same
2118 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2119 in its include files, and this causes problems building Exim.
2121 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2122 header_syntax) but Exim was just ignoring anything given after a slash.
2123 In particular, this caused confusion with an attempt to use "verify =
2124 reverse_host_lookup/defer_ok". An error is now given when options are
2125 supplied for verify items that do not have them. (Maybe reverse_host_
2126 lookup should have a defer_ok option, but that's a different point.)
2128 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2129 defined by RFC 821) to 2048, because there were problems with some AUTH
2130 commands, and RFC 1869 says the size should be increased for extended
2131 SMTP commands that take arguments.
2133 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2136 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2137 "unknown" error; now it says that the functionality isn't in the binary.
2139 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2140 an address' error message when a string expansion fails (syntax or
2141 whatever). Otherwise the password may appear in the log. Following change
2142 PH/42 below, there is no longer a chance of it appearing in a bounce
2145 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2147 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2148 of its addresses was ever tried. (Bugzilla bug #2.)
2150 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2151 the result incorrectly in the debug output. (It correctly added a newline
2152 to what was transported.)
2154 TF/01 Added $received_time.
2156 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2157 commented out examples of how to interface to a virus scanner and to
2158 SpamAssassin. Also added commented examples of av_scanner and
2159 spamd_address settings.
2161 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2162 and controls are allowed in which ACLs. There were a couple of minor
2163 errors. Some of the entries in the conditions table (which is a table of
2164 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2165 negation of where the condition IS allowed.
2167 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2169 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2170 header file does not have a version number, so I've had to invent a new
2171 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2172 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2173 radiusclient), but it was contributed by a Radius user.
2175 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2176 files or directories, for OpenSSL.
2178 PH/19 When an Exim process that is running as root has to create an Exim log
2179 file, it does so in a subprocess that runs as exim:exim so as to get the
2180 ownership right at creation (otherwise, other Exim processes might see
2181 the file with the wrong ownership). There was no test for failure of this
2182 fork() call, which would lead to the process getting stuck as it waited
2183 for a non-existent subprocess. Forks do occasionally fail when resources
2184 run out. I reviewed all the other calls to fork(); they all seem to check
2187 PH/20 When checking for unexpected SMTP input at connect time (before writing
2188 the banner), Exim was not dealing correctly with a non-positive return
2189 from the read() function. If the client had disconnected by this time,
2190 the result was a log entry for a synchronization error with an empty
2191 string after "input=" when read() returned zero. If read() returned -1
2192 (an event I could not check), uninitialized data bytes were printed.
2193 There were reports of junk text (parts of files, etc) appearing after
2196 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2198 PH/22 Added support for macro redefinition, and (re)definition in between
2199 driver and ACL definitions.
2201 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2202 forgetting to use the resulting value; it was using the unexpanded value.
2204 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2205 hadn't been configured. The fix is from Juergen Kreileder, who
2206 understands it better than I do:
2208 "Here's what I see happening with three configured cyrus_sasl
2209 authenticators configured (plain, login, cram-md5):
2211 On startup auth_cyrus_sasl_init() gets called for each of these.
2212 This means three calls to sasl_listmech() without a specified mech_list.
2213 => SASL tests which mechs of all available mechs actually work
2214 => three warnings about OTP not working
2215 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2217 With the patch, sasl_listmech() also gets called three times. But now
2218 SASL's mech_list option is set to the server_mech specified in the the
2219 authenticator. Or in other words, the answer from sasl_listmech()
2220 gets limited to just the mech you're testing for (which is different
2222 => the return list contains just 'plain' or 'login', 'cram-md5' or
2223 nothing depending on the value of ob->server_mech.
2225 I've just tested the patch: Authentication still works fine,
2226 unavailable mechs specified in the exim configuration are still
2227 caught, and the auth.log warnings about OTP are gone."
2229 PH/25 When debugging is enabled, the contents of the command line are added
2230 to the debugging output, even when log_selector=+arguments is not
2233 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2234 answer is "GNU", and only if the return is "GNU/something" is the answer
2237 PH/27 $acl_verify_message is now set immediately after the failure of a
2238 verification in an ACL, and so is available in subsequent modifiers. In
2239 particular, the message can be preserved by coding like this:
2241 warn !verify = sender
2242 set acl_m0 = $acl_verify_message
2244 Previously, $acl_verify_message was set only while expanding "message"
2245 and "log_message" when a very denied access.
2247 PH/28 Modified OS/os.c-Linux with
2249 -#ifndef OS_LOAD_AVERAGE
2250 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2252 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2253 nomenclature these days.)
2255 PH/29 Installed patch from the Sieve maintainer that adds the options
2256 sieve_useraddress and sieve_subaddress to the redirect router.
2258 PH/30 In these circumstances:
2259 . Two addresses routed to the same list of hosts;
2260 . First host does not offer TLS;
2261 . First host accepts first address;
2262 . First host gives temporary error to second address;
2263 . Second host offers TLS and a TLS session is established;
2264 . Second host accepts second address.
2265 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2266 and peerdn, if requested) that were in fact used only for the second
2269 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2270 attention to any local part prefix or suffix that was matched by the
2271 router that accepted the address. It now behaves in the same way as it
2272 does for delivery: the affixes are removed from the local part unless
2273 rcpt_include_affixes is set on the transport.
2275 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2276 timeout during the DATA phase of an incoming message.
2278 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2279 applied a patch sanctioned by the Sieve maintainer.
2281 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2282 the uid or gid is negative. A case of a negative gid caused this to be
2283 noticed. The fix allows for either to be negative.
2285 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2286 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2289 PH/36 The change PH/12 above was broken. Fixed it.
2291 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2292 the grounds that routing the same address twice would always produce the
2293 same answer. This might have been true once, but it is certainly no
2294 longer true now. Routing a child address may depend on the previous
2295 routing that produced that child. Some complicated redirection strategies
2296 went wrong when messages had multiple recipients, and made Exim's
2297 behaviour dependent on the order in which the addresses were given.
2299 I have moved the duplicate checking until after the routing is complete.
2300 Exim scans the addresses that are assigned to local and remote
2301 transports, and removes any duplicates. This means that more work will be
2302 done, as duplicates will always all be routed, but duplicates are
2303 presumably rare, so I don't expect this is of any significance.
2305 For deliveries to pipes, files, and autoreplies, the duplicate checking
2306 still happens during the routing process, since they are not going to be
2309 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2310 It corrects a timeout issue with spamd. This is Ian's comment: "The
2311 background is that sometimes spamd either never reads data from a
2312 connection it has accepted, or it never writes response data. The exiscan
2313 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2314 blindly assumes that writes won't block so it may never time out."
2316 PH/39 Allow G after quota size as well as K and M.
2318 PH/40 The value set for $authenticated_id in an authenticator may not contain
2319 binary zeroes or newlines because the value is written to log lines and
2320 to spool files. There was no check on this. Now the value is run through
2321 the string_printing() function so that such characters are converted to
2322 printable escape sequences.
2324 PH/41 $message_linecount is a new variable that contains the total number of
2325 lines in the message. Compare $body_linecount, which is the count for the
2328 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2329 bounce and delay warning messages, except in certain special cases, which
2332 (a) An SMTP error message from a remote host;
2333 (b) A message specified in a :fail: redirection;
2334 (c) A message specified in a "fail" command in a system filter;
2335 (d) A message specified in a FAIL return from the queryprogram router;
2336 (e) A message specified by the cannot_route_message router option.
2338 In these cases only, Exim does include the error details in bounce and
2339 warning messages. There are also a few cases where bland messages such
2340 as "unrouteable address" or "local delivery error" are given.
2342 PH/43 $value is now also set for the "else" part of a ${run expansion.
2344 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2345 being worked on, but at least Exim now implements the latest version to
2348 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2349 process to complete was treated as a delivery failure, a timeout while
2350 writing the message to the pipe was logged, but erroneously treated as a
2351 successful delivery. Such timeouts include transport filter timeouts. For
2352 consistency with the overall process timeout, these timeouts are now
2353 treated as errors, giving rise to delivery failures by default. However,
2354 there is now a new Boolean option for the pipe transport called
2355 timeout_defer, which, if set TRUE, converts the failures into defers for
2356 both kinds of timeout. A transport filter timeout is now identified in
2359 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2360 systems where "make" and "gmake" are different, calling "gmake" at top
2361 level broke things. I've arranged for the value of $(MAKE) to be passed
2362 from the Makefile to this script so that it can call the same version of
2366 A note about Exim versions 4.44 and 4.50
2367 ----------------------------------------
2369 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2370 changes of various kinds. As a consequence, a big documentation update was
2371 needed. This delayed the release for rather longer than seemed good, especially
2372 in the light of a couple of (minor) security issues. Therefore, the changes
2373 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2374 release. So 4.44 and 4.50 are in effect two different branches that both start
2377 I have left the 4.50 change log unchanged; it contains all the changes since
2378 4.43. The change log for 4.44 is below; many of its items are identical to
2379 those for 4.50. This seems to be the most sensible way to preserve the
2380 historical information.
2386 1. Minor wording change to the doc/README.SIEVE file.
2388 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2389 computation of the current number of files was incorrect.
2391 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2392 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2395 4. Give more explanation in the error message when the command for a transport
2396 filter fails to execute.
2398 5. There are several places where Exim runs a non-Exim command in a
2399 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2400 was being done only for the command run by the queryprogram router. It is
2401 now done for all such subprocesses. The other cases are: ${run, transport
2402 filters, and the commands run by the lmtp and pipe transports.
2404 6. Added CONFIGURE_GROUP build-time option.
2406 7. Some older OS have a limit of 256 on the maximum number of file
2407 descriptors. Exim was using setrlimit() to set 1000 as a large value
2408 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2409 systems. I've change it so that if it can't get 1000, it tries for 256.
2411 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2412 was an oversight, and furthermore, ever since the addition of extra
2413 controls (e.g. 4.43/32), the checks on when to allow different forms of
2414 "control" were broken. There should now be diagnostics for all cases when a
2415 control that does not make sense is encountered.
2417 9. Added the /retain_sender option to "control=submission".
2419 10. $recipients is now available in the predata ACL (oversight).
2421 11. Tidy the search cache before the fork to do a delivery from a message
2422 received from the command line. Otherwise the child will trigger a lookup
2423 failure and thereby defer the delivery if it tries to use (for example) a
2424 cached ldap connection that the parent has called unbind on.
2426 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2427 of $address_data from the recipient verification was clobbered by the
2428 sender verification.
2430 13. The value of address_data from a sender verification is now available in
2431 $sender_address_data in subsequent conditions in the ACL statement.
2433 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2435 15. Added a new option "connect=<time>" to callout options, to set a different
2438 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2439 was its contents. (It was OK if the option was not defined at all.)
2441 17. A "Completed" log line is now written for messages that are removed from
2442 the spool by the -Mrm option.
2444 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2445 information about exactly what failed.
2447 19. Added -dd to debug only the daemon process.
2449 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2450 handles timeouts, both on the server side and network timeouts. Renamed the
2451 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2453 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2454 It is now set to "smtps".
2456 22. $host_address is now set to the target address during the checking of
2457 ignore_target_hosts.
2459 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2460 being passed; this would have caused $sender_host_name to have been used if
2461 matching the list had actually called for a host name (not very likely,
2462 since this list is usually IP addresses). A host name is now passed as
2465 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2466 code that specifies a non-restarting handler (typically sigaction() in
2467 modern systems) in an attempt to fix a rare and obscure crash bug.
2469 25. Narrowed the window for a race in the daemon that could cause it to ignore
2470 SIGCHLD signals. This is not a major problem, because they are used only to
2471 wake it up if nothing else does.
2473 26. A malformed maildirsize file could cause Exim to calculate negative values
2474 for the mailbox size or file count. Odd effects could occur as a result.
2475 The maildirsize information is now recalculated if the size or filecount
2478 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2479 support for a long time. Removed HAVE_SYS_VFS_H.
2481 28. Installed the latest version of exipick from John Jetmore.
2483 29. In an address list, if the pattern was not a regular expression, an empty
2484 subject address (from a bounce message) matched only if the pattern was an
2485 empty string. Non-empty patterns were not even tested. This was the wrong
2486 because it is perfectly reasonable to use an empty address as part of a
2487 database query. An empty address is now tested by patterns that are
2488 lookups. However, all the other forms of pattern expect the subject to
2489 contain a local part and a domain, and therefore, for them, an empty
2490 address still always fails if the pattern is not itself empty.
