1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.557 2008/12/12 14:36:37 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
80 NM/09 Bugzilla 787: Potential buffer overflow in string_format
81 Patch provided by Eugene Bujak
87 TK/01 Add preliminary DKIM support. Currently requires a forked version of
88 ALT-N's libdkim that I have put here:
89 http://duncanthrax.net/exim-experimental/
91 Note to Michael Haardt: I had to rename some vars in sieve.c. They
92 were called 'true' and it seems that C99 defines that as a reserved
93 keyword to be used with 'bool' variable types. That means you could
94 not include C99-style headers which use bools without triggering
95 build errors in sieve.c.
97 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
98 as mailq or other aliases. Changed the --help handling significantly
99 to do whats expected. exim_usage() emits usage/help information.
101 SC/01 Added the -bylocaldomain option to eximstats.
103 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
105 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
107 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
113 PH/01 Another patch from the Sieve maintainer.
115 PH/02 When an IPv6 address is converted to a string for single-key lookup
116 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
117 dots are used instead of colons so that keys in lsearch files need not
118 contain colons. This was done some time before quoting was made available
119 in lsearch files. However, iplsearch files do require colons in IPv6 keys
120 (notated using the quote facility) so as to distinguish them from IPv4
121 keys. This meant that lookups for IP addresses in host lists did not work
122 for iplsearch lookups.
124 This has been fixed by arranging for IPv6 addresses to be expressed with
125 colons if the lookup type is iplsearch. This is not incompatible, because
126 previously such lookups could never work.
128 The situation is now rather anomolous, since one *can* have colons in
129 ordinary lsearch keys. However, making the change in all cases is
130 incompatible and would probably break a number of configurations.
132 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
135 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
136 conversion specification without a maximum field width, thereby enabling
137 a rogue spamd server to cause a buffer overflow. While nobody in their
138 right mind would setup Exim to query an untrusted spamd server, an
139 attacker that gains access to a server running spamd could potentially
140 exploit this vulnerability to run arbitrary code as the Exim user.
142 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
143 $primary_hostname instead of what libspf2 thinks the hosts name is.
145 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
146 a directory entry by the name of the lookup key. Previously, if a
147 symlink pointed to a non-existing file or a file in a directory that
148 Exim lacked permissions to read, a lookup for a key matching that
149 symlink would fail. Now it is enough that a matching directory entry
150 exists, symlink or not. (Bugzilla 503.)
152 PH/03 The body_linecount and body_zerocount variables are now exported in the
155 PH/04 Added the $dnslist_matched variable.
157 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
158 This means they are set thereafter only if the connection becomes
161 PH/06 Added the client_condition to authenticators so that some can be skipped
162 by clients under certain conditions.
164 PH/07 The error message for a badly-placed control=no_multiline_responses left
165 "_responses" off the end of the name.
167 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
169 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
170 (without spaces) instead of just copying the configuration text.
172 PH/10 Added the /noupdate option to the ratelimit ACL condition.
174 PH/11 Added $max_received_linelength.
176 PH/12 Added +ignore_defer and +include_defer to host lists.
178 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
179 way in which PCRE > 7.0 is built.
181 PH/14 Implemented queue_only_load_latch.
183 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
184 MAIL command. The effect was to mangle the value on 64-bit systems.
186 PH/16 Another patch from the Sieve maintainer.
188 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
190 PH/18 If a system quota error occurred while trying to create the file for
191 a maildir delivery, the message "Mailbox is full" was not appended to the
192 bounce if the delivery eventually timed out. Change 4.67/27 below applied
193 only to a quota excession during the actual writing of the file.
195 PH/19 It seems that peer DN values may contain newlines (and other non-printing
196 characters?) which causes problems in log lines. The DN values are now
197 passed through string_printing() before being added to log lines.
199 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
200 and InterBase are left for another time.)
202 PH/21 Added message_body_newlines option.
204 PH/22 Guard against possible overflow in moan_check_errorcopy().
206 PH/23 POSIX allows open() to be a macro; guard against that.
208 PH/24 If the recipient of an error message contained an @ in the local part
209 (suitably quoted, of course), incorrect values were put in $domain and
210 $local_part during the evaluation of errors_copy.
216 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
217 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
220 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
221 issue a MAIL command.
223 PH/02 In an ACL statement such as
225 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
227 if a client was not listed at all, or was listed with a value other than
228 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
229 the condition was not true (as it should be), so access was not denied.
230 The bug was that the ! inversion was incorrectly passed on to the second
231 item. This has been fixed.
233 PH/03 Added additional dnslists conditions == and =& which are different from
234 = and & when the dns lookup returns more than one IP address.
236 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
237 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
239 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
240 FSYNC, which compiles an option called disable_fsync that allows for
241 bypassing fsync(). The documentation is heavily laced with warnings.
243 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
245 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
246 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
247 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
248 including adding "make clean"; (3) Added -fPIC when compiling the test
249 dynamically loaded module, to get rid of a warning.
251 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
252 message fails, move_frozen_messages = true and ignore_bounce_errors_after
253 = 0s. The bug is otherwise harmless.
255 PH/07 There was a bug in the dovecot authenticator such that the value of
256 $auth1 could be overwritten, and so not correctly preserved, after a
257 successful authentication. This usually meant that the value preserved by
258 the server_setid option was incorrect.
260 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
262 PH/09 Installed PCRE release 7.0.
264 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
265 run for batched SMTP input. It is now run at the start of every message
266 in the batch. While fixing this I discovered that the process information
267 (output by running exiwhat) was not always getting set for -bs and -bS
268 input. This is fixed, and it now also says "batched" for BSMTP.
270 PH/11 Added control=no_pipelining.
272 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
273 patch, slightly modified), and move the expansion of helo_data till after
274 the connection is made in the smtp transport (so it can use these
277 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
279 PH/14 Added log_selector = +pid.
281 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
283 PH/16 Add ${if forany and ${if forall.
285 PH/17 Added dsn_from option to vary the From: line in DSNs.
287 PH/18 Flush SMTP output before performing a callout, unless control =
288 no_callout_flush is set.
290 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
291 was true (the default) a successful delivery failed to delete the retry
292 item, thus causing premature timeout of the address. The bug is now
295 PH/20 Added hosts_avoid_pipelining to the smtp transport.
297 PH/21 Long custom messages for fakedefer and fakereject are now split up
298 into multiline reponses in the same way that messages for "deny" and
299 other ACL rejections are.
301 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
302 with slight modification.
304 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
305 draft, changing the syntax and factoring some duplicate code".
307 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
308 for deliveries of the second and subsequent messages over the same SMTP
311 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
312 ${reduce, with only minor "tidies".
314 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
316 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
317 expansion side effects.
319 PH/27 When a message times out after an over-quota error from an Exim-imposed
320 quota, the bounce message says "mailbox is full". This message was not
321 being given when it was a system quota that was exceeded. It now should
324 MH/03 Made $recipients available in local_scan(). local_scan() already has
325 better access to the recipient list through recipients_list[], but
326 $recipients can be useful in postmaster-provided expansion strings.
328 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
329 in the case of a MAIL command with additional options following the
330 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
331 were accidentally chopped off.
333 PH/29 SMTP synchronization checks are implemented when a command is read -
334 there is a check that no more input is waiting when there shouldn't be
335 any. However, for some commands, a delay in an ACL can mean that it is
336 some time before the response is written. In this time, more input might
337 arrive, invalidly. So now there are extra checks after an ACL has run for
338 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
339 pipelining has not been advertised.
341 PH/30 MH's patch to allow iscntrl() characters to be list separators.
343 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
344 returned in the SMTP response when smtp_return_error_details was false.