2492 30. Exim went into a mad DNS loop when attempting to do a callout where the
2493 host was specified on an smtp transport, and looking it up yielded more
2494 than one IP address.
2496 31. Re-factored the code for checking spool and log partition space into a
2497 function that finds that data and another that does the check. The former
2498 is then used to implement four new variables: $spool_space, $log_space,
2499 $spool_inodes, and $log_inodes.
2501 32. The RFC2047 encoding function was originally intended for short strings
2502 such as real names; it was not keeping to the 75-character limit for
2503 encoded words that the RFC imposes. It now respects the limit, and
2504 generates multiple encoded words if necessary. To be on the safe side, I
2505 have increased the buffer size for the ${rfc2047: expansion operator from
2508 33. It is now permitted to omit both strings after an "if" condition; if the
2509 condition is true, the result is "true". As before, when the second string
2510 is omitted, a false condition yields an empty string. This makes it less
2511 cumbersome to write custom ACL and router conditions.
2513 34. Failure to deliver a bounce message always caused it to be frozen, even if
2514 there was an errors_to setting on the router. The errors_to setting is now
2517 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2518 canonical form (fully expanded) before being placed in
2519 $sender_host_address.
2521 36. The table in the code that translates DNS record types into text (T_A to
2522 "A" for instance) was missing entries for NS and CNAME. It is just possible
2523 that this could have caused confusion if both these types were looked up
2524 for the same domain, because the text type is used as part of Exim's
2525 per-process caching. But the chance of anyone hitting this buglet seems
2528 37. The dnsdb lookup has been extended in a number of ways.
2530 (1) There is a new type, "zns", which walks up the domain tree until it
2531 finds some nameserver records. It should be used with care.
2533 (2) There is a new type, "mxh", which is like "mx" except that it returns
2534 just the host names, not the priorities.
2536 (3) It is now possible to give a list of domains (or IP addresses) to be
2537 looked up. The behaviour when one of the lookups defers can be
2538 controlled by a keyword.
2540 (4) It is now possible to specify the separator character for use when
2541 multiple records are returned.
2543 38. The dnslists ACL condition has been extended: it is now possible to supply
2544 a list of IP addresses and/or domains to be looked up in a particular DNS
2547 39. Added log_selector=+queue_time_overall.
2549 40. When running the queue in the test harness, wait just a tad after forking a
2550 delivery process, to get repeatability of debugging output.
2552 41. Include certificate and key file names in error message when GnuTLS fails
2553 to set them up, because the GnuTLS error message doesn't include the name
2554 of the failing file when there is a problem reading it.
2556 42. Allow both -bf and -bF in the same test run.
2558 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2560 44. The "Exiscan patch" is now merged into the mainline Exim source.
2562 45. Sometimes the final signoff response after QUIT could fail to get
2563 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2564 < 0 before doing a fflush(). This bug looks as though it goes back to the
2565 introduction of TLS in release 3.20, but "sometimes" must have been rare
2566 because the tests only now provoked it.
2568 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2569 (this can affect the format of dates).
2571 47. exim_tidydb, when checking for the continued existence of a message for
2572 which it has found a message-specific retry record, was not finding
2573 messages that were in split spool directories. Consequently, it was
2574 deleting retry records that should have stayed in existence.
2576 48. Steve fixed some bugs in eximstats.
2578 49. The SPA authentication driver was not abandoning authentication and moving
2579 on to the next authenticator when an expansion was forced to fail,
2580 contradicting the general specification for all authenticators. Instead it
2581 was generating a temporary error. It now behaves as specified.
2583 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2584 (the order specifies the preference for clients). The order is now AES256,
2585 AES128, 3DES, ARCFOUR128.
2587 51. Small patch to Sieve code - explicitly set From: when generating an
2590 52. Exim crashed if a remote delivery caused a very long error message to be
2591 recorded - for instance if somebody sent an entire SpamAssassin report back
2592 as a large number of 550 error lines. This bug was coincidentally fixed by
2593 increasing the size of one of Exim's internal buffers (big_buffer) that
2594 happened as part of the Exiscan merge. However, to be on the safe side, I
2595 have made the code more robust (and fixed the comments that describe what
2598 53. Now that there can be additional text after "Completed" in log lines (if
2599 the queue_time_overall log selector is set), a one-byte patch to exigrep
2600 was needed to allow it to recognize "Completed" as not the last thing in
2603 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2604 patch that reportedly fixes this has been added. I am not expert enough to
2605 create a test for it. This is what the patch creator wrote:
2607 "I found a little strange behaviour of ldap code when working with
2608 Windows 2003 AD Domain, where users was placed in more than one
2609 Organization Units. When I tried to give exim partial DN, the exit code
2610 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2611 But simultaneously result of request was absolutely normal ldap result,
2612 so I produce this patch..."
2614 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2615 REFERENCE, so I have modified the code to exclude the patch when that macro
2618 55. Some experimental protocols are using DNS PTR records for new purposes. The
2619 keys for these records are domain names, not reversed IP addresses. The
2620 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2621 leaves it alone. Component reversal etc. now happens only for IP addresses.
2624 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2626 57. Double the size of the debug message buffer (to 2048) so that more of very
2627 long debug lines gets shown.
2629 58. The exicyclog utility now does better if the number of log files to keep
2630 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2632 59. Two changes related to the smtp_active_hostname option:
2634 (1) $smtp_active_hostname is now available as a variable.
2635 (2) The default for smtp_banner uses $smtp_active_hostname instead
2636 of $primary_hostname.
2638 60. The host_aton() function is supposed to be passed a string that is known
2639 to be a valid IP address. However, in the case of IPv6 addresses, it was
2640 not checking this. This is a hostage to fortune. Exim now panics and dies
2641 if the condition is not met. A case was found where this could be provoked
2642 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2643 components; fortuitously, this particular loophole had already been fixed
2644 by change 4.50/55 above.
2646 If there are any other similar loopholes, the new check in host_aton()
2647 itself should stop them being exploited. The report I received stated that
2648 data on the command line could provoke the exploit when Exim was running as
2649 exim, but did not say which command line option was involved. All I could
2650 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2651 running as the user.
2654 61. There was a buffer overflow vulnerability in the SPA authentication code
2655 (which came originally from the Samba project). I have added a test to the
2656 spa_base64_to_bits() function which I hope fixes it.
2659 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2660 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2662 63. The daemon start-up calls getloadavg() while still root for those OS that
2663 need the first call to be done as root, but it missed one case: when
2664 deliver_queue_load_max is set with deliver_drop_privilege. This is
2665 necessary for the benefit of the queue runner, because there is no re-exec
2666 when deliver_drop_privilege is set.
2668 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2669 This has been fixed.
2671 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2672 was in use, was not putting the data itself into the right store pool;
2673 consequently, it could be overwritten for a subsequent message in the same
2674 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2677 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2679 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2680 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2681 the function were treating the return as a boolean value, which happened to
2682 work because 0=false and not-0=true, but is not correct code.
2684 68. The host_aton() function was not handling scoped IPv6 addresses (those
2685 with, for example, "%eth0" on the end) correctly.
2687 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2688 negated items (that is, ~something) in unsigned ints. Some compilers
2689 apparently mutter when there is no cast.
2691 70. If an address verification called from an ACL failed, and did not produce a
2692 user-specific message (i.e. there was only a "system" message), nothing was
2693 put in $acl_verify_message. In this situation, it now puts the system
2696 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2697 session; change 4.31/43 added the unwanted input to the log line - except
2698 that it did not do this in the start of session case. It now does.
2700 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2701 This is wrong and can cause the other end to generate a synchronization
2702 error if it is another Exim or anything else that does the synchronization
2703 check. A QUIT command is no longer sent after a timeout.
2705 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2706 during host lookups.
2708 74. The defer_ok option of callout verification was not working if it was used
2709 when verifying addresses in header lines, that is, for this case:
2711 verify = header_sender/callout=defer_ok
2713 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2714 those file descriptors could be used for SMTP connections. If anything
2715 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2716 could be sent to the SMTP client, causing chaos. The daemon now opens
2717 stdin, stdout, and stderr to /dev/null when it puts itself into the
2720 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2721 log by default. The user can override this with suitable Perl magic.
2723 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2724 the log message when discard triggers, was not working for the DATA ACL or
2725 for the non-SMTP ACL.
2727 78. Error message wording change in sieve.c.
2729 79. If smtp_accept_max_per_host was set, the number of connections could be
2730 restricted to fewer than expected, because the daemon was trying to set up
2731 a new connection before checking whether the processes handling previous
2732 connections had finished. The check for completed processes is now done
2733 earlier. On busy systems, this bug wouldn't be noticed because something
2734 else would have woken the daemon, and it would have reaped the completed
2737 80. If a message was submitted locally by a user whose login name contained one
2738 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2739 It caused a spool format error. I have fixed the spool reading code. A
2740 related problem was that the "from" clause in the Received: line became
2741 illegal because of the space(s). It is now covered by ${quote_local_part.
2743 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2746 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2748 83. Patch from Sieve maintainer to fix unterminated string problem in
2749 "vacation" handling.
2751 84. Some minor changes to the Linux configuration files to help with other
2752 OS variants using glibc.
2754 85. One more patch for Sieve to update vacation handling to latest spec.
2757 ----------------------------------------------------
2758 See the note above about the 4.44 and 4.50 releases.
2759 ----------------------------------------------------
2765 1. Change 4.43/35 introduced a bug that caused file counts to be
2766 incorrectly computed when quota_filecount was set in an appendfile
2769 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2770 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2773 3. Give more explanation in the error message when the command for a transport
2774 filter fails to execute.
2776 4. There are several places where Exim runs a non-Exim command in a
2777 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2778 was being done only for the command run by the queryprogram router. It is
2779 now done for all such subprocesses. The other cases are: ${run, transport
2780 filters, and the commands run by the lmtp and pipe transports.
2782 5. Some older OS have a limit of 256 on the maximum number of file
2783 descriptors. Exim was using setrlimit() to set 1000 as a large value
2784 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2785 systems. I've change it so that if it can't get 1000, it tries for 256.
2787 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2788 was an oversight, and furthermore, ever since the addition of extra
2789 controls (e.g. 4.43/32), the checks on when to allow different forms of
2790 "control" were broken. There should now be diagnostics for all cases when a
2791 control that does not make sense is encountered.
2793 7. $recipients is now available in the predata ACL (oversight).
2795 8. Tidy the search cache before the fork to do a delivery from a message
2796 received from the command line. Otherwise the child will trigger a lookup
2797 failure and thereby defer the delivery if it tries to use (for example) a
2798 cached ldap connection that the parent has called unbind on.
2800 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2801 of $address_data from the recipient verification was clobbered by the
2802 sender verification.
2804 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2805 was its contents. (It was OK if the option was not defined at all.)
2807 11. A "Completed" log line is now written for messages that are removed from
2808 the spool by the -Mrm option.
2810 12. $host_address is now set to the target address during the checking of
2811 ignore_target_hosts.
2813 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2814 being passed; this would have caused $sender_host_name to have been used if
2815 matching the list had actually called for a host name (not very likely,
2816 since this list is usually IP addresses). A host name is now passed as
2819 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2820 code that specifies a non-restarting handler (typically sigaction() in
2821 modern systems) in an attempt to fix a rare and obscure crash bug.
2823 15. Narrowed the window for a race in the daemon that could cause it to ignore
2824 SIGCHLD signals. This is not a major problem, because they are used only to
2825 wake it up if nothing else does.
2827 16. A malformed maildirsize file could cause Exim to calculate negative values
2828 for the mailbox size or file count. Odd effects could occur as a result.
2829 The maildirsize information is now recalculated if the size or filecount
2832 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2833 support for a long time. Removed HAVE_SYS_VFS_H.
2835 18. Updated exipick to current release from John Jetmore.
2837 19. Allow an empty sender to be matched against a lookup in an address list.
2838 Previously the only cases considered were a regular expression, or an
2841 20. Exim went into a mad DNS lookup loop when doing a callout where the
2842 host was specified on the transport, if the DNS lookup yielded more than
2845 21. The RFC2047 encoding function was originally intended for short strings
2846 such as real names; it was not keeping to the 75-character limit for
2847 encoded words that the RFC imposes. It now respects the limit, and
2848 generates multiple encoded words if necessary. To be on the safe side, I
2849 have increased the buffer size for the ${rfc2047: expansion operator from
2852 22. Failure to deliver a bounce message always caused it to be frozen, even if
2853 there was an errors_to setting on the router. The errors_to setting is now
2856 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2857 canonical form (fully expanded) before being placed in
2858 $sender_host_address.
2860 24. Updated eximstats to version 1.33
2862 25. Include certificate and key file names in error message when GnuTLS fails
2863 to set them up, because the GnuTLS error message doesn't include the name
2864 of the failing file when there is a problem reading it.
2866 26. Expand error message when OpenSSL has problems setting up cert/key files.
2869 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2870 (this can affect the format of dates).
2872 28. exim_tidydb, when checking for the continued existence of a message for
2873 which it has found a message-specific retry record, was not finding
2874 messages that were in split spool directories. Consequently, it was
2875 deleting retry records that should have stayed in existence.
2877 29. eximstats updated to version 1.35
2878 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2879 1.35 - bugfix such that pie charts by volume are generated correctly
2881 30. The SPA authentication driver was not abandoning authentication and moving
2882 on to the next authenticator when an expansion was forced to fail,
2883 contradicting the general specification for all authenticators. Instead it
2884 was generating a temporary error. It now behaves as specified.
2886 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2887 (the order specifies the preference for clients). The order is now AES256,
2888 AES128, 3DES, ARCFOUR128.
2890 31. Small patch to Sieve code - explicitly set From: when generating an
2893 32. Exim crashed if a remote delivery caused a very long error message to be
2894 recorded - for instance if somebody sent an entire SpamAssassin report back
2895 as a large number of 550 error lines. This bug was coincidentally fixed by
2896 increasing the size of one of Exim's internal buffers (big_buffer) that
2897 happened as part of the Exiscan merge. However, to be on the safe side, I
2898 have made the code more robust (and fixed the comments that describe what
2901 33. Some experimental protocols are using DNS PTR records for new purposes. The
2902 keys for these records are domain names, not reversed IP addresses. The
2903 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2904 leaves it alone. Component reversal etc. now happens only for IP addresses.
2907 34. The host_aton() function is supposed to be passed a string that is known
2908 to be a valid IP address. However, in the case of IPv6 addresses, it was
2909 not checking this. This is a hostage to fortune. Exim now panics and dies
2910 if the condition is not met. A case was found where this could be provoked
2911 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2912 components; fortuitously, this particular loophole had already been fixed
2913 by change 4.50/55 or 4.44/33 above.
2915 If there are any other similar loopholes, the new check in host_aton()
2916 itself should stop them being exploited. The report I received stated that
2917 data on the command line could provoke the exploit when Exim was running as
2918 exim, but did not say which command line option was involved. All I could
2919 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2920 running as the user.
2923 35. There was a buffer overflow vulnerability in the SPA authentication code
2924 (which came originally from the Samba project). I have added a test to the
2925 spa_base64_to_bits() function which I hope fixes it.
2928 36. The daemon start-up calls getloadavg() while still root for those OS that
2929 need the first call to be done as root, but it missed one case: when
2930 deliver_queue_load_max is set with deliver_drop_privilege. This is
2931 necessary for the benefit of the queue runner, because there is no re-exec
2932 when deliver_drop_privilege is set.
2934 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2935 was in use, was not putting the data itself into the right store pool;
2936 consequently, it could be overwritten for a subsequent message in the same
2937 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2940 38. Sometimes the final signoff response after QUIT could fail to get
2941 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2942 < 0 before doing a fflush(). This bug looks as though it goes back to the
2943 introduction of TLS in release 3.20, but "sometimes" must have been rare
2944 because the tests only now provoked it.
2950 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2951 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2952 Now it returns an integer. A number of calls were still expecting a T/F
2953 return. Fortuitously, in all cases, the tests worked in OK situations,
2954 which is the norm. However, things would have gone wrong on any write
2955 failures on the smtp file descriptor. This function is used when sending
2956 messages over SMTP and also when doing verify callouts.
2958 2. When Exim is called to do synchronous delivery of a locally submitted
2959 message (the -odf or -odi options), it no longer closes stderr before doing
2962 3. Implemented the mua_wrapper option.
2964 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2966 5. Implemented the functions header_remove(), header_testname(),
2967 header_add_at_position(), and receive_remove_recipient(), and exported them
2970 6. If an ACL "warn" statement specified the addition of headers, Exim already
2971 inserted X-ACL-Warn: at the start if there was no header name. However, it
2972 was not making this test for the second and subsequent header lines if
2973 there were newlines in the string. This meant that an invalid header could
2974 be inserted if Exim was badly configured.
2976 7. Allow an ACL "warn" statement to add header lines at the start or after all
2977 the Received: headers, as well as at the end.
2979 8. Added the rcpt_4xx retry error code.
2981 9. Added postmaster_mailfrom=xxx to callout verification option.
2983 10. Added mailfrom=xxxx to the callout verification option, for verify=
2986 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2987 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2989 12. Inserted some casts to stop certain compilers warning when using pointer
2990 differences as field lengths or precisions in printf-type calls (mostly
2991 affecting debugging statements).
2993 13. Added optional readline() support for -be (dynamically loaded).
2995 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2996 same clock tick as a message's arrival, so that its received time was the
2997 same as the "first fail" time on the retry record, and that message
2998 remained on the queue past the ultimate address timeout, every queue runner
2999 would try a delivery (because it was past the ultimate address timeout) but
3000 after another failure, the ultimate address timeout, which should have then
3001 bounced the address, did not kick in. This was a "< instead of <=" error;
3002 in most cases the first failure would have been in the next clock tick
3003 after the received time, and all would be well.
3005 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3006 being recognized when the domain list was tested by the match_domain
3007 condition in an expansion string.
3009 16. Added the ${str2b64: operator.
3011 17. Exim was always calling setrlimit() to set a large limit for the number of
3012 processes, without checking whether the existing limit was already
3013 adequate. (It did check for the limit on file descriptors.) Furthermore,
3014 errors from getrlimit() and setrlimit() were being ignored. Now they are
3015 logged to the main and panic logs, but Exim does carry on, to try to do its
3016 job under whatever limits there are.
3018 18. Imported PCRE 5.0.
3020 19. Trivial typo in log message " temporarily refused connection" (the leading
3023 20. If the log selector return_path_on_delivery was set and an address was
3024 redirected to /dev/null, the delivery process crashed because it assumed
3025 that a return path would always be set for a "successful" delivery. In this
3026 case, the whole delivery is bypassed as an optimization, and therefore no
3029 21. Internal re-arrangement: the function for sending a challenge and reading
3030 a response while authentication was assuming a zero-terminated challenge
3031 string. It's now changed to take a pointer and a length, to allow for
3032 binary data in such strings.
3034 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3036 23. Exim was not respecting finduser_retries when seeking the login of the
3037 uid under which it was called; it was always trying 10 times. (The default
3038 setting of finduser_retries is zero.) Also, it was sleeping after the final
3039 failure, which is pointless.
3041 24. Implemented tls_on_connect_ports.
3043 25. Implemented acl_smtp_predata.
3045 26. If the domain in control=submission is set empty, Exim assumes that the
3046 authenticated id is a complete email address when it generates From: or
3047 Sender: header lines.
3049 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3050 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3051 chown and chgrp in /bin and hostname in /usr/bin.
3053 28. Exim was keeping the "process log" file open after each use, just as it
3054 does for the main log. This opens the possibility of it remaining open for
3055 long periods when the USR1 signal hits a daemon. Occasional processlog
3056 errors were reported, that could have been caused by this. Anyway, it seems
3057 much more sensible not to leave this file open at all, so that is what now
3060 29. The long-running daemon process does not normally write to the log once it
3061 has entered its main loop, and it closes the log before doing so. This is
3062 so that log files can straightforwardly be renamed and moved. However,
3063 there are a couple of unusual error situations where the daemon does write
3064 log entries, and I had neglected to close the log afterwards.
3066 30. The text of an SMTP error response that was received during a remote
3067 delivery was being truncated at 512 bytes. This is too short for some of
3068 the long messages that one sometimes sees. I've increased the limit to
3071 31. It is now possible to make retry rules that apply only when a message has a
3072 specific sender, in particular, an empty sender.
3074 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3075 it possible to be selective about when SMTP synchronization is enforced.
3077 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3079 32. Implemented hosts_connection_nolog.
3081 33. Added an ACL for QUIT.
3083 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3086 35. Added mailbox_size and mailbox_filecount to appendfile.
3088 36. Added control = no_multiline_responses to ACLs.
3090 37. There was a bug in the logic of the code that waits for the clock to tick
3091 in the case where the clock went backwards by a substantial amount such
3092 that the microsecond fraction of "now" was more than the microsecond
3093 fraction of "then" (but the whole seconds number was less).
3095 38. Added support for the libradius Radius client library this is found on
3096 FreeBSD (previously only the radiusclient library was supported).
3102 1. When certain lookups returned multiple values in the form name=value, the
3103 quoting of the values was not always being done properly. Specifically:
3104 (a) If the value started with a double quote, but contained no whitespace,
3106 (b) If the value contained whitespace other than a space character (i.e.
3107 tabs or newlines or carriage returns) it was not quoted.
3108 This fix has been applied to the mysql and pgsql lookups by writing a
3109 separate quoting function and calling it from the lookup code. The fix
3110 should probably also be applied to nisplus, ibase and oracle lookups, but
3111 since I cannot test any of those, I have not disturbed their existing code.
3113 2. A hit in the callout cache for a specific address caused a log line with no
3114 reason for rejecting RCPT. Now it says "Previous (cached) callout
3115 verification failure".
3117 3. There was an off-by-one bug in the queryprogram router. An over-long
3118 return line was truncated at 256 instead of 255 characters, thereby
3119 overflowing its buffer with the terminating zero. As well as fixing this, I
3120 have increased the buffer size to 1024 (and made a note to document this).
3122 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3123 when Exim is waiting for an SMTP response from a remote server, Exim
3124 restarts its select() call on the socket, thereby resetting its timeout.
3125 This is not a problem when such interrupts are rare. Somebody set up a cron
3126 job to run exiwhat every 2 minutes, which is less than the normal select()
3127 timeout (5 or 10 minutes). This meant that the select() timeout never
3128 kicked in because it was always reset. I have fixed this by comparing the
3129 time when an interrupt arrives with the time at the start of the first call
3130 to select(). If more time than the timeout has elapsed, the interrupt is
3131 treated as a timeout.
3133 5. Some internal re-factoring in preparation for the addition of Sieve
3134 extensions (by MH). In particular, the "personal" test is moved to a
3135 separate function, and given an option for scanning Cc: and Bcc: (which is
3136 not set for Exim filters).
3138 6. When Exim created an email address using the login of the caller as the
3139 local part (e.g. when creating a From: or Sender: header line), it was not
3140 quoting the local part when it contained special characters such as @.
3142 7. Installed new OpenBSD configuration files.
3144 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3145 try to make them clearer.
3147 9. Callout options, other than the timeout value, were being ignored when
3148 verifying sender addresses in header lines. For example, when using
3150 verify = header_sender/callout=no_cache
3152 the cache was (incorrectly) being used.
3154 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3155 only the Cygwin environment.
3157 11. When return_path_on_delivery was set as a log selector, if different remote
3158 addresses in the same message used different return paths and parallel
3159 remote delivery occurred, the wrong values would sometimes be logged.
3160 (Whenever a remote delivery process finished, the return path value from
3161 the most recently started remote delivery process was logged.)
3163 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3164 lines when AUTH and/or TLS are in use. This is the "received protocol"
3165 field. Exim used to use "asmtp" for authenticated SMTP, without any
3166 indication (in the protocol name) for TLS use. Now it follows the RFC and
3167 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3168 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3169 names appear in log lines as well as in Received: header lines.
3171 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3172 extensions, and comparison tests, and to fix some bugs.
3174 14. Changes to the "personal" filter test:
3176 (1) The test was buggy in that it was just doing the equivalent of
3177 "contains" tests on header lines. For example, if a user's address was
3178 anne@some.where, the "personal" test would incorrectly be true for
3180 To: susanne@some.where
3182 This test is now done by extracting each address from the header in turn,
3183 and checking the entire address. Other tests that are part of "personal"
3184 are now done using regular expressions (for example, to check local parts
3185 of addresses in From: header lines).
3187 (2) The list of non-personal local parts in From: addresses has been
3188 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3189 taken from the Sieve specification recommendations.
3191 (3) If the message contains any header line starting with "List-" it is
3192 treated as non-personal.