347 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
348 instead of the C I/O that was originally supplied, because problems were
351 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
352 Exim which did not show up earlier: it was assuming that a call to
353 SSL_CTX_set_info_callback() might give an error value. In fact, there is
354 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
355 was a macro that became an assignment, so it seemed to work. This has
356 changed to a proper function call with a void return, hence the compile
357 error. Exim's code has been fixed.
359 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
362 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
364 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
366 PH/37 If a message is not accepted after it has had an id assigned (e.g.
367 because it turns out to be too big or there is a timeout) there is no
368 "Completed" line in the log. When some messages of this type were
369 selected by exigrep, they were listed as "not completed". Others were
370 picked up by some special patterns. I have improved the selection
371 criteria to be more general.
373 PH/38 The host_find_failed option in the manualroute router can now be set
374 to "ignore", to completely ignore a host whose IP address cannot be
375 found. If all hosts are ignored, the behaviour is controlled by the new
376 host_all_ignored option.
378 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
379 homing or because of multiple MX records with /mx) generated more than
380 one IP address, and the following item turned out to be the local host,
381 all the secondary addresses of the first item were incorrectly removed
382 from the list, along with the local host and any following hosts (which
383 is what is supposed to happen).
385 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
386 whoever called Exim into the -H file. In the case of the daemon it was
387 behaving confusingly. When first started, it used values for whoever
388 started the daemon, but after a SIGHUP it used the Exim user (because it
389 calls itself on a restart). I have changed the code so that it now always
392 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
393 message are rejected with the same error (e.g. no authentication or bad
394 sender address), and a DATA command is nevertheless sent (as can happen
395 with PIPELINING or a stupid MUA), the error message that was given to the
396 RCPT commands is included in the rejection of the DATA command. This is
397 intended to be helpful for MUAs that show only the final error to their
400 PH/42 Another patch from the Sieve maintainer.
402 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
403 Eximstats - Fixed some broken HTML links and added missing column headers
405 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
406 columns for Rejects, Temp Rejects, Ham, and Spam rows.
408 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
410 PH/43 Yet another patch from the Sieve maintainer.
412 PH/44 I found a way to check for a TCP/IP connection going away before sending
413 the response to the final '.' that terminates a message, but only in the
414 case where the client has not sent further data following the '.'
415 (unfortunately, this is allowed). However, in many cases there won't be
416 any further data because there won't be any more messages to send. A call
417 to select() can be used: if it shows that the input is "ready", there is
418 either input waiting, or the socket has been closed. An attempt to read
419 the next input character can distinguish the two cases. Previously, Exim
420 would have sent an OK response which the client would never have see.
421 This could lead to message repetition. This fix should cure that, at
422 least in a lot of common cases.
424 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
425 advertised in response to EHLO.
431 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
432 fixed by 4.65/MH/01 (is this a record?) are fixed:
434 (i) An empty string was always treated as zero by the numeric comparison
435 operators. This behaviour has been restored.
437 (ii) It is documented that the numeric comparison operators always treat
438 their arguments as decimal numbers. This was broken in that numbers
439 starting with 0 were being interpreted as octal.
441 While fixing these problems I realized that there was another issue that
442 hadn't been noticed. Values of message_size_limit (both the global option
443 and the transport option) were treated as octal if they started with 0.
444 The documentation was vague. These values are now always treated as
445 decimal, and I will make that clear in the documentation.
451 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
452 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
455 MH/01 Don't check that the operands of numeric comparison operators are
456 integers when their expansion is in "skipping" mode (fixes bug
457 introduced by 4.64-PH/07).
459 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
460 child addresses, Exim now panics and dies. Previously, because the count
461 is held in a short int, deliveries were likely to be lost. As such a
462 large number of recipients for a single message is ridiculous
463 (performance will be very, very poor), I have chosen to impose a limit
464 rather than extend the field.
470 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
471 leftover -K file (the existence of which was triggered by #402).
472 While we were at it, introduced process PID as part of the -K
473 filename. This should rule out race conditions when creating
476 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
477 processing considerably. Previous code took too long for large mails,
478 triggering a timeout which in turn triggers #401.
480 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
481 in the DK code in transports.c. sendfile() is not really portable,
482 hence the _LINUX specificness.
484 TF/01 In the add_headers option to the mail command in an Exim filter,
485 there was a bug that Exim would claim a syntax error in any
486 header after the first one which had an odd number of characters
489 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
490 callout verification, Exim cached a "reject" for the entire domain. This
491 is correct for most verifications, but it is not correct for a recipient
492 verification with use_sender or use_postmaster set, because in that case
493 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
494 case of MAIL FROM:<> rejection from other early rejections (e.g.
495 rejection of HELO). When verifying a recipient using a non-null MAIL
496 address, the cache is ignored if it shows MAIL FROM:<> rejection.
497 Whatever the result of the callout, the value of the domain cache is
498 left unchanged (for any other kind of callout, getting as far as trying
499 RCPT means that the domain itself is ok).
501 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
504 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
505 manifest itself as EPIPE rather than ECONNECT. When tidying away a
506 session, the daemon ignores ECONNECT errors and logs others; it now
507 ignores EPIPE as well.
509 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
510 (quoted-printable decoding).
512 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
513 later the small subsequent patch to fix an introduced bug.
515 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
517 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
519 PH/08 An error is now given if message_size_limit is specified negative.
521 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
522 to be given (somewhat) arbitrary names.
524 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
527 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
528 miscellaneous code fixes
530 PH/10 Added the log_reject_target ACL modifier to specify where to log
533 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
534 hostname. This is wrong, because it relates to the incoming message (and
535 probably the interface on which it is arriving) and not to the outgoing
536 callout (which could be using a different interface). This has been
537 changed to use the value of the helo_data option from the smtp transport
538 instead - this is what is used when a message is actually being sent. If
539 there is no remote transport (possible with a router that sets up host
540 addresses), $smtp_active_hostname is used.
542 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
543 tweaks were necessary in order to get it to work (see also 21 below):
544 (a) The code assumed that strncpy() returns a negative number on buffer
545 overflow, which isn't the case. Replaced with Exim's string_format()
547 (b) There were several signed/unsigned issues. I just did the minimum
548 hacking in of casts. There is scope for a larger refactoring.
549 (c) The code used strcasecmp() which is not a standard C function.
550 Replaced with Exim's strcmpic() function.
551 (d) The code set only $1; it now sets $auth1 as well.
552 (e) A simple test gave the error "authentication client didn't specify
553 service in request". It would seem that Dovecot has changed its
554 interface. Fortunately there's a specification; I followed it and
555 changed what the client sends and it appears to be working now.
557 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
560 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
561 address A is aliased to B and C, where B exists and C does not. Without
562 -v the output is "A verified" because verification stops after a
563 successful redirection if more than one address is generated. However,
564 with -v the child addresses are also verified. Exim was outputting "A
565 failed to verify" and then showing the successful verification for C,
566 with its parentage. It now outputs "B failed to verify", showing B's
567 parentage before showing the successful verification of C.
569 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
570 look up a TXT record in a specific list after matching in a combined
573 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
574 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
575 they consult the DNS. I had assumed they would set it the way they
576 wanted; and indeed my experiments on Linux seem to show that in some
577 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
578 To be on the safe side, however, I have now made the interface to
579 host_find_byname() similar to host_find_bydns(), with an argument
580 containing the DNS resolver options. The host_find_byname() function now
581 sets these options at its start, just as host_find_bydns() does. The smtp
582 transport options dns_qualify_single and dns_search_parents are passed to
583 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
584 of host_find_byname() use the default settings of RES_DEFNAMES
585 (qualify_single) but not RES_DNSRCH (search_parents).
587 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
588 spool_read_header() do less string testing, by means of a preliminary
589 switch on the second character of optional "-foo" lines. (This is
590 overdue, caused by the large number of possibilities that now exist.