3194 (4) The test for "circular" in the Subject: header line has been removed
3195 because it now seems ill-conceived.
3197 15. Minor typos in src/EDITME comments corrected.
3199 16. Installed latest exipick from John Jetmore.
3201 17. If headers_add on a router specified a text string that was too long for
3202 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3203 of string_sprintf() is now avoided.
3205 18. $message_body_size was not set (it was always zero) when running the DATA
3206 ACL and the local_scan() function.
3208 19. For the "mail" command in an Exim filter, no default was being set for
3209 the once_repeat time, causing a random time value to be used if "once" was
3210 specified. (If the value happened to be <= 0, no repeat happened.) The
3211 default is now 0s, meaning "never repeat". The "vacation" command was OK
3212 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3213 (I found it when inspecting the code).
3215 20. There is now an overall timeout for performing a callout verification. It
3216 defaults to 4 times the callout timeout, which applies to individual SMTP
3217 commands during the callout. The overall timeout applies when there is more
3218 than one host that can be tried. The timeout is checked before trying the
3219 next host. This prevents very long delays if there are a large number of
3220 hosts and all are timing out (e.g. when the network connections are timing
3221 out). The value of the overall timeout can be changed by specifying an
3222 additional sub-option for "callout", called "maxwait". For example:
3224 verify = sender/callout=5s,maxwait=20s
3226 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3227 to the end before writing, but this should make it even safer).
3229 22. Exim was forgetting that it had advertised PIPELINING for the second and
3230 subsequent messages on an SMTP connection. It was also not resetting its
3231 memory on STARTTLS and an internal HELO.
3233 23. When Exim logs an SMTP synchronization error within a session, it now
3234 records whether PIPELINING has been advertised or not.
3236 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3237 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3238 rather than long int.
3240 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3242 26. Added the never_mail option to autoreply.
3248 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3249 crash if the getsockname() call failed; this can happen if a connection is
3250 closed very soon after it is established. The problem was simply in the
3251 order in which certain operations were done, causing Exim to try to write
3252 to the SMTP stream before it had set up the file descriptor. The bug has
3253 been fixed by making things happen in the correct order.
3259 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3260 before the connection was closed, thus losing the rejection response.
3262 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3263 some early Solaris releases, but causes trouble in current releases where
3264 socklen_t is defined.
3266 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3269 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3272 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3273 that releasing the top of it at the end releases what was used for sub-
3274 expansions (unless the block got too big). However, discard this block if
3275 the first thing is a variable or header, so that we can use its block when
3276 it is dynamic (useful for very large $message_headers, for example).
3278 6. Lookups now cache *every* query, not just the most recent. A new, separate
3279 store pool is used for this. It can be recovered when all lookup caches are
3280 flushed. Lookups now release memory at the end of their result strings.
3281 This has involved some general refactoring of the lookup sources.
3283 7. Some code has been added to the store_xxx() functions to reduce the amount
3284 of flapping under certain conditions.
3286 8. log_incoming_interface used to affect only the <= reception log lines. Now
3287 it causes the local interface and port to be added to several more SMTP log
3288 lines, for example "SMTP connection from", and rejection lines.
3290 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3292 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3294 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3295 could be overwritten at the end of the current message (or the start of a
3296 new message if it was set in a HELO ACL). The value is now preserved for
3297 the duration of the SMTP connection.
3299 12. If a transport had a headers_rewrite setting, and a matching header line
3300 contained an unqualified address, that address was qualified, even if it
3301 did not match any rewriting rules. The underlying bug was that the values
3302 of the flags that permit the existence of unqualified sender and recipient
3303 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3304 non-local messages, and by -bnq for local messages) were not being
3305 preserved with the message after it was received.
3307 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3308 "next input=" as part of the text comprising the host identity instead of
3309 the correct text. The code was using the same buffer for two different
3310 strings. However, depending on which order the printing function evaluated
3311 its arguments, the bug did not always show up. Under Linux, for example, my
3312 test suite worked just fine.
3314 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3315 doesn't work with some older versions of Perl. It has been changed to "my",
3316 which in any case is probably the better facility to use.
3318 15. A really picky compiler found some instances of statements for creating
3319 error messages that either had too many or two few arguments for the format
3322 16. The size of the buffer for calls to the DNS resolver has been increased
3323 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3324 for addresses that have a lot of PTR records. This alleviates a problem; it
3325 does not fully solve it.
3327 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3328 buffer now truncates the list and logs the incident, which is the same
3329 action as happens when Exim is looking up a host name and its aliases.
3330 Previously in this situation something unpredictable would happen;
3331 sometimes it was "internal error: store_reset failed".
3333 18. If a server dropped the connection unexpectedly when an Exim client was
3334 using GnuTLS and trying to read a response, the client delivery process
3335 crashed while trying to generate an error log message.
3337 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3338 string, for example:
3340 warn message = H1: something\nH2: something
3342 the text was added as a single header line from Exim's point of view
3343 though it ended up OK in the delivered message. However, searching for the
3344 second and subsequent header lines using $h_h2: did not work. This has been
3345 fixed. Similarly, if a system filter added multiple headers in this way,
3346 the routers could not see them.
3348 20. Expanded the error message when iplsearch is called with an invalid key to
3349 suggest using net-iplsearch in a host list.
3351 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3352 ACLs are no longer actually imposed (and a message to that effect is
3355 22. If a "gecos" field in a passwd entry contained escaped characters, in
3356 particular, if it contained a \" sequence, Exim got it wrong when building
3357 a From: or a Sender: header from that name. A second bug also caused
3358 incorrect handling when an unquoted " was present following a character
3359 that needed quoting.
3361 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3362 was not being matched caselessly.
3364 24. Arranged for all hyphens in the exim.8 source to be escaped with
3367 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3368 properly. Recipient callout cache records were still being keyed to include
3369 the sender, even when use_sender was set false. This led to far more
3370 callouts that were necessary. The sender is no longer included in the key
3371 when use_sender is false.
3373 26. Added "control = submission" modifier to ACLs.
3375 27. Added the ${base62d: operator to decode base 62 numbers.
3377 28. dnsdb lookups can now access SRV records.
3379 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3380 the configuration file.
3382 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3383 (-v) mode. This makes the output for a verbose queue run more intelligible.
3385 31. Added a use_postmaster feature to recipient callouts.
3387 32. Added the $body_zerocount variable, containing the number of binary zero
3388 bytes in the message body.
3390 33. The time of last modification of the "new" subdirectory is now used as the
3391 "mailbox time last read" when there is a quota error for a maildir
3394 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3396 35. Added +ignore_unknown as a special item in host lists.
3398 36. Code for decoding IPv6 addresses in host lists is now included, even if
3399 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3400 address was recognized as an IP address, but was then not correctly decoded
3401 into binary, causing unexpected and incorrect effects when compared with
3408 1. Very minor rewording of debugging text in manualroute to say "list of
3409 hosts" instead of "hostlist".
3411 2. If verify=header_syntax was set, and a header line with an unqualified
3412 address (no domain) and a large number of spaces between the end of the
3413 name and the colon was received, the reception process suffered a buffer
3414 overflow, and (when I tested it) crashed. This was caused by some obsolete
3415 code that should have been removed. The fix is to remove it!
3417 3. When running in the test harness, delay a bit after writing a bounce
3418 message to get a bit more predictability in the log output.
3420 4. Added a call to search_tidyup() just before forking a reception process. In
3421 theory, someone could use a lookup in the expansion of smtp_accept_max_
3422 per_host which, without the tidyup, could leave open a database connection.
3424 5. Added the variables $recipient_data and $sender_data which get set from a
3425 lookup success in an ACL "recipients" or "senders" condition, or a router
3426 "senders" option, similar to $domain_data and $local_part_data.
3428 6. Moved the writing of debug_print from before to after the "senders" test
3431 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3432 problems for message scanning, either using a data ACL, or using
3433 local_scan() because the Received: header was not generated till after they
3434 were called (in order to set the time as the time of reception completion).
3435 I have revised the way this works. The header is now generated after the
3436 body is received, but before the ACL or local_scan() are called. After they
3437 are run, the timestamp in the header is updated.
3443 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3444 before starting a queue runner without re-exec. This happened only when
3445 deliver_drop_privilege was set or when the Exim user was set to root. The
3446 effect of the bug was that timeouts during subsequent deliveries caused
3447 crashes instead of being properly handled. The handler is now left at its
3448 default (and expected) setting.
3450 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3451 message, again when deliver_drop_privilege is set or Exim is run as root.
3452 The bug described in (1) was not present in this case, but the tidying up
3453 of the other signals was missing. I have made the two cases consistent.
3455 3. The ignore_target_hosts setting on a manualroute router was being ignored
3456 for hosts that were looked up using the /MX notation.
3458 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3461 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3462 operated on the sender address. After changing the $sender_address to <>
3463 for the sender address verify, Exim was re-instated it as the original
3464 (before rewriting) address, but remembering that it had rewritten it, so it
3465 wasn't rewriting it again. This bug also had the effect of breaking the
3466 sender address verification caching when the sender address was rewritten.
3468 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3469 This has been changed so that if the ip literal address matches
3470 ignore_target_hosts, the router declines.
3472 7. Added expansion conditions match_domain, match_address, and match_local_
3473 part (NOT match_host).
3475 8. The placeholder for the Received: header didn't have a length field set.
3477 9. Added code to Exim itself and to exim_lock to test for a specific race
3478 condition that could lead to file corruption when using MBX delivery. The
3479 issue is with the lockfile that is created in /tmp. If this file is removed
3480 after a process has opened it but before that process has acquired a lock,
3481 there is the potential for a second process to recreate the file and also
3482 acquire a lock. This could lead to two Exim processes writing to the file
3483 at the same time. The added code performs the same test as UW imapd; it
3484 checks after acquiring the lock that its file descriptor still refers to
3485 the same named file.
3487 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3488 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3491 11. Added the smtp_active_hostname option. If used, this will typically be made
3492 to depend on the incoming interface address. Because $interface_address is
3493 not set up until the daemon has forked a reception process, error responses
3494 that can happen earlier (such as "too many connections") no longer contain
3497 12. If an expansion in a condition on a "warn" statement fails because a lookup
3498 defers, the "warn" statement is abandoned, and the next ACL statement is
3499 processed. Previously this caused the whole ACL to be aborted.
3501 13. Added the iplsearch lookup type.
3503 14. Added ident_timeout as a log selector.
3505 15. Added tls_certificate_verified as a log selector.
3507 16. Added a global option tls_require_ciphers (compare the smtp transport
3508 option of the same name). This controls incoming TLS connections.
3510 17. I finally figured out how to make tls_require_ciphers do a similar thing
3511 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3512 before starting the TLS session.
3514 18. Tabs are now shown as \t in -bP output.
3516 19. If the log selector return_path_on_delivery was set, Exim crashed when
3517 bouncing a message because it had too many Received: header lines.
3519 20. If two routers both had headers_remove settings, and the first one included
3520 a superfluous trailing colon, the final name in the first list and the
3521 first name in the second list were incorrectly joined into one item (with a
3522 colon in the middle).
3528 1. Added -C and -D options to the exinext utility, mainly to make it easier
3529 to include in the automated testing, but these could be helpful when
3530 multiple configurations are in use.
3532 2. The exinext utility was not formatting the output nicely when there was
3533 an alternate port involved in the retry record key, nor when there was a
3534 message id as well (for retries that were specific to a specific message
3535 and a specific host). It was also confused by IPv6 addresses, because of
3536 the additional colons they contain. I have fixed the IPv4 problem, and
3537 patched it up to do a reasonable job for IPv6.
3539 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3540 delivery, the log line now contains "pipelined" if PIPELINING was used.
3542 4. An SMTP transport process used to panic and die if the bind() call to set
3543 an explicit outgoing interface failed. This has been changed; it is now
3544 treated in the same way as a connect() failure.
3546 5. A reference to $sender_host_name in the part of a conditional expansion
3547 that was being skipped was still causing a DNS lookup. This no longer
3550 6. The def: expansion condition was not recognizing references to header lines
3551 that used bh_ and bheader_.
3553 7. Added the _cache feature to named lists.
3555 8. The code for checking quota_filecount in the appendfile transport was
3556 allowing one more file than it should have been.
3558 9. For compatibility with Sendmail, the command line option
3566 and sets the incoming protocol and host name (for trusted callers). The
3567 host name and its colon can be omitted when only the protocol is to be set.
3568 Note the Exim already has two private options, -pd and -ps, that refer to
3569 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3570 "s", but I don't think that's a major issue.
3572 10. A number of refactoring changes to the code, none of which should affect
3575 (a) The number of logging options was getting close to filling up the
3576 32-bit word that was used as a bit map. I have split them into two classes:
3577 those that are passed in the argument to log_write(), and those that are
3578 only ever tested independently outside of that function. These are now in
3579 separate 32-bit words, so there is plenty of room for expansion again.
3580 There is no change in the user interface or the logging behaviour.
3582 (b) When building, for example, log lines, the code previously used a
3583 macro that called string_cat() twice, in order to add two strings. This is
3584 not really sufficiently general. Furthermore, there was one instance where
3585 it was actually wrong because one of the argument was used twice, and in
3586 one call a function was used. (As it happened, calling the function twice
3587 did not affect the overall behaviour.) The macro has been replaced by a
3588 function that can join an arbitrary number of extra strings onto a growing
3591 (c) The code for expansion conditions now uses a table and a binary chop
3592 instead of a serial search (which was left over from when there were very
3593 few conditions). Also, it now recognizes conditions like "pam" even when
3594 the relevant support is not compiled in: a suitably worded error message is
3595 given if an attempt is made to use such a condition.
3597 11. Added ${time_interval:xxxxx}.
3599 12. A bug was causing one of the ddress fields not to be passed back correctly
3600 from remote delivery subprocesses. The field in question was not being
3601 subsequently used, so this caused to problems in practice.
3603 13. Added new log selectors queue_time and deliver_time.
3605 14. Might have fixed a bug in maildirsizefile handling that threw up
3606 "unexpected character" debug warnings, and recalculated the data
3607 unnecessarily. In any case, I expanded the warning message to give more
3610 15. Added the message "Restricted characters in address" to the statements in
3611 the default ACL that block characters like @ and % in local parts.
3613 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3614 Three changes have been made:
3616 (a) There was a serious bug; a negative response to MAIL caused the whole
3617 recipient domain to be cached as invalid, thereby blocking all messages
3618 to all local parts at the same domain, from all senders. This bug has
3619 been fixed. The domain is no longer cached after a negative response to
3620 MAIL if the sender used is not empty.
3622 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3625 (c) A new callout option, "use_sender" has been added for people who want
3626 the modified behaviour.
3632 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3635 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3636 indeed breaks things for older releases.
3638 3. Added additional logging to the case where there is a problem reading data
3639 from a filter that is running in a subprocess using a pipe, in order to
3640 try to track down a specific problem.
3642 4. Testing facility fudge: when running in the test harness and attempting
3643 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3644 getting "No route to host". Convert this to a timeout.
3646 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3649 6. Some OS don't have socklen_t but use size_t instead. This affects the
3650 fifth argument of getsockopt() amongst other things. This is now
3651 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3652 can be set for individual OS. I have set it for SunOS5, OSF1, and
3653 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3654 some earlier ones do not.
3656 7. Change 4.30/15 was not doing the test caselessly.
3658 8. The standard form for an IPv6 address literal was being rejected by address
3659 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3660 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3661 this, as well as the form without the "IPv6" on the front (but only when
3662 address literals are enabled, of course).
3664 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3666 10. Exim crashed if a message with an empty sender address specified by -f
3667 encountered a router with an errors_to setting. This could be provoked only
3668 by a command such as
3672 where an empty string was supplied; "<>" did not hit this bug.
3674 11. Installed PCRE release 4.5.
3676 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3677 remained set. It is now erased.
3679 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3680 times from message ids (which are base 36 rather than the normal 62).
3682 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3683 were being counted as actual protocol errors, and logged if the log
3684 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3685 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3686 and DATA following a set of rejected RCPTs do not count as protocol errors.
3687 In other words, Exim assumes they were pipelined, though this may not
3688 actually be the case. Of course, in all cases the client gets an
3689 appropriate error code.
3691 15. If a lookup fails in an ACL condition, a message about the failure may
3692 be available; it is used if testing the ACL cannot continue, because most
3693 such messages specify what the cause of the deferral is. However, some
3694 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3695 that caused an old message to be retained and used if a later statement
3696 caused a defer, replacing the real cause of the deferral.
3698 16. If an IP address had so many PTR records that the DNS lookup buffer
3699 was not large enough to hold them, Exim could crash while trying to process
3700 the truncated data. It now detects and logs this case.
3702 17. Further to 4.21/58, another change has been made: if (and only if) the
3703 first line of a message (the first header line) ends with CRLF, a bare LF
3704 in a subsequent header line has a space inserted after it, so as not to
3705 terminate the header.
3707 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3708 unnecessarily, used atoi() instead of strtol(), and didn't check the
3709 termination when getting file sizes from file names by regex.
3711 19. Completely re-implemented the support for maildirsize files, in the light
3712 of a number of problems with the previous contributed implementation
3713 (4.30/29). In particular:
3715 . If the quota is zero, the maildirsize file is maintained, but no quota is
3718 . If the maildir directory does not exist, it is created before any attempt
3719 to write a maildirsize file.
3721 . The quota value in the file is just a cache; if the quota is changed in
3722 the transport, the new value overrides.
3724 . A regular expression is available for excluding directories from the
3727 20. The autoreply transport checks the characters in options that define the
3728 message's headers; it allows continued headers, but it was checking with
3729 isspace() after an embedded newline instead of explicitly looking for a
3732 21. If all the "regular" hosts to which an address was routed had passed their
3733 expiry times, and had not reached their retry times, the address was
3734 bounced, even if fallback hosts were defined. Now Exim should go on to try
3737 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3738 equivalent code in the SMTP transport. Some hosts send humungous responses
3739 to HELO/EHLO, more than 1024 it seems.
3741 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3742 gives compiler warnings in some environments. I've now done it "properly",
3745 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3746 (because of problems with the built-in one) was declared to return uschar *
3747 instead of char *, causing compiler failure.
3749 25. Fixed a file descriptor leak when processing alias/forward files.
3751 26. Fixed a minor format string issue in dbfn.c.
3753 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3755 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3756 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3759 29. When a sender address is verified, it is cached, to save repeating the test
3760 when there is more than one recipient in a message. However, when the
3761 verification involves a callout, it is possible for different callout
3762 options to be set for different recipients. It is too complicated to keep
3763 track of this in the cache, so now Exim always runs a verification when a
3764 callout is required, relying on the callout cache for the optimization.
3765 The overhead is duplication of the address routing, but this should not be
3768 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3769 to be verified with callout=postmaster, and the main callout worked but the
3770 postmaster check failed, the verification correctly failed. However, if a
3771 subsequent RCPT command asked for sender verification *without* the
3772 postmaster check, incorrect caching caused this verification also to fail,
3775 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3776 it was not caching the DNS options (qualify_single, search_parents) that
3777 were used when the lookup failed. A subsequent lookup with different
3778 options therefore always gave the same answer, though there were cases
3779 where it should not have. (Example: a "domains = !$mx_any" option on a
3780 dnslookup router: the "domains" option is always processed without any
3781 widening, but the router might have qualify_single set.) Now Exim uses the
3782 cached value only when the same options are set.
3784 32. Added John Jetmore's "exipick" utility to the distribution.
3786 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3787 than a timeout (e.g. a certificate is required, and is not provided), an
3788 Exim server now closes the connection immediately. Previously it waited for
3789 the client to close - but if the client is SSL, it seems that they each
3790 wait for each other, leading to a delay before one of them times out.
3792 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3793 maintained 0.8.x compatibility because I don't think many are using it, and
3794 it is clearly obsolete.
3796 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3799 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3800 client certificate was expired. A simple patch fixes this, though I don't
3801 understand the full logic of why the verify callback is called multiple
3804 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3805 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3806 which causes problems with some clients (such as the Certicom SSL Plus
3807 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3808 disables the coutermeasure allowing Eudora to connect."
3810 38. Exim was not checking that a write() to a log file succeeded. This could
3811 lead to Bad Things if a log got too big, in particular if it hit a file
3812 size limit. Exim now panics and dies if it cannot write to a log file, just
3813 as it does if it cannot open a log file.
3815 39. Modified OS/Makefile-Linux so that it now contains
3817 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3819 The two -D definitions ensure that Exim is compiled with large file
3820 support, which makes it possible to handle log files that are bigger than
3823 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3824 instance) a domain was checked against a named list that involved a lookup,
3825 causing $domain_data to be set, then another domain was checked against the
3826 same list, then the first domain was re-checked, the value of $domain_data
3827 after the final check could be wrong. In particular, if the second check
3828 failed, it could be set empty. This bug probably also applied to
3831 41. The strip_trailing_dot option was not being applied to the address given
3832 with the -f command-line option.
3834 42. The code for reading a message's header from the spool was incrementing
3835 $received_count, but never initializing it. This meant that the value was
3836 incorrect (doubled) while delivering a message in the same process in which
3837 it was received. In the most common configuration of Exim, this never
3838 happens - a fresh exec is done - but it can happen when
3839 deliver_drop_privilege is set.
3841 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3842 it now includes up to 150 characters of the unexpected data in the log
3845 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3846 and building data strings. The size of both of these buffers was 10 000
3847 bytes - far larger than anybody would *ever* want, thought I. Needless to
3848 say, somebody hit the limit. I have increased the maximum line length to
3849 20 000 and the maximum data length of concatenated lines to 100 000. I have
3850 also fixed two bugs, because there was no checking on these buffers. Tsk,
3851 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3852 buffer is too small.
3854 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3855 lsearch lookups. Now it does.
3857 46. When parsing a route_list item in a manualroute router, a fixed-length
3858 buffer was used for the list of hosts. I made this 1024 bytes long,
3859 thinking that nobody would ever have a list of hosts that long. Wrong.
3860 Somebody had a whole pile of complicated expansion conditions, and the
3861 string was silently truncated, leading to an expansion error. It turns out
3862 that it is easier to change to an unlimited length (owing to other changes
3863 that have happened since this code was originally written) than to build
3864 structure for giving a limitation error. The length of the item that
3865 expands into the list of hosts is now unlimited.
3867 47. The lsearch lookup could not handle data where the length of text line was
3868 more than 4095 characters. Such lines were truncated, leading to shortened
3869 data being returned. It should now handle lines of any length.
3871 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3872 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3875 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3876 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3877 no longer happens in comment lines. A list of replacements is now placed
3878 at the head of all of the source files, except those whose only change is
3879 to replace PERL_COMMAND in the very first #! line.
3881 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3882 messages on the queue, with a bottom-up merge sort, using code contributed
3883 by Michael Haardt. This should make operations like -bp somewhat faster on
3884 large queues. It won't affect queue runners, except when queue_run_in_order
3887 51. Installed eximstats 1.31 in the distribution.
3889 52. Added support for SRV lookups to the dnslookup router.
3891 53. If an ACL referred to $message_body or $message_body_end, the value was not
3892 reset for any messages that followed in the same SMTP session.
3894 54. The store-handling optimization for building very long strings was not
3895 differentiating between the different store pools. I don't think this
3896 actually made any difference in practice, but I've tidied it.
3898 55. While running the routers to verify a sender address, $sender_address
3899 was still set to the sender address. This is wrong, because when routing to
3900 send a bounce to the sender, it would be empty. Therefore, I have changed
3901 it so that, while verifying a sender address, $sender_address is set to <>.
3902 (There is no change to what happens when verifying a recipient address.)
3904 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3905 target A or AAAA records (if not already returned) without resetting the
3906 qualify_single or search_parents options of the DNS resolver. These are
3907 inappropriate in this case because the targets of MX and SRV records must
3908 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3909 target that, when qualified, matched something in the local domain. These
3910 two options are now turned off when doing these lookups.
3912 57. It seems that at least some releases of Reiserfs (which does not have the
3913 concept of a fixed number of inodes) returns zero and not -1 for the
3914 number of available inodes. This interacted badly with check_spool_inodes,
3915 which assumed that -1 was the "no such thing" setting. What I have done is
3916 to check that the total number of inodes is greater than zero before doing
3917 the test of how many are available.
3919 58. When a "warn" ACL statement has a log_message modifier, the message is
3920 remembered, and not repeated. This is to avoid a lot of repetition when a
3921 message has many recipients that cause the same warning to be written.
3922 Howewer, Exim was preserving the list of already written lines for an
3923 entire SMTP session, which doesn't seem right. The memory is now reset if a
3924 new message is started.