591 Originally there were few.) While I was there, I also converted the
592 str(n)cmp tests so they don't re-test the leading "-" and the first
593 character, in the hope this might squeeze out yet more improvement.
595 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
596 flag allowing group syntax was set by the header_syntax check but not
597 turned off, possible causing trouble later; (2) The flag was not being
598 set at all for the header_verify test, causing "group"-style headers to
599 be rejected. I have now set it in this case, and also caused header_
600 verify to ignore an empty address taken from a group. While doing this, I
601 came across some other cases where the code for allowing group syntax
602 while scanning a header line wasn't quite right (mostly, not resetting
603 the flag correctly in the right place). These bugs could have caused
604 trouble for malformed header lines. I hope it is now all correct.
606 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
607 with the "reply" argument non-NULL. The code, however (which originally
608 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
609 but it didn't always do it. This confused somebody who was copying the
610 code for some other use. I have removed all the tests.
612 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
613 feature that was used to support insecure browsers during the U.S. crypto
614 embargo. It requires special client support, and Exim is probably the
615 only MTA that supported it -- and would never use it because real RSA is
616 always available. This code has been removed, because it had the bad
617 effect of slowing Exim down by computing (never used) parameters for the
618 RSA_EXPORT functionality.
620 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
621 authenticator to fail if there's a tab character in the incoming data
622 (there should never be unless someone is messing about, as it's supposed
623 to be base64-encoded). Also added, on Timo's advice, the "secured" option
624 if the connection is using TLS or if the remote IP is the same as the
625 local IP, and the "valid-client-cert option" if a client certificate has
628 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
629 authenticators. This can be used for authorization after authentication
630 succeeds. (In the case of plaintext, it servers for both authentication
633 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
634 if any retry times were supplied.
636 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
637 connection, where there is no client IP address to check. In this
638 situation, the verify now always succeeds.
640 PH/25 Applied John Jetmore's -Mset patch.
642 PH/26 Added -bem to be like -Mset, but loading a message from a file.
644 PH/27 In a string expansion for a processed (not raw) header when multiple
645 headers of the same name were present, leading whitespace was being
646 removed from all of them, but trailing whitespace was being removed only
647 from the last one. Now trailing whitespace is removed from each header
648 before concatenation. Completely empty headers in a concatenation (as
651 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
652 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
654 PH/29 [Removed. This was a change that I later backed out, and forgot to
655 correct the ChangeLog entry (that I had efficiently created) before
656 committing the later change.]
658 PH/30 Exim was sometimes attempting to deliver messages that had suffered
659 address errors (4xx response to RCPT) over the same connection as other
660 messages routed to the same hosts. Such deliveries are always "forced",
661 so retry times are not inspected. This resulted in far too many retries
662 for the affected addresses. The effect occurred only when there were more
663 hosts than the hosts_max_try setting in the smtp transport when it had
664 the 4xx errors. Those hosts that it had tried were not added to the list
665 of hosts for which the message was waiting, so if all were tried, there
666 was no problem. Two fixes have been applied:
668 (i) If there are any address or message errors in an SMTP delivery, none
669 of the hosts (tried or untried) are now added to the list of hosts
670 for which the message is waiting, so the message should not be a
671 candidate for sending over the same connection that was used for a
672 successful delivery of some other message. This seems entirely
673 reasonable: after all the message is NOT "waiting for some host".
674 This is so "obvious" that I'm not sure why it wasn't done
675 previously. Hope I haven't missed anything, but it can't do any
676 harm, as the worst effect is to miss an optimization.
678 (ii) If, despite (i), such a delivery is accidentally attempted, the
679 routing retry time is respected, so at least it doesn't keep
680 hammering the server.
682 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
683 in ${readsocket because some servers need this prod.
685 PH/32 Added some extra debug output when updating a wait-xxx database.
687 PH/33 The hint "could be header name not terminated by colon", which has been
688 given for certain expansion errors for a long time, was not being given
689 for the ${if def:h_colon_omitted{... case.
691 PH/34 The spec says: "With one important exception, whenever a domain list is
692 being scanned, $domain contains the subject domain." There was at least
693 one case where this was not true.
695 PH/35 The error "getsockname() failed: connection reset by peer" was being
696 written to the panic log as well as the main log, but it isn't really
697 panic-worthy as it just means the connection died rather early on. I have
698 removed the panic log writing for the ECONNRESET error when getsockname()
701 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
702 runs only) independently of the message's sender address. This meant
703 that, if the 4xx error was in fact related to the sender, a different
704 message to the same recipient with a different sender could confuse
705 things. In particualar, this can happen when sending to a greylisting
706 server, but other circumstances could also provoke similar problems.
707 I have changed the default so that the retry time for these errors is now
708 based a combination of the sender and recipient addresses. This change
709 can be overridden by setting address_retry_include_sender=false in the
712 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
713 remote server are returned as part of bounce messages. This was not
714 happening for LMTP over a pipe (the lmtp transport), but now it is the
715 same for both kinds of LMTP.
717 PH/38 Despite being documented as not happening, Exim was rewriting addresses
718 in header lines that were in fact CNAMEs. This is no longer the case.
720 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
721 and queue runs started by the daemon processed all messages. This has
722 been fixed so that -R and -S can now usefully be given with -q<time>.
724 PH/40 Import PCRE release 6.7 (fixes some bugs).
726 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
728 PH/42 Give an error if -q is specified more than once.
730 PH/43 Renamed the variables $interface_address and $interface_port as
731 $received_ip_address and $received_port, to make it clear that these
732 values apply to message reception, and not to the outgoing interface when
733 a message is delivered. (The old names remain recognized, of course.)
735 PH/44 There was no timeout on the connect() call when using a Unix domain
736 socket in the ${readsocket expansion. There now is.
738 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
739 be meaningful with "accept".
741 SC/01 Eximstats V1.43
742 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
744 SC/02 Eximstats V1.44
745 Use a glob alias rather than an array ref in the generated
746 parser. This improves both readability and performance.
748 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
749 Collect SpamAssassin and rejection statistics.
750 Don't display local sender or destination tables unless
751 there is data to show.
752 Added average volumes into the top table text output.
754 SC/04 Eximstats V1.46
755 Collect data on the number of addresses (recipients)
756 as well as the number of messages.
758 SC/05 Eximstats V1.47
759 Added 'Message too big' to the list of mail rejection
760 reasons (thanks to Marco Gaiarin).
762 SC/06 Eximstats V1.48
763 Mainlog lines which have GMT offsets and are too short to
764 have a flag are now skipped.
766 SC/07 Eximstats V1.49 (Alain Williams)
767 Added the -emptyok flag.
769 SC/08 Eximstats V1.50
770 Fixes for obtaining the IP address from reject messages.
772 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
773 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
774 whitesspace changes from 4.64-PH/27
776 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
779 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
780 are found, allow negative numbers in numeric criteria)
782 JJ/06 exipick.20061117.2, added new $message_body_missing variable
784 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
785 to match changes made in 4.64-PH/43
787 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
789 PH/47 Put in an explicit test for a DNS lookup of an address record where the
790 "domain" is actually an IP address, and force a failure. This locks out
791 those revolvers/nameservers that support "A-for-A" lookups, in
792 contravention of the specifications.
794 PH/48 When a host name was looked up from an IP address, and the subsequent
795 forward lookup of the name timed out, the host name was left in
796 $sender_host_name, contrary to the specification.
798 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
799 restricted to single-key lookups, Exim was not diagnosing an error if
800 * or *@ was used with a query-style lookup.
802 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
804 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
805 long ago, but noone interested enough thought of it. Let's just say that
806 the "1.1" means that there are some new functions that weren't there at
807 some point in the past.
809 PH/51 Error processing for expansion failure of helo_data from an smtp
810 transport during callout processing was broken.
812 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
813 tested/used via the -bh/-bhc/-bs options.