3926 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3927 debugging output unless log_rewrite was also set.
3929 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3930 of (int)(handle) when we know that handle contains (void *)(-1).
3932 61. The Exim daemon panic-logs an error return when it closes the incoming
3933 connection. However "connection reset by peer" seems to be common, and
3934 isn't really an error worthy of noting specially, so that particular error
3937 62. When Exim is trying to find all the local interfaces, it used to panic and
3938 die if the ioctl to get the interface flags failed. However, it seems that
3939 on at least one OS (Solaris 9) it is possible to have an interface that is
3940 included in the list of interfaces, but for which you get a failure error
3941 for this call. This happens when the interface is not "plumbed" into a
3942 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3943 failure of the "get flags" call assumes that the interface is down.
3945 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3946 makes life easier for people who are doing arithmetic on fields extracted
3947 from dates, where you often get leading zeros that should not be
3948 interpreted as octal.
3950 64. Added qualify_domain to the redirect router, to override the global
3953 65. If a pathologically long header line contained very many addresses (the
3954 report of this problem mentioned 10 000) and each of them was rewritten,
3955 Exim could use up a very large amount of memory. (It kept on making new
3956 copies of the header line as it rewrote, and never released the old ones.)
3957 At the expense of a bit more processing, the header rewriting function has
3958 been changed so that it no longer eats memory in this way.
3960 66. The generation of the Received: header has been moved from the time that a
3961 message starts to be received, to the time that it finishes. The timestamp
3962 in the Received: header should now be very close to that of the <= log
3963 line. There are two side-effects of this change:
3965 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3966 logged header lines no longer include the local Received: line, because
3967 it has not yet been created. The same applies to a copy of the message
3968 that is returned to a non-SMTP sender when a message is rejected.
3970 (b) When a filter file is tested using -bf, no additional Received: header
3971 is added to the test message. After some thought, I decided that this
3974 This change does not affect the value of $received_for. It is still set
3975 after address rewriting, but before local_scan() is called.
3977 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3979 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3980 gave an unhelpful panic error message, and a defer error. I have managed to
3981 change this behaviour so that it now rejects any supplied certificate,
3982 which seems right, as the list of acceptable certificates is empty.
3984 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3985 gave an unhelpful defer error. I have not managed to make this reject any
3986 supplied certificates, but the error message it gives is "no certificate
3987 supplied", which is not helpful.
3989 70. exigrep's output now also includes lines that are not associated with any
3990 message, but which match the given pattern. Implemented by a patch from
3991 Martin Sluka, which also tidied up the Perl a bit.
3993 71. Recipient callout verification, like sender verification, was using <> in
3994 the MAIL FROM command. This isn't really the right thing, since the actual
3995 sender may affect whether the remote host accepts the recipient or not. I
3996 have changed it to use the actual sender in the callout; this means that
3997 the cache record is now keyed on a recipient/sender pair, not just the
3998 recipient address. There doesn't seem to be a real danger of callout loops,
3999 since a callout by the remote host to check the sender would use <>.
4000 [SEE ABOVE: changed after hitting problems.]
4002 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4003 temporary errors. However, in the case of such a code being given after
4004 the end of a data transmission (i.e. after ".") Exim was failing to write
4005 a retry record for the message. (Yes, there was some broken host that was
4006 actually sending 8xx at this point.)
4008 73. An unknown lookup type in a host list could cause Exim to panic-die when
4009 the list was checked. (An example that provoked this was putting <; in the
4010 middle of a list instead of at the start.) If this happened during a DATA
4011 ACL check, a -D file could be left lying around. This kind of configuration
4012 error no longer causes Exim to die; instead it causes a defer errror. The
4013 incident is still logged to the main and panic logs.
4015 74. Buglet left over from Exim 3 conversion. The message "too many messages
4016 in one connection" was written to the rejectlog but not the mainlog, except
4017 when address rewriting (yes!) was being logged.
4019 75. Added write_rejectlog option.
4021 76. When a system filter was run not as root (that is, when system_filter_user
4022 was set), the values of the $n variables were not being returned to the
4023 main process; thus, they were not subsequently available in the $sn
4026 77. Added +return_path_on_delivery log selector.
4028 78. A connection timeout was being treated differently from recipients deferred
4029 when testing hosts_max_try with a message that was older than the host's
4030 retry timeout. (The host should not be counted, thus allowing all hosts to
4031 be tried at least once before bouncing.) This may have been the cause of an
4032 occasionally reported bug whereby a message would remain on the queue
4033 longer than the retry timeout, but would be bounced if a delivery was
4034 forced. I say "may" because I never totally pinned down the problem;
4035 setting up timeout/retry tests is difficult. See also the next item.
4037 79. The ultimate address timeout was not being applied to errors that involved
4038 a combination of host plus message (for example, a timeout on a MAIL
4039 command). When an address resolved to a number of possible hosts, and they
4040 were not all tried for each delivery (e.g. because of hosts_max_try), a
4041 message could remain on the queue longer than the retry timeout.
4043 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4046 81. Fixed an obscure SMTP outgoing bug which required at least the following
4047 conditions: (a) there was another message waiting for the same server;
4048 (b) the server returned 5xx to all RCPT commands in the first message so
4049 that the message was not completed; (c) the server dropped the connection
4050 or gave a negative response to the RSET that Exim sends to abort the
4051 transaction. The observed case was a dropped connection after DATA that had
4052 been sent in pipelining mode. That is, the server had advertised PIPELINING
4053 but was not implementing it correctly. The effect of the bug was incorrect
4054 behaviour, such as trying another host, and this could lead to a crash.
4060 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4061 and daemon.c were passed as pointers to ints; they should have been
4062 pointers to socklen_t variables (which are typically unsigned ints).
4064 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4067 3. Fixed a really odd bug that affected only the testing scheme; patching a
4068 certain fixed string in the binary changed the value of another string that
4069 happened to be identical to the end of the original first string.
4071 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4072 name", it returns that address as the IP address. On some operating
4073 systems (e.g. Solaris), it also passes back the IP address string as the
4074 "host name". However, on others (e.g. Linux), it passes back an empty
4075 string. Exim wasn't checking for this, and was changing the host name to an
4076 empty string, assuming it had been canonicized.
4078 5. Although rare, it is permitted to have more than one PTR record for a given
4079 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4080 all the names associated with an address, because they do in Solaris.
4081 However, it seems that they do not in Linux for data that comes from the
4082 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4083 I found this out when I moved to a new Linux workstation and tried to run
4084 the Exim test suite.
4086 To get round this problem I have changed the code so that it now does its
4087 own call to the DNS to look up PTR records when searching for a host name.
4088 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4089 addresses that are only in /etc/hosts are still found.
4091 This behaviour is, however, controlled by an option called host_lookup_
4092 order, which defaults to "bydns:byaddr". If people want to use the other
4093 order, or indeed, just use one or the other means of lookup, they can
4094 specify it in this variable.
4096 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4097 some operating systems, this comes back from gethostbyaddr() as an empty
4098 string, and this is what Exim used to test for. However, it seems that in
4099 other systems, "." is yielded. Exim now tests for this case too.
4101 7. The values of check_spool_space and check_log_space are now held internally
4102 as a number of kilobytes instead of an absolute number of bytes. If a
4103 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4104 kilobyte. This means that much larger values can be stored.
4106 8. Exim monitor: an attempt to get the action menu when not actually pointing
4107 at a message produces an empty menu entitled "No message selected". This
4108 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4109 no entries in it ("Shell widget menu has zero width and/or height"). So I
4110 have added a single, blank menu entry in this case.
4112 9. Added ${quote_local_part.
4114 10. MIME decoding is now applied to the contents of Subject: header lines when
4117 11. Now that a reference to $sender_host_address automatically causes a reverse
4118 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4119 host lookup before query-style lookups in lists that might use this
4120 variable. This has therefore been abolished, and the "net-" prefix is no
4121 longer necessary for query-style lookups.
4123 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4124 have been a typo for LFLAGS, so it has been changed.
4126 13. The install script calls Exim with "-C /dev/null" in order to find the
4127 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4128 to be output. Howeve, since Exim outputs its version number before the
4129 error, it didn't break the script. It just looked ugly. I fixed this by
4130 always allowing "-C /dev/null" if the caller is root.
4132 14. Ignore overlarge ACL variable number when reading spool file - insurance
4133 against a later release with more variables having written the file.
4135 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4136 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4137 this, as well as the form without the "IPv6" on the front.
4139 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4140 OS/Makefile-Darwin file.
4142 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4143 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4145 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4146 variable. However, some OS use ints for inodes. I've added cast to long int
4147 to get rid of the compiler warning.
4149 19. I had forgotten to lock out "/../" in configuration file names when
4150 ALT_CONFIG_PREFIX was set.
4152 20. Routers used for verification do not need to specify transports. However,
4153 if such a router generated a host list, and callout was configured, Exim
4154 crashed, because it could not find a port number from the (non-existent)
4155 transport. It now assumes port 25 in this circumstance.
4157 21. Added the -t option to exigrep.
4159 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4160 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4161 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4162 them is set, LOOKUP_LSEARCH is forced.
4164 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4166 24. Added sender and host information to the "rejected by local_scan()" log
4167 line; previously there was no indication of these.
4169 25. Added .include_if_exists.
4171 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4172 Linux. It turns out that not all file systems support this. Apparently some
4173 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4174 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4175 not supported on the file descriptor, is now ignored when Exim is trying to
4176 sync a directory. This applies only to Linux.
4178 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4180 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4181 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4184 29. Added support for maildirsize files from supplied patch (modified a bit).
4186 30. The use of :fail: followed by an empty string could lead Exim to respond to
4187 sender verification failures with (e.g.):
4189 550 Verification failed for <xxx>
4190 550 Sender verify failed
4192 where the first response line was missing the '-' that indicates it is not
4193 the final line of the response.
4195 31. The loop for finding the name of the user that called Exim had a hardwired
4196 limit of 10; it now uses the value of finduser_retries, which is used for
4197 all other user lookups.
4199 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4202 33. Exim was neglecting to zero errno before one call of strtol() when
4203 expanding a string and expecting an integer value. On some systems this
4204 resulted in spurious "integer overflow" errors. Also, it was casting the
4205 result into an int without checking.
4207 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4208 did not work. The code looks as if it has *never* worked, though it appears
4209 to have been documented since at least releast 1.62. I have made it work.
4211 35. The "timeout_DNS" error in retry rules, also documented since at least
4212 1.62, also never worked. As it isn't clear exactly what this means, and
4213 clearly it isn't a major issue, I have abolished the feature by treating it
4214 as "timeout", and writing a warning to the main and panic logs.
4216 36. The display of retry rules for -brt wasn't always showing the error code
4219 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4220 timeout_connect_A, timeout_connect_MX.
4222 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4223 to the empty sender.
4225 39. The daemon was not analysing the content of -oX till after it had closed
4226 stderr and disconnected from the controlling terminal. This meant that any
4227 syntax errors were only noted on the panic log, and the return code from
4228 the command was 0. By re-arranging the code a little, I've made the
4229 decoding happen first, so such errors now appear on stderr, and the return
4230 code is 1. However, the actual setting up of the sockets still happens in
4231 the disconnected process, so errors there are still only recorded on the
4234 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4235 connections (as happens on some IP stacks) was logged at start up time as
4236 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4237 it from "IPv6 and IPv4", which means that two separate sockets are being
4240 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4241 says whether AF_INET or AF_INET6 was passed as an argument.
4243 42. Exiwhat output was messed up when time zones were included in log
4246 43. Exiwhat now gives more information about the daemon's listening ports,
4247 and whether -tls-on-connect was used.
4249 44. The "port" option of the smtp transport is now expanded.
4251 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4252 silently ignored. Now an error message is written to the main and panic
4255 46. There's a new ACL modifier called "logwrite" which writes to a log file
4256 as soon as it is encountered.
4258 47. Added $local_user_uid and $local_user_gid at routing time.
4260 48. Exim crashed when trying to verify a sender address that was being
4263 49. Exim was recognizing only a space character after ".include". It now also
4264 recognizes a tab character.
4266 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4267 extracting the relevant information from the specification. The man page no
4268 longer contains scrambled data for the -d option, and I've added a section
4269 at the front about calling Exim under different names.
4271 51. Added "extra_headers" argument to the "mail" command in filter files.
4273 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4276 53. Installed eximstats 1.29.
4278 54. Added transport_filter_timeout as a generic transport option.
4280 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4281 Cc: header lines. This was required by RFC 822, but it not required by RFC
4284 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4285 incoming messages that did not have them. Now it does so only if the
4286 message originates locally, that is, if there is no associated remote host
4287 address. When Resent- header lines are present, this applies to the Resent-
4288 lines rather than the non-Resent- lines.
4290 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4291 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4293 58. Messages for configuration errors now include the name of the main
4294 configuration file - useful now that there may be more than one file in a
4295 list (.included file names were always shown).
4297 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4298 for those rare installations that do not start the daemon as root or run it
4299 setuid root. I've cut out the call to initgroups() if the daemon is not
4302 60. The Exim user and group can now be bound into the binary as text strings
4303 that are looked up at the start of Exim's processing.
4305 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4307 62. Added $mailstore_basename variable.
4309 63. Installed patch to sieve.c from Michael Haardt.
4311 64. When Exim failed to open the panic log after failing to open the main log,
4312 the original message it was trying to log was written to stderr and debug
4313 output, but if they were not available (the usual case in production), it
4314 was lost. Now it is written to syslog before the two lines that record the
4315 failures to open the logs.
4317 65. Users' Exim filters run in subprocesses under the user's uid. It is
4318 possible for a "deliver" command or an alias in a "personal" command to
4319 provoke an address rewrite. If logging of address rewriting is configured,
4320 this fails because the process is not running as root or exim. There may be
4321 a better way of dealing with this, but for the moment (because 4.30 needs
4322 to be released), I have disabled address rewrite logging when running a
4323 filter in a non-root, non-exim process.
4329 1. The buildconfig auxiliary program wasn't quoting the value set for
4330 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4331 not defined. This bug was masked in 4.22 by the effect that was fixed in
4334 2. Some messages that were rejected after a message id was allocated were
4335 shown as "incomplete" by exigrep. It no longer does this for messages that
4336 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4338 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4339 have allow_domain_literals set, the ID did not get logged in the <= line.
4340 Domain literals are now always recognized in Message-ID: header lines.
4342 4. The first argument for a ${extract expansion item is the key name or field
4343 number. Leading and trailing spaces in this item were not being ignored,
4344 causing some misleading effects.
4346 5. When deliver_drop_privilege was set, single queue runner processes started
4347 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4348 same command in the process it spins off) were not dropping privilege.
4350 6. When the daemon running as "exim" started a queue runner, it always
4351 re-executed Exim in the spun-off process. This is a waste of effort when
4352 deliver_drop_privilege is set. The new process now just calls the
4353 queue-runner function directly.
4359 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4362 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4363 ignored. Though the use of -oP was forcing the writing of a pid file, it
4364 was always written to the default place.
4366 3. If the message "no IP address found for host xxxx" is generated during
4367 incoming verification, it is now followed by identification of the incoming
4368 connection (so you can more easily find what provoked it).
4370 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4372 5. Added some features to "harden" Exim a bit more against certain attacks:
4374 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4375 be put in Local/Makefile. This is like the never_users runtime option,
4376 but it cannot be overridden. The default setting is "root".
4378 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4379 prefix string with which any file named in a -C command line option
4382 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4383 is retained for -C and -D only if the caller of Exim is root. Without
4384 it, the exim user may also use -C and -D and retain privilege.
4386 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4387 command line option is disabled.
4389 6. Macro names set by the -D option must start with an upper case letter, just
4390 like macro names defined in the configuration file.
4392 7. Added "dereference=" facility to LDAP.
4394 8. Two instances of the typo "uknown" in the source files are fixed.
4396 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4397 the Configure-Makefile script screwed up while processing it.
4399 10. Incorporated PCRE 4.4.
4401 11. The SMTP synchronization check was not operating right at the start of an
4402 SMTP session. For example, it could not catch a HELO sent before the client
4403 waited for the greeting. There is now a check for outstanding input at the
4404 point when the greeting is written. Because of the duplex, asynchronous
4405 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4406 way, but not yet received, when the check is performed.
4408 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4409 on TCP/IP sockets, because this apparently causes some broken clients to
4412 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4413 unchanged) from the Cygwin maintainer.
4415 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4416 was supported instead of testing for mbx. Effectively a typo.
4418 15. The spa authenticator server code was not checking that the input it
4419 received was valid base64.
4421 16. The debug output line for the "set" modifier in ACLs was not showing the
4422 name of the variable that was being set.
4424 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4426 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4427 lookup on its own. Something else was needed to trigger the lookup. For
4428 example, a match in host_lookup or the need for a host name in a host list.
4429 Now, if $sender_host_name is referenced and the host name has not yet been
4430 looked up, a lookup is performed. If the lookup fails, the variable remains
4431 empty, and $host_lookup_failed is set to "1".
4433 19. Added "eqi" as a case-independent comparison operator.
4435 20. The saslauthd authentication condition could segfault if neither service
4436 nor realm was specified.
4438 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4439 error message that was logged was misleading, and incoming SMTP
4440 connections were dropped. The message is now more accurate, and temporary
4441 errors are given to SMTP connections.
4443 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4444 (except RSET) with a 421 error, until QUIT is received. However, it was
4445 failing to send a response to QUIT.
4447 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4448 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4449 ACL is now run after the helo_try_verify_hosts code.
4451 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4452 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4453 case-independent, so other case variants are also recognized). Apparently
4454 some systems use these upper case variants.
4456 25. If more than two messages were waiting for the same host, and a transport
4457 filter was specified for the transport, Exim sent two messages over the
4458 same TCP/IP connection, and then failed with "socket operation on non-
4459 socket" when it tried to send the third.
4461 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4463 27. The extern definition of crypt16() in expand.c was not being excluded when
4464 the OS had its own crypt16() function.
4466 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4467 as a preferred synonym for return_size_limit, both as an option and as an
4470 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4472 30. Changed the default configuration ACL to relax the local part checking rule
4473 for addresses that are not in any local domains. For these addresses,
4474 slashes and pipe symbols are allowed within local parts, but the sequence
4475 /../ is explicitly forbidden.
4477 31. SPA server authentication was not clearing the challenge buffer before
4480 32. log_message in a "warn" ACL statement was writing to the reject log as
4481 well as to the main log, which contradicts the documentation and doesn't
4482 seem right (because no rejection is happening). So I have stopped it.
4484 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4485 However, I am unable to do any testing of this.
4487 34. Fixed an infelicity in the appendfile transport. When checking directories
4488 for a mailbox, to see if any needed to be created, it was accidentally
4489 using path names with one or more superfluous leading slashes; tracing
4490 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4492 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4493 contents are added to the log line that is written for every discarded
4494 recipient. (Previously a log_message setting was ignored.)
4496 36. The ${quote: operator now quotes the string if it is empty.
4498 37. The install script runs exim in order to find its version number. If for
4499 some reason other than non-existence or emptiness, which it checks, it
4500 could not run './exim', it was installing it with an empty version number,
4501 i.e. as "exim-". This error state is now caught, and the installation is
4504 38. An argument was missing from the function that creates an error message
4505 when Exim fails to connect to the socket for saslauthd authentication.
4506 This could cause Exim to crash, or give a corrupted message.
4508 39. Added isip, isip4, and isip6 to ${if conditions.
4510 40. The ACL variables $acl_xx are now saved with the message, and can be
4511 accessed later in routers, transports, and filters.
4513 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4514 strings in the file are not string-expanded.
4516 42. If a MAIL command specified a SIZE value that was too large to fit into an
4517 int variable, the check against message_size_limit failed. Such values are
4518 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4519 day this will have to be increased, but I don't think I want to be around
4520 when emails are that large.
4527 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4528 iconv() is not standard in FreeBSD.
4530 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4531 IPv6 enabled. The observed symptom was a segmentation fault on return from
4532 the function os_common_find_running_interfaces() in src/os.c.
4534 3. In the check_special_case() function in daemon.c I had used "errno" as an
4535 argument name, which causes warnings on some systems. This was basically a
4536 typo, since it was named "eno" in the comments!
4538 4. The code that waits for the clock to tick (at a resolution of some fraction
4539 of a second) so as to ensure message-id uniqueness was always waiting for
4540 at least one whole tick, when it could have waited for less. [This is
4541 almost certainly not relevant at current processor speeds, where it is
4542 unlikely to ever wait at all. But we try to future-proof.]
4544 5. The function that sleeps for a time interval that includes fractions of a
4545 second contained a race. It did not block SIGALRM between setting the
4546 timer, and suspending (a couple of lines later). If the interval was short
4547 and the sigsuspend() was delayed until after it had expired, the suspension
4548 never ended. On busy systems this could lead to processes getting stuck for
4551 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4552 process, before it forks any delivery processes. The open lookup caching
4553 mechanism meant that the open file or database connection was passed into
4554 the delivery process. The problem was that delivery processes always tidy
4555 up cached lookup data. This could cause a problem for the next delivery
4556 process started by the queue runner, because the external queue runner
4557 process does not know about the closure. So the next delivery process
4558 still has data in the lookup cache. In the case of a file lookup, there was
4559 no problem because closing a file descriptor in a subprocess doesn't affect
4560 the parent. However, if the lookup was caching a connection to a database,
4561 the connection was closed, and the second delivery process was likely to
4562 see errors such as "PGSQL: query failed: server closed the connection
4563 unexpectedly". The problem has been fixed by closing all cached lookups
4564 in a queue runner before running a delivery process.
4566 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4567 seem to have the "const" qualifier which it has on other OS. I've
4570 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4571 *of the same type* (client or server) with the same public name that an
4572 error should be diagnosed.
4574 9. When Exim looked up a host name for an IP address, but failed to find the
4575 original IP address when looking up the host name (a safety check), it
4576 output the message "<ip address> does not match any IP for NULL", which was
4577 confusing, to say the least. The bug was that the host name should have
4578 appeared instead of "NULL".
4580 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4581 user that is built into the binary, and the -C or -D options is used, root
4582 privilege is dropped before the configuration file is read. In addition,
4583 logging is switched to stderr instead of the normal log files. If the
4584 configuration then re-defines the Exim user, the unprivileged environment
4585 is probably not what is expected, so Exim logs a panic warning message (but
4588 However, if deliver_drop_privilege is set, the unprivileged state may well
4589 be exactly what is intended, so the warning has been cut out in that case,
4590 and Exim is allowed to try to write to its normal log files.
4596 1. smtp_return_error_details was not giving details for temporary sender
4597 or receiver verification errors.
4599 2. Diagnose a configuration error if two authenticators have the same public
4602 3. Exim used not to create the message log file for a message until the first
4603 delivery attempt. This could be confusing when incoming messages were held
4604 for policy or load reasons. The message log file is now created at the time
4605 the message is received, and an initial "Received" line is written to it.
4607 4. The automatically generated man page for command line options had a minor
4608 bug that caused no ill effects; however, a more serious problem was that
4609 the procedure for building the man page automatically didn't always
4610 operate. Consequently, release 4.20 contains an out-of-date version. This
4611 shouldn't happen again.
4613 5. When building Exim with embedded Perl support, the script that builds the
4614 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4615 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4617 6. The freeze_tell option was not being used for messages that were frozen on
4618 arrival, either by an ACL or by local_scan().
4620 7. Added the smtp_incomplete_transaction log selector.
4622 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4623 was accepting AUTH without a new EHLO.
4625 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4626 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4629 10. Logging of TCP/IP connections (when configured) now happens in the main
4630 daemon process instead of the child process, so that the TCP/IP connection
4631 count is more accurate (but it can never be perfect).
4633 11. The use of "drop" in a nested ACL was not being handled correctly in the
4634 outer ACL. Now, if condition failure induced by the nested "drop" causes
4635 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4636 or "require"), the connection is dropped.
4638 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4639 that yield "discard" can now be used with an "accept" or a "discard" verb,
4640 but an error is generated for any others (because I can't see a useful way
4641 to define what should happen).
4643 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4644 are now processed in the same way as lines in the Exim configuration file.
4645 In particular, continuation lines are supported.
4647 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4649 15. Added -ti meaning -t -i.
4651 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4652 domains, and warn by logging if others are found.
4654 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4655 returned from GIFCONF when Exim is trying to find the list of interfaces on
4656 a host. The code in os.c has been modified to copy each ifreq to an aligned
4657 structure in all cases.
4659 Also, in some cases, the returned ifreq's were being copied to a 'struct
4660 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4661 means the last couple of bytes of an IPv6 address could be chopped if the
4662 ifreq contained only a normal sockaddr (14 bytes storage).