815 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
816 bug, fixed in subsequent PCRE releases).
818 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
819 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
821 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
827 SC/01 Use a glob alias rather than an array ref in eximstats generated
828 parser. This improves both readability and performance.
830 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
831 Don't display local sender or destination tables in eximstats unless
832 there is data to show.
833 Added average volumes into the eximstats top table text output.
835 SC/03 Collect data on the number of addresses (recipients) as well
836 as the number of messages in eximstats.
838 TF/01 Correct an error in the documentation for the redirect router. Exim
839 does (usually) call initgroups() when daemonizing.
841 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
842 with consistent privilege compared to when running as a daemon.
844 TF/03 Note in the spec that $authenticated_id is not set for local
845 submissions from trusted users.
847 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
848 Thanks to Dean Brooks <dean@iglou.com> for the patch.
850 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
851 by adding some example configuration directives to the default
852 configuration file. A little bit of work is required to uncomment the
853 directives and define how usernames and passwords are checked, but
854 there is now a framework to start from.
856 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
857 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
858 without this. I don't know how relevant this is to other LDAP libraries.
860 PH/02 Add the verb name to the "unknown ACL verb" error.
862 PH/03 Magnus Holmgren's patch for filter_prepend_home.
864 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
866 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
867 directory not expanded when it should be if an expanded home directory
868 was set for the address (which is overridden by the transport).
870 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
873 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
874 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
875 because it is too late at that time, and has no effect.
877 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
878 security issue with \' (bugzilla #107). I could not use the
879 PQescapeStringConn() function, because it needs a PGconn value as one of
882 PH/08 When testing addresses using -bt, indicate those final addresses that
883 are duplicates that would not cause an additional delivery. At least one
884 person was confused, thinking that -bt output corresponded to deliveries.
885 (Suppressing duplicates isn't a good idea as you lose the information
886 about possibly different redirections that led to the duplicates.)
888 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
889 systems where poll() doesn't work, in particular OS X.
891 PH/10 Added more information to debugging output for retry time not reached.
893 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
894 operations in malware.c.
896 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
899 PH/13 If write_rejectlog was set false when logging was sent to syslog with
900 syslog_duplication set false, log lines that would normally be written
901 both the the main log and to the reject log were not written to syslog at
904 PH/14 In the default configuration, change the use of "message" in ACL warn
905 statements to "add_header".
907 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
908 not followed by a command (e.g. "seen endif").
910 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
911 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
914 PH/17 Added extra conditions to the default value of delay_warning_condition
918 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
919 { match{$h_precedence:}{(?i)bulk|list|junk} } \
920 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
923 The Auto-Submitted: and various List- headers are standardised, whereas I
924 don't think Precedence: ever was.
926 PH/18 Refactored debugging code in route_finduser() to show more information,
927 in particular, the error code if getpwnam() issues one.
929 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
930 This is apparently needed in addition to the PH/07 change above to avoid
931 any possible encoding problems.
933 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
934 but not after initializing Perl.
936 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
937 output them only if debugging. By default they are written stderr,
938 apparently, which is not desirable.
940 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
943 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
946 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
948 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
949 authenticated or an ident call has been made. Suppress the default
950 values for $authenticated_id and $authenticated_sender (but permit -oMai
951 and -oMas) when testing with -bh.
953 PH/24 Re-jigged the order of the tests in the default configuration so that the
954 tests for valid domains and recipients precede the DNS black list and CSA
955 tests, on the grounds that those ones are more expensive.
957 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
958 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
959 This bug exists in every version of Exim that I still have, right back to
962 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
963 However, an attempt to turn on case-sensitivity in a regex key by
964 including (?-i) didn't work because the subject string was already
965 lowercased, and the effects were non-intuitive. It turns out that a
966 one-line patch can be used to allow (?-i) to work as expected.
972 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
973 other effects) broke the use of negated acl sub-conditions.
975 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
978 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
979 "Deny" causes Exim to reject the incoming connection with a 554 error.
980 Unfortunately, if there is a major crisis, such as a disk failure,
981 tcp-wrappers gives "deny", whereas what one would like would be some
982 kind of temporary error. A kludge has been added to help with this.
983 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
984 554 error is used if errno is still zero or contains ENOENT (which occurs
985 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
988 PH/03 Add -lutil to the default FreeBSD LIBS setting.
990 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
991 errors. Otherwise a message that provokes a temporary error (when other
992 messages do not) can cause a whole host to time out.
994 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
995 addresses were routed directly to files or pipes from a redirect router.
996 File deliveries just didn't batch; pipe deliveries might have suffered
999 PH/06 A failure to get a lock for a hints database would erroneously always say
1000 "Failed to get write lock", even when it was really a read lock.
1002 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1003 of 0600. This has been changed to use the value of the lockfile_mode
1004 option (which defaults to 0600).
1006 PH/08 Applied small patch from the Sieve maintainer.
1008 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1009 folder from quota calculations, a direct delivery into this folder messed
1010 up the contents of the maildirsize file. This was because the regex was
1011 used only to exclude .Trash (or whatever) when the size of the mailbox
1012 was calculated. There was no check that a delivery was happening into an
1013 excluded directory. This bug has been fixed by ignoring all quota
1014 processing for deliveries into excluded directories.
1016 PH/10 Added the maildirfolder_create_regex option to appendfile.
1022 PH/01 The code for finding all the local interface addresses on a FreeBSD
1023 system running IPv6 was broken. This may well have applied to all BSD
1024 systems, as well as to others that have similar system calls. The broken
1025 code found IPv4 interfaces correctly, but gave incorrect values for the
1026 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1027 that it would not match correctly against @[] and not recognize the IPv6
1030 PH/02 The ipliteral router was not recognizing addresses of the form user@
1031 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1033 PH/03 Added disable_ipv6.
1035 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1036 decoded form, because it is most often used to construct To: headers
1037 lines in autoreplies, and the decoded form may well be syntactically
1038 invalid. However, $reply_address has leading white space removed, and all
1039 newlines turned into spaces so that the autoreply transport does not
1042 PH/05 If group was specified without a user on a router, and no group or user
1043 was specified on a transport, the group from the router was ignored.
1045 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1046 for visible compile-time settings that can be used to change these
1047 numbers, for those that want even more. Backwards compatibility with old
1048 spool files has been maintained. However, going back to a previous Exim
1049 release will lost any variables that are in spool files.
1051 PH/07 Two small changes when running in the test harness: increase delay when
1052 passing a TCP/IP connection to a new process, in case the original
1053 process has to generate a bounce, and remove special handling of
1054 127.0.0.2 (sic), which is no longer necessary.
1056 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1057 be the same on different OS.
1059 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1062 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1063 whether --show-vars was specified or not
1065 JJ/02 exipick: Added support for new ACL variable spool format introduced
1068 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1069 syntactically invalid From: or Reply-to: line, and a filter used this to
1070 generate an autoreply, and therefore failed to obtain an address for the
1071 autoreply, Exim could try to deliver to a non-existent relative file
1072 name, causing unrelated and misleading errors. What now happens is that
1073 it logs this as a hard delivery error, but does not attempt to create a
1076 PH/11 The exinext utility has a -C option for testing purposes, but although
1077 the given file was scanned by exinext itself; it wasn't being passed on
1078 when Exim was called.
1080 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1081 an end-of-file indication when reading a command response.
1083 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1084 compiled. In many other places in Exim, IPv6 addresses are always
1085 recognized, so I have changed this. It also means that IPv4 domain
1086 literals of the form [IPV4:n.n.n.n] are now always recognized.
1088 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1089 used if the router is not running as root, for example, when verifying at
1090 ACL time, or when using -bh. The debugging output from this situation was
1091 non-existent - all you got was a failure to exec. I have made two
1094 (a) Failures to set uid/gid, the current directory, or a process leader
1095 in a subprocess such as that created by queryprogram now generate
1096 suitable debugging ouput when -d is set.
1098 (b) The queryprogram router detects when it is not running as root,
1099 outputs suitable debugging information if -d is set, and then runs
1100 the subprocess without attempting to change uid/gid.
1102 PH/15 Minor change to Makefile for building test_host (undocumented testing
1105 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1106 additional section of a DNS packet that returns MX or SRV records.
1107 Instead, it always explicitly searches for A/AAAA records. This avoids
1108 major problems that occur when a DNS server includes only records of one
1109 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1110 fixed another bug: if SRV records were looked up and the corresponding
1111 address records were *not* found in the additional section, the port
1112 values from the SRV records were lost.
1114 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1115 using the correct key (the original address) when searching the retry
1116 rules in order to find which one to use for generating the retry hint.
1118 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1119 adding the default one. Similarly, if it contains a Reply-To: header, the
1120 errors_reply_to option, if set, is not used.
1122 PH/19 When calculating a retry time, Exim used to measure the "time since
1123 failure" by looking at the "first failed" field in the retry record. Now
1124 it does not use this if it is later than than the arrival time of the
1125 message. Instead it uses the arrival time. This makes for better
1126 behaviour in cases where some deliveries succeed, thus re-setting the
1127 "first failed" field. An example is a quota failure for a huge message
1128 when small messages continue to be delivered. Without this change, the
1129 "time since failure" will always be short, possible causing more frequent
1130 delivery attempts for the huge message than are intended.
1131 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1133 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1134 $1, $2, $3) because the numerical variables can be reset during some
1135 expansion items (e.g. "match"), thereby losing the authentication data.
1137 PH/21 Make -bV show the size of off_t variables so that the test suite can
1138 decide whether to run tests for quotas > 2G.
1140 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1141 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1142 filecount value is greater than 2G or if a quota value is greater than 2G
1143 on a system where the size of off_t is not greater than 4, a panic error
1146 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1147 never match. The debug and -bh output now contains an explicit error
1148 message indicating a malformed IPv4 address or mask.
1150 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1151 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1152 PH/23 above applies.
1154 PH/25 Do not write to syslog when running in the test harness. The only
1155 occasion when this arises is a failure to open the main or panic logs
1156 (for which there is an explicit test).
1158 PH/26 Added the /no_tell option to "control=freeze".
1160 PH/27 If a host name lookup failed very early in a connection, for example, if
1161 the IP address matched host_lookup and the reverse lookup yielded a name
1162 that did not have a forward lookup, an error message of the form "no IP
1163 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1164 could be logged. Now it outputs the IP address instead of "NULL".
1166 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1167 allows the sender and the authenticated sender to be set when
1168 submitting a message from within Exim. Since child_open_exim() is
1169 documented for local_scan(), the new function should be too.
1171 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1172 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1173 results in an empty string is now treated as unset.
1175 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1177 PH/31 Added sender_verify_fail logging option.
1179 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1180 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1181 tidied the source and removed it altogether.
1183 PH/33 When a queue run was abandoned because the load average was too high, a
1184 log line was always written; now it is written only if the queue_run log
1185 selector is set. In addition, the log line for abandonment now contains
1186 information about the queue run such as the pid. This is always present
1187 in "start" and "stop" lines but was omitted from the "abandon" line.
1189 PH/34 Omit spaces between a header name and the colon in the error message that
1190 is given when verify = headers_syntax fails (if there are lots of them,
1191 the message gets confusing).
1193 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1194 names, as there are now some PTR records that contain slashes. This check
1195 is only to protect against broken name servers that fall over on strange
1196 characters, so the fact that it applies to all lookups doesn't matter.
1198 PH/36 Now that the new test suite is complete, we can remove some of the
1199 special code in Exim that was needed for the old test suite. For example,
1200 sorting DNS records because real resolvers return them in an arbitrary
1201 order. The new test suite's fake resolver always returns records in the
1204 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1205 bounces) except when queue_only is set, to avoid logging races between
1206 the different processes.
1208 PH/38 Panic-die if .include specifies a non-absolute path.
1210 PH/39 A tweak to the "H" retry rule from its user.
1212 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1213 a label. They prevented compilation on older perls.
1215 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1216 a warning to be raised on newish perls.
1218 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1219 on queue. Changes to match documented behaviour of showing count of
1220 messages matching specified criteria.
1222 PH/40 Changed the default ident timeout from 30s to 5s.
1224 PH/41 Added support for the use of login_cap features, on those BSD systems
1225 that have them, for controlling the resources used by pipe deliveries.
1227 PH/42 The content-scanning code uses fopen() to create files in which to put
1228 message data. Previously it was not paying any attention to the mode of
1229 the files. Exim runs with umask(0) because the rest of the code creates
1230 files with open(), and sets the required mode explicitly. Thus, these
1231 files were ending up world-writeable. This was not a big issue, because,
1232 being within the spool directory, they were not world-accessible. I have
1233 created a function called modefopen, which takes an additional mode
1234 argument. It sets umask(777), creates the file, chmods it to the required
1235 mode, then resets the umask. All the relevant calls to fopen() in the
1236 content scanning code have been changed to use this function.
1238 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1239 to 24 hours. This avoids potential overflow problems when processing G
1240 and H retry rules. I suspect nobody ever tinkers with this value.
1242 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1244 PH/45 When the plaintext authenticator is running as a client, the server's
1245 challenges are checked to ensure they are valid base64 strings. By
1246 default, the authentication attempt is cancelled if an invalid string is
1247 received. Setting client_ignore_invalid_base64 true ignores these errors.
1248 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1249 they are received. Thus, the responses can be made to depend on the
1250 challenges. If an invalid string is ignored, an empty string is placed in
1253 PH/46 Messages that are created by the autoreply transport now contains a
1254 References: header, in accordance with RFCs 2822 and 3834.
1256 PH/47 Added authenticated_sender_force to the smtp transport.
1258 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1260 PH/49 Installed latest patch from the Sieve maintainer.
1262 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1263 was also set, the appendfile transport was unnecessarily scanning a
1264 directory of message files (e.g. for maildir delivery) to find the count
1265 of files (along with the size), even though it did not need this
1266 information. It now does the scan only if it needs to find either the
1267 size of the count of files.
1269 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1271 PH/52 Two bugs concerned with error handling when the smtp transport is
1274 (i) Exim was not creating retry information for temporary errors given
1275 for individual recipients after the DATA command when the smtp transport
1276 was used in LMTP mode. This meant that they could be retried too
1277 frequently, and not timed out correctly.
1279 (ii) Exim was setting the flag that allows error details to be returned
1280 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1281 recipients that were returned after the DATA command.
1283 PH/53 This is related to PH/52, but is more general: for any failing address,
1284 when detailed error information was permitted to be returned to the
1285 sender, but the error was temporary, then after the final timeout, only
1286 "retry timeout exceeded" was returned. Now it returns the full error as
1287 well as "retry timeout exceeded".
1289 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1290 do this, and (what is worse) MTAs that accept it.
1292 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1293 will now be deprecated.
1295 PH/56 New os.c-cygwin from the Cygwin maintainer.
1297 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1298 formats (previously only available in exim formats via -bpr, -bpru,
1299 and -bpra. Now also available in native and exiqgrep formats)
1301 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1302 with very large, slow to parse queues
1304 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1306 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1308 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1309 responses to authentication challenges, though it was showing the
1310 challenges; (ii) I've removed the CR characters from the debug output for
1313 PH/58 Allow for the insertion of a newline as well as a space when a string
1314 is turned into more than one encoded-word during RFC 2047 encoding. The
1315 Sieve code now uses this.
1317 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1318 data_4xx, lost_connection, tls_required.
1320 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1321 message was being sent as an SMTP response.
1323 PH/61 Add -l and -k options to exicyclog.
1325 PH/62 When verifying, if an address was redirected to one new address, so that
1326 verification continued, and the new address failed or deferred after
1327 having set something in $address_data, the value of $address_data was not
1328 passed back to the ACL. This was different to the case when no
1329 redirection occurred. The value is now passed back in both cases.
1331 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1332 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1333 use login_cap.h, so on its own it isn't the distinguishing feature. The
1334 new name refers directly to the setclassresources() function.
1336 PH/65 Added configuration files for NetBSD3.
1338 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1340 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1341 is preferred over IPv4.
1343 PH/68 The bounce_return_message and bounce_return_body options were not being
1344 honoured for bounces generated during the reception of non-SMTP messages.
1345 In particular, this applied to messages rejected by the ACL. This bug has
1346 been fixed. However, if bounce_return_message is true and bounce_return_
1347 body is false, the headers that are returned for a non-SMTP message
1348 include only those that have been read before the error was detected.
1349 (In the case of an ACL rejection, they have all been read.)
1351 PH/69 The HTML version of the specification is now built in a directory called
1352 spec_html instead of spec.html, because the latter looks like a path with
1353 a MIME-type, and this confuses some software.
1355 PH/70 Catch two compiler warnings in sieve.c.
1357 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1358 function verify_get_ident() calls ip_connect() to connect a socket, but
1359 if the "connect()" function timed out, ip_connect() used to close the
1360 socket. However, verify_get_ident() also closes the socket later, and in
1361 between Exim writes to the log, which may get opened at this point. When
1362 the socket was closed in ip_connect(), the log could get the same file
1363 descriptor number as the socket. This naturally causes chaos. The fix is
1364 not to close the socket in ip_connect(); the socket should be closed by
1365 the function that creates it. There was only one place in the code where
1366 this was missing, in the iplookup router, which I don't think anybody now
1367 uses, but I've fixed it anyway.
1369 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1370 well as to direct DNS lookups. Otherwise the handling of names in host
1371 lists is inconsistent and therefore confusing.
1377 PH/01 Two changes to the default runtime configuration:
1379 (1) Move the checks for relay_from_hosts and authenticated clients from
1380 after to before the (commented out) DNS black list checks.
1382 (2) Add control=submission to the relay_from_hosts and authenticated
1383 clients checks, on the grounds that messages accepted by these
1384 statements are most likely to be submissions.
1386 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1388 (1) Generate an error if the third argument for the ${prvs expansion is
1391 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1394 (3) Reset the variables that are obtained from the first argument of
1395 ${prvscheck and used in the second argument before leaving the code,
1396 because their memory is reclaimed, so using them afterwards may do
1399 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1400 one (it's much easier than Tom thought :-).
1402 (5) Because of (4), we can now allow for the use of $prvscheck_result
1403 inside the third argument.
1405 PH/03 For some reason, the default setting of PATH when running a command from
1406 a pipe transport was just "/usr/bin". I have changed it to
1409 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1410 anything to be listed in the output from -bV.
1412 PH/05 When a filter generated an autoreply, the entire To: header line was
1413 quoted in the delivery log line, like this:
1415 => >A.N.Other <ano@some.domain> <original@ddress> ...
1417 This has been changed so that it extracts the operative address. There
1418 may be more than one such address. If so, they are comma-separated, like
1421 => >ano@some.domain,ona@other.domain <original@ddress> ...
1423 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1424 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1425 not being looked up in the rDNS to get a host name, Exim was showing the
1426 IP address twice in Received: lines, even though the IP addresses were
1427 identical. For example:
1429 Received: from [1.2.3.4] (helo=[1.2.3.4])
1431 However, if the real host name was known, it was omitting the HELO data
1432 if it matched the actual IP address. This has been tidied up so that it
1433 doesn't show the same IP address twice.
1435 PH/07 When both +timestamp and +memory debugging was on, the value given by
1436 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1437 called by the debug printing, thereby overwriting the timestamp buffer.
1438 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1440 PH/08 When the original message was included in an autoreply transport, it
1441 always said "this is a copy of the message, including all the headers",
1442 even if body_only or headers_only was set. It now gives an appropriate
1445 PH/09 Applied a patch from the Sieve maintainer which:
1447 o fixes some comments
1448 o adds the (disabled) notify extension core
1449 o adds some debug output for the result of if/elsif tests
1450 o points to the current vacation draft in the documentation
1451 and documents the missing references header update
1455 o fixes a bug in processing the envelope test (when testing
1456 multiple envelope elements, the last element determinted the
1459 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1460 Electronic Mail") by including:
1462 Auto-submitted: auto-generated
1464 in the messages that it generates (bounce messages and others, such as
1465 warnings). In the case of bounce messages for non-SMTP mesages, there was
1466 also a typo: it was using "Auto_submitted" (underscore instead of
1467 hyphen). Since every message generated by Exim is necessarily in response
1468 to another message, thes have all been changed to:
1470 Auto-Submitted: auto-replied
1472 in accordance with these statements in the RFC:
1474 The auto-replied keyword:
1476 - SHOULD be used on messages sent in direct response to another
1477 message by an automatic process,
1479 - MUST NOT be used on manually-generated messages,
1481 - MAY be used on Delivery Status Notifications (DSNs) and Message
1482 Disposition Notifications (MDNs),
1484 - MUST NOT be used on messages generated by automatic or periodic
1485 processes, except for messages which are automatic responses to
1488 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1489 to the default Received: header definition.
1491 PH/12 Added log selector acl_warn_skipped (default on).
1493 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1494 variables because (a) they are in the wrong storage pool and (b) even if
1495 they were copied, it wouldn't work properly because of the caching.
1497 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1498 checking when decoding. Apparently there are clients that generate
1499 overlong encoded strings. Why am I not surprised?
1501 PH/15 If the first argument of "${if match_address" was not empty, but did not
1502 contain an "@" character, Exim crashed. Now it writes a panic log message
1503 and treats the condition as false.
1505 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1507 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1508 extension "envelope-auth". The code is finished and in agreement with
1509 other implementations, but there is no documentation so far and in fact,
1510 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1511 not changing the active code.
1513 Print executed "if" and "elsif" statements when debugging is used. This
1514 helps a great deal to understand what a filter does.
1516 Document more things not specified clearly in RFC3028. I had all this
1517 sorted out, when out of a sudden new issues came to my mind. Oops."
1519 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1522 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1523 canonical form (as documented). However, after a host name lookup from
1524 the IP address, check_host() was doing a simple string comparison with
1525 addresses acquired from the DNS when checking that the found name did
1526 have the original IP as one of its addresses. Since any found IPv6
1527 addresses are likely to be in abbreviated form, the comparison could
1528 fail. Luckily, there already exists a function for doing the comparison
1529 by converting both addresses to binary, so now that is used instead of
1530 the text comparison.
1532 PH/20 There was another similar case to PH/19, when a complete host name was
1533 given in a host list; looking up its IP address could give an abbreviated
1534 form, whereas the current host's name might or might not be abbreviated.
1535 The same fix has been applied.
1541 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1542 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1545 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1546 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1548 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1550 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1551 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1552 Also, for "Auto-Submitted", treat anything other than "no" as
1553 non-personal, in accordance with RFC 3834. (Previously it treated
1554 anything starting "auto-" as non-personal.)
1556 TF/01 The control=submission/name=... option had a problem with syntax
1557 errors if the name included a slash character. The /name= option
1558 now slurps the rest of the string, so it can include any characters
1559 but it must come last in the list of options (after /sender_retain
1562 PH/05 Some modifications to the interface to the fake nameserver for the new
1570 TK/01 Added the "success_on_redirect" address verification option. See
1571 NewStuff for rationale and an example.
1573 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1575 PH/02 Patch to exigrep to allow it to work on syslog lines.
1577 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1578 fread() to skip over the body file's header line, because in Cygwin the
1579 header line is locked and is inaccessible.
1581 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1582 co-exist for some time) to make it clear that it is the Exim ID that is
1583 referenced, not the Message-ID: header line.
1585 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1586 string_format() function, because snprintf() does not exist on all
1589 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1590 ${stat: expansion item.
1592 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1593 protocol synchronization error", to keep the pedants happy.
1595 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1596 well as for IRIX systems, when gcc is being used. See the host.c source
1599 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1601 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1604 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1605 transport and to the smtp transport in LMTP mode.
1607 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1609 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1610 run a filter in a subprocess. This could lead to confusion in subsequent
1611 lookups in the parent process. There should also be a search_tidyup() at
1612 the end of the subprocess.
1614 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1615 only if the host matched helo_try_verify_hosts, which caused the
1616 verification to occur when the EHLO/HELO command was issued. The ACL just
1617 tested the remembered result. Now, if a previous verification attempt has
1618 not happened, "verify = helo" does it there and then.
1620 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1622 TK/03 Fix log output including CR from clamd.
1624 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1625 exist provoked a memory error which could cause a segfault.
1627 PH/15 Installed PCRE 6.2
1629 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1631 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1632 of the problem. Specifically, suggested +O2 rather than +O1 for the
1635 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1637 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1638 record, the port was not correctly logged when the outgoing_port log
1639 selector was set (it logged the transort's default port).
1641 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1642 fallback_hosts, and "hosts" in the smtp transport.
1644 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1645 host errors such as "Connection refused".
1647 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1648 authentication with radiusclient 0.4.9:
1650 - Error returned from rc_read_config was caught wrongly
1651 - Username/password not passed on to radius server due to wrong length.
1653 The presumption is that some radiusclient API changes for 4.51/PH/17
1654 were not taken care of correctly. The code is still untested by me (my
1655 Linux distribution still has 0.3.2 of radiusclient), but it was
1656 contributed by a Radius user.
1658 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1659 expanding the "port" option of the smtp transport.
1661 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1662 while reading a MIME header. Thanks to Tom Hughes for a patch.
1664 PH/24 Include config.h inside local_scan.h so that configuration settings are
1667 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1668 that in an ACL for RCPT (for example), you can examine exactly what was
1671 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1672 commands, but it was not correctly comparing the address with the actual
1673 client host address. Thus, it would show the EHLO address in Received:
1674 header lines when this was not necessary.
1676 PH/27 Added the % operator to ${eval:}.
1678 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1679 it should be ignoring failures (because with -C, for example, it has lost
1680 privilege). It wasn't ignoring creation failures other than "already
1683 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1686 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1687 because an input line was too long, either on its own, or by virtue of
1688 too many continuations, the temporary file was not being removed, and the
1689 return code was incorrect.
1691 PH/31 Missing "BOOL" in function definition in filtertest.c.
1693 PH/32 Applied Sieve patches from the maintainer.
1695 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1697 PH/33 Added "verify = not_blind".
1699 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1700 Local/Makefile (with some defaults set). These are used in built scripts
1701 such as exicyclog, but they have never been used in the exim_install
1702 script (though there are many overriding facilities there). I have
1703 arranged that the exim_install script now takes note of these two
1706 PH/35 Installed configuration files for Dragonfly.
1708 PH/36 When a locally submitted message by a trusted user did not contain a
1709 From: header, and the sender address was obtained from -f or from an SMTP
1710 MAIL command, and the trusted user did not use -F to supply a sender
1711 name, $originator_name was incorrectly used when constructing a From:
1712 header. Furthermore, $originator_name was used for submission mode
1713 messages from external hosts without From: headers in a similar way,
1714 which is clearly wrong.
1716 PH/37 Added control=suppress_local_fixups.
1718 PH/38 When log_selector = +received_sender was set, and the addition of the
1719 sender made the log line's construction buffer exactly full, or one byte
1720 less than full, an overflow happened when the terminating "\n" was
1723 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1724 when the result of a list match is failure because a DNS lookup failed.
1726 PH/40 RM_COMMAND is now used in the building process.
1728 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1729 the "build-* directories that it finds.
1731 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1732 address in a domain literal was a prefix of an interface address.
1734 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1735 when verifying a sender address, unless rewrite_headers is false.
1737 PH/44 Wrote a long comment about why errors_to addresses are verified as
1738 recipients, not senders.
1740 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1741 the ratelimit ACL was added.
1743 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1745 PH/46 Added extra information about PostgreSQL errors to the error string.
1747 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1748 avoiding the need to install special zones in a real server. This is
1749 backwards compatible; if it can't find the fake resolver, it drops back.
1750 Thus, both old and new test suites can be run.
1752 TF/02 Added util/ratelimit.pl
1754 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1755 clock is set back in time.
1757 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1758 Candler <B.Candler@pobox.com>.
1760 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1761 OK for addresses that are the result of redirections.
1763 PH/48 A number of further additions for the benefit of the new test suite,
1764 including a fake gethostbyname() that interfaces to the fake DNS resolver
1767 TF/06 The fix for widen_domains has also been applied to qualify_single and
1768 search_parents which are the other dnslookup options that can cause
1771 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1774 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1776 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1777 service" if it encountered an MX record with an empty target hostname.
1778 The message is now "an MX or SRV record indicated no SMTP service".
1780 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1781 if the DNS of the sending site is misconfigured. This is quite a
1782 common situation. This change restores the behaviour of treating a
1783 helo verification defer as a failure.
1785 PH/51 If self=fail was set on a router, the bounce message did not include the
1786 actual error message.
1792 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1794 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1795 command itself ended in error, the underlying message about the transport
1796 filter timeout was being overwritten with the pipe command error. Now the
1797 underlying error message should be appended to the second error message.
1799 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1801 PH/02 Reduce the amount of output that "make" produces by default. Full output
1802 can still be requested.
1804 PH/03 The warning log line about a condition test deferring for a "warn" verb
1805 was being output only once per connection, rather than after each
1806 occurrence (because it was using the same function as for successful
1807 "warn" verbs). This seems wrong, so I have changed it.
1809 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1810 it should not have, which might have caused a crash in the right
1811 circumstances, but probably never did.
1813 PH/04 Installed a modified version of Tony Finch's patch to make submission
1814 mode fix the return path as well as the Sender: header line, and to
1815 add a /name= option so that you can make the user's friendly name appear
1818 TF/03 Added the control = fakedefer ACL modifier.
1820 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1821 Mark Lowes for thorough testing.
1823 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1825 TK/03 Merged latest SRS patch from Miles Wilton.
1827 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1828 with the definition in sysexits.h (which is #included earlier).
1829 Fortunately, Exim does not actually use EX_OK. The code used to try to
1830 preserve the sysexits.h value, by assumimg that macro definitions were
1831 scanned for macro replacements. I have been disabused of this notion,
1832 so now the code just undefines EX_OK before #including unistd.h.
1834 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1835 in the smtp transport. When a block could not be written in a single
1836 write() function, the timeout was being re-applied to each part-write.
1837 This seems wrong - if the receiver was accepting one byte at a time it
1838 would take for ever. The timeout is now adjusted when this happens. It
1839 doesn't have to be particularly precise.
1841 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1842 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1844 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1845 without a domain if the check to <postmaster@domain> fails.
1847 SC/01 Eximstats: added -xls and the ability to specify output files
1848 (patch written by Frank Heydlauf).
1850 SC/02 Eximstats: use FileHandles for outputing results.
1852 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1854 SC/04 Eximstats: fixed display of large numbers with -nvr option
1856 SC/05 Eximstats: fixed merging of reports with empty tables.
1858 SC/06 Eximstats: added the -include_original_destination flag
1860 SC/07 Eximstats: removed tabs and trailing whitespace.
1862 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1864 TK/06 MBOX spool code: Add real "From " MBOX separator line
1865 so the .eml file is really in mbox format (even though
1866 most programs do not really care). Patch from Alex Miller.
1868 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1869 The latter is generated from $received_to and is only set if the
1870 message has one envelope recipient. SA can use these headers,
1871 obviously out-of-the-box. Patch from Alex Miller.
1873 PH/08 The ${def test on a variable was returning false if the variable's
1874 value was "0", contrary to what the specification has always said!
1875 The result should be true unless the variable is empty.
1877 PH/09 The syntax error of a character other than { following "${if
1878 def:variable_name" (after optional whitespace) was not being diagnosed.
1879 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1880 accidental colon was present, for example, could give incorrect results.
1882 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1883 result is used (not including appendfile, where other changes are about
1886 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1887 This involved changing a lot of size variables from int to off_t. It
1888 should work with maildirs and everything.
1890 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1891 spamd dying while we are connected to it.
1893 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1896 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1897 function for BDB 4.3.
1899 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1901 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1902 the sources for PCRE have been re-arranged and more files are now
1905 PH/15 The code I had for printing potentially long long variables in PH/11
1906 above was not the best (it lost precision). The length of off_t variables
1907 is now inspected at build time, and an appropriate printing format (%ld
1908 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1909 to be "long long int" or "long int". This is needed for the internal
1910 formatting function string_vformat().
1912 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1913 the configuration file to be ":syslog", then the script "guesses" where
1914 the logs files are, rather than using the compiled in default. In our
1915 case the guess is not the same as the compiled default, so the script
1916 suddenly stopped working when I started to use syslog. The patch checks
1917 to see if log_file_path is "". If so, it attempts to read it from exim
1918 with no configuration file to get the compiled in version, before it
1919 falls back to the previous guessing code."
1921 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1922 implementing BATV in an Exim configuration. See NewStuff for the gory
1925 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1926 Makefile that are specific to HP-UX.
1928 PH/18 If the "use_postmaster" option was set for a recipient callout together
1929 with the "random" option, the postmaster address was used as the MAIL
1930 FROM address for the random test, but not for the subsequent recipient
1931 test. It is now used for both.
1933 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1934 patch removes a few documentation additions to RFC 3028, because the
1935 latest draft now contains them. It adds the new en;ascii-case comparator
1936 and a new error check for 8bit text in MIME parts. Comparator and
1937 require names are now matched exactly. I enabled the subaddress
1938 extension, but it is not well tested yet (read: it works for me)."
1940 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1941 rework some of the code of TK/09 above to avoid the hardwired use of
1942 "%lld" and "long long". Replaced the call to snprintf() with a call to
1945 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1946 records point to non-existent hosts", "retry timeout exceeded", and
1947 "retry time not reached for any host after a long failure period".
1949 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1950 experimental DomainKeys support:
1952 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1953 (2) On an error such as an illegally used "control", the wrong name for
1954 the control was given.
1956 These problems did NOT occur unless DomainKeys support was compiled.
1958 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1960 PH/24 Added ${if match_ip condition.
1962 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1963 databases so that it will be absolutely obvious if a crash occurs in the
1964 DB library. This is a regular occurrence (often caused by mis-matched
1967 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1968 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1969 picked up on a user's system that detects such things. There doesn't seem
1970 to be a gcc warning option for this - only an attribute that has to be
1971 put on the function's prototype. It seems that in Fedora Core 4 they have
1972 set this on a number of new functions. No doubt there will be more in due
1975 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1976 specify a transport. However, if an address that was verified by such a
1977 router was the subject of a callout, Exim crashed because it tried to
1978 read the rcpt_include_affixes from the non-existent transport. Now it
1979 just assumes that the setting of that option is false. This bug was
1980 introduced by 4.51/PH/31.
1982 PH/28 Changed -d+all to exclude +memory, because that information is very
1983 rarely of interest, but it makes the output a lot bigger. People tend to
1984 do -d+all out of habit.
1986 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1987 code in os-type was giving problems when libc.so lives in lib64, like on
1990 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1991 aren't the modern standard, and it seems that some systems' include files
1992 don't always have them. Exim was already checking for some of the newer
1993 ones like T_AAAA, and defining it itself. I've added checks for all the
1994 record types that Exim uses.
1996 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1997 not automatically generating a new one, as it is supposed to. This
1998 prevented TLS from working. If the file did exist, but contained invalid
1999 data, a new version was generated, as expected. It was only the case of a
2000 non-existent file that was broken.
2002 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2003 with a change in libdomainkeys > 0.64.
2005 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2006 from DNS. If the selector record carries the flag, it now has
2007 precedence over the domain-wide flag.
2009 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2011 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2012 the use of a port name that isn't defined in /etc/services) occurred, the
2013 message was deferred as in a normal delivery, and thus remained on the
2014 spool, instead of being failed because of the mua_wrapper setting. This
2015 is now fixed, and I tidied up some of the mua_wrapper messages at the
2018 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2019 the messages in a hash of arrays rather than using individual hashes.
2020 This is a bit cleaner and results in dramatic memory savings, albeit
2021 at a slight CPU cost.
2023 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2024 as requested by Marc Sherman.
2026 SC/10 Eximstats: added histograms for user specified patterns as requested
2029 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2031 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2032 fopen() in the content-scanning modules that did not already have it.
2038 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2039 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2041 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2043 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2045 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2046 to test Sieve filters that use "vacation".
2048 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2049 that changes the way the GnuTLS parameters are stored in the cache file.
2050 The new format can be generated externally. For backward compatibility,
2051 if the data in the cache doesn't make sense, Exim assumes it has read an
2052 old-format file, and it generates new data and writes a new file. This
2053 means that you can't go back to an older release without removing the
2056 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2057 work if there are any delivery delays because "one_time" forces the
2058 parent to be marked "delivered", so its unseen clone is never tried
2059 again. For this reason, Exim now forbids the simultaneous setting of
2062 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2063 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2064 entry if you want to know the details. The fix, however, neglected to
2065 consider the case when local delivery batching is involved. The test for
2066 "previously delivered" was not happening when checking to see if an
2067 address could be batched with a previous (undelivered) one; under
2068 certain circumstances this could lead to multiple deliveries to the same
2071 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2072 in its include files, and this causes problems building Exim.
2074 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2075 header_syntax) but Exim was just ignoring anything given after a slash.
2076 In particular, this caused confusion with an attempt to use "verify =
2077 reverse_host_lookup/defer_ok". An error is now given when options are
2078 supplied for verify items that do not have them. (Maybe reverse_host_
2079 lookup should have a defer_ok option, but that's a different point.)
2081 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2082 defined by RFC 821) to 2048, because there were problems with some AUTH
2083 commands, and RFC 1869 says the size should be increased for extended
2084 SMTP commands that take arguments.
2086 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2089 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2090 "unknown" error; now it says that the functionality isn't in the binary.
2092 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2093 an address' error message when a string expansion fails (syntax or
2094 whatever). Otherwise the password may appear in the log. Following change
2095 PH/42 below, there is no longer a chance of it appearing in a bounce
2098 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2100 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2101 of its addresses was ever tried. (Bugzilla bug #2.)
2103 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2104 the result incorrectly in the debug output. (It correctly added a newline
2105 to what was transported.)
2107 TF/01 Added $received_time.
2109 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2110 commented out examples of how to interface to a virus scanner and to
2111 SpamAssassin. Also added commented examples of av_scanner and
2112 spamd_address settings.
2114 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2115 and controls are allowed in which ACLs. There were a couple of minor
2116 errors. Some of the entries in the conditions table (which is a table of
2117 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2118 negation of where the condition IS allowed.
2120 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2122 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2123 header file does not have a version