4664 18. Named domain lists were not supported in the hosts_treat_as_local option.
4665 An entry such as +xxxx was not recognized, and was treated as a literal
4668 19. Ensure that header lines added by a DATA ACL are included in the reject log
4669 if the ACL subsequently rejects the message.
4671 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4672 MD5 (which is deprecated).
4674 21. When testing a filter file using -bf, Exim was writing a message when it
4675 took the sender from a "From " line in the message, but it was not doing so
4676 when it took $return_path from a Return-Path: header line. It now does.
4678 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4679 with a valid header line field name (a series of printing characters
4680 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4682 23. Changed "disc" in the source to "disk" to conform to the documentation and
4683 the book and for uniformity.
4685 24. Ignore Sendmail's -Ooption=value command line item.
4687 25. When execve() failed while trying to run a command in a pipe transport,
4688 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4689 could be confused with a return value of 69 from the command itself. This
4690 has been changed to 127, the value the shell returns if it is asked to run
4691 a non-existent command. The wording for the related log line suggests a
4692 non-existent command as the problem.
4694 26. If received_header_text expands to an empty string, do not add a Received:
4695 header line to the message. (Well, it adds a token one on the spool, but
4696 marks it "old" so that it doesn't get used or transmitted.)
4698 27. Installed eximstats 1.28 (addition of -nt option).
4700 28. There was no check for failure on the call to getsockname() in the daemon
4701 code. This can fail if there is a shortage of resources on the system, with
4702 ENOMEM, for example. A temporary error is now given on failure.
4704 29. Contrary to the C standard, it seems that in some environments, the
4705 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4706 program. Exim now does this explicitly; it affects the formatting of
4707 timestamps using strftime().
4709 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4710 complaints. I've now initialized all the variables, to avoid this.
4712 32. Header lines added by a system filter were not being "seen" during
4713 transport-time rewrites.
4715 33. The info_callback() function passed to OpenSSL is set up with type void
4716 (*)(SSL *, int, int), as described somewhere. However, when calling the
4717 function (actually a macro) that sets it up, the type void(*)() is
4718 expected. I've put in a cast to prevent warnings from picky compilers.
4720 34. If a DNS black list lookup found a CNAME record, but there were no A
4721 records associated with the domain it pointed at, Exim crashed.
4723 35. If a DNS black list lookup returned more than one A record, Exim ignored
4724 all but the first. It now scans all returned addresses if a particular IP
4725 value is being sought. In this situation, the contents of the
4726 $dnslist_value variable are a list of all the addresses, separated by a
4729 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4730 to accept a host name and all its aliases if the forward lookup for any of
4731 them yielded the IP address of the incoming connection. Now it accepts only
4732 those names whose forward lookup yields the correct IP address. Any other
4733 names are discarded. This closes a loophole whereby a rogue DNS
4734 administrator could create reverse DNS records to break through a
4735 wildcarded host restriction in an ACL.
4737 37. If a user filter or a system filter that ran in a subprocess used any of
4738 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4739 the wrong values were passed to the pipe command ($thisaddress had the
4740 value of $0, $0 had the value of $1, etc). This bug was introduced by
4741 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4743 38. Improved the line breaking for long SMTP error messages from ACLs.
4744 Previously, if there was no break point between 40 and 75 characters, Exim
4745 left the rest of the message alone. Two changes have been made: (a) I've
4746 reduced the minimum length to 35 characters; (b) if it can't find a break
4747 point between 35 and 75 characters, it looks ahead and uses the first one
4748 that it finds. This may give the occasional overlong line, but at least the
4749 remaining text gets split now.
4751 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4752 file descriptors might be low, and that setting 1000 would always raise it.
4753 It turns out that in some environments, the limit is already over 1000 and
4754 that lowering it causes trouble. So now Exim takes care not to decrease it.
4756 40. When delivering a message, the value of $return_path is set to $sender_
4757 address at the start of routing (routers may change the value). By an
4758 oversight, this default was not being set up when an address was tested by
4759 -bt or -bv, which affected the outcome if any router or filter referred to
4762 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4763 log only when "message" or "log_message" are set. However, if one of the
4764 conditions was an address verification, or a call to a nested ACL, the
4765 messages generated by the underlying test were being passed through. This
4766 no longer happens. The underlying message is available in $acl_verify_
4767 message for both "message" and "log_message" expansions, so it can be
4768 passed through if needed.
4770 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4771 new expansion $bh_ to give the encoded byte string without charset
4772 translation. Translation happens only if iconv() is available; HAVE_ICONV
4773 indicates this at build time. HEADERS_CHARSET gives the charset to
4774 translate to; headers_charset can change it in the configuration, and
4775 "headers charset" can change it in an individual filter file.
4777 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4778 that creates RFC 2047 encoded "words" labels them as that charset instead
4779 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4780 expansion operator; (ii) when Exim creates a From: line for a local
4781 message; (iii) when a header line is rewritten to include a "phrase" part.
4783 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4784 buggy, causing it to skip the first lines of messages whose message ID
4785 ended in 'D'. This would not have bitten before Exim release 4.14, because
4786 message IDs were unlikely to end in 'D' before then. The effect was to have
4787 incorrect size information for certain domains.
4789 45. #include "config.h" was missing at the start of the crypt16.c module. This
4790 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4793 46. If there was a timeout during a "random" callout check, Exim treated it as
4794 a failure of the random address, and carried on sending RSET and the real
4795 address. If the delay was just some slowness somewhere, the response to the
4796 original RCPT would be taken as a response to RSET and so on, causing
4797 mayhem of various kinds.
4799 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4800 when I implemented it. It didn't allow for the fact that some option values
4801 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4802 the right test for positive values.
4804 48. Domain names in DNS records are case-independent. Exim always looks them up
4805 in lower case. Some resolvers return domain names in exactly the case they
4806 appear in the zone file, that is, they may contain uppercase letters. Not
4807 all resolvers do this - some return always lower case. Exim was treating a
4808 change of case by a resolver as a change of domain, similar to a widening
4809 of a domain abbreviation. This triggered its re-routing code and so it was
4810 trying to route what was effectively the same domain again. This normally
4811 caused routing to fail (because the router wouldn't handle the domain
4812 twice). Now Exim checks for this case specially, and just changes the
4813 casing of the domain that it ultimately uses when it transmits the message
4816 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4819 50. If a filter generated a file delivery with a non-absolute name (possible if
4820 no home directory exists for the router), the forbid_file option was not
4823 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4824 the existing equality matching.
4826 52. Exim was using ints instead of ino_t variables in some places where it was
4827 dealing with inode numbers.
4829 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4830 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4831 called TMPDIR, and if it finds it is different, it changes its value.
4833 54. The smtp_printf() function is now made available to local_scan() so
4834 additional output lines can be written before returning. There is also an
4835 smtp_fflush() function to enable the detection of a dropped connection.
4836 The variables smtp_input and smtp_batched_input are exported to
4839 55. Changed the default runtime configuration: the message "Unknown user"
4840 has been removed from the ACL, and instead placed on the localuser router,
4841 using the cannot_route_message feature. This means that any verification
4842 failures that generate their own messages won't get overridden. Similarly,
4843 the "Unrouteable address" message that was in the ACL for unverifiable
4844 relay addresses has also been removed.
4846 56. Added hosts_avoid_esmtp to the smtp transport.
4848 57. The exicyclog script was not checking for the esoteric option
4849 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4850 will work only if exicyclog is run under the appropriate euid.
4852 58. Following a discussion on the list, the rules by which Exim recognises line
4853 endings on incoming messages have been changed. The -dropcr and drop_cr
4854 options are now no-ops, retained only for backwards compatibility. The
4855 following line terminators are recognized: LF CRLF CR. However, special
4856 processing applies to CR:
4858 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4859 nor a local message in the state where . is a terminator.
4861 (ii) If a bare CR is encountered in a header line, an extra space is added
4862 after the line terminator so as not to end the header. The reasoning
4863 behind this is that bare CRs in header lines are most likely either
4864 to be mistakes, or people trying to play silly games.
4866 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4867 was being incorrectly given as 18 bytes larger than it should have been.
4868 This is a VOB (very old bug).
4870 60. This may never have affected anything current, but just in case it has:
4871 When the local host is found other than at the start of a list of hosts,
4872 the local host, those with the same MX, and any that follow, are discarded.
4873 When the list in question was part of a longer list of hosts, the following
4874 hosts (not currently being processed) were also being discarded. This no
4875 longer happens. I'm not sure if this situation could ever has previously
4878 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4881 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4882 header. This is something that is recommended in a new Internet Draft, and
4883 is something that is documented as being done by Sendmail. There are two
4884 possible values. For messages generated by the autoreply transport, Exim
4887 Auto-Submitted: auto-replied
4889 whereas for all other generated messages (e.g. bounces) it adds
4891 Auto-Submitted: auto-generated
4893 63. The "personal" condition in filters now includes a test for the
4894 Auto-Submitted: header. If it contains the string "auto-" the message it
4895 not considered personal.
4897 64. Added rcpt_include_affixes as a generic transport option.
4899 65. Added queue_only_override (default true).
4901 66. Added the syslog_duplication option.
4903 67. If what should have been the first header line of a message consisted of
4904 a space followed by a colon, Exim was mis-interpreting it as a header line.
4905 It isn't of course - it is syntactically invalid and should therefore be
4906 treated as the start of the message body. The misbehaviour could have
4907 caused a number of strange effects, including loss of data in subsequent
4908 header lines, and spool format errors.
4910 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4911 client host had authenticated. This control can now be exercised by an ACL
4912 for more flexibility.
4914 69. By default, callouts do not happen when testing with -bh. There is now a
4915 variant, -bhc, which does actually run the callout code, including
4916 consulting and updating the callout cache.
4918 70. Added support for saslauthd authentication, courtesy of Alexander
4921 71. If statvfs() failed on the spool or log directories while checking their
4922 size for availability, Exim confusingly gave the error "space shortage".
4923 Furthermore, in debugging mode it crashed with a floating point exception.
4924 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4925 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4926 serious problem, Exim now writes to the main and panic logs when this
4927 happens, with details of the failure. It then refuses to accept the
4928 incoming message, giving the message "spool directory problem" or "log
4929 directory problem" with a 421 code for SMTP messages.
4931 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4932 0, 1, and 2 exist, because some OS complain for execs without them (see
4933 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4934 descriptors. However, the code omitted to check that the open succeeded,
4935 causing mysterious errors if for some reason the permissions on /dev/null
4936 got screwed. Now Exim writes a message to the main and panic logs, and
4937 bombs out if it can't open /dev/null.
4939 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4940 interact so that it is all more flexible. It is supposed to remain
4941 backwards compatible. Also added extra_local_interfaces.
4943 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4944 to bomb out with an assertion failure - to the client this appears as a
4945 connection drop. This problem occurs in the part of the code that was taken
4946 from the Samba project. Fortunately, the assertion is in a very simple
4947 function, so I have fixed this by reproducing the function inline in the
4948 one place where it is called, and arranging for authentication to fail
4949 instead of killing the process with assert().
4951 75. The SPA client code was not working when the server requested OEM rather
4952 than Unicode encoding.
4954 76. Added code to make require_files with a specific uid setting more usable in
4955 the case where statting the file as root fails - usually a non-root-mounted
4956 NFS file system. When this happens and the failure is EACCES, Exim now
4957 forks a subprocess and does the per-uid checking as the relevant uid.
4959 77. Added process_log_path.
4961 78. If log_file_path was not explicitly set, a setting of check_log_space or
4962 check_log_inodes was ignored.
4964 79. If a space check for the spool or log partitions fails, the incident is now
4965 logged. Of course, in the latter case the data may get lost...
4967 80. Added the %p formatting code to string_format() so that it can be used to
4968 print addresses in debug_print(). Adjusted all the address printing in the
4969 debugging in store.c to use %p rather than %d.
4971 81. There was a concern that a line of code in smtp_in.c could overflow a
4972 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4973 initially expressed, the concern was not well-founded, because trailing
4974 spaces are removed early. However, if the trailing spaces were followed by
4975 a NULL, they did not get removed, so the overflow was possible. Two fixes
4978 (a) I re-wrote the offending code in a cleaner fashion.
4979 (b) If an incoming SMTP command contains a NULL character, it is rejected
4982 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4983 runs initgroups(), so that if the Exim user is in any additional groups,
4984 they will be used during message reception.
4990 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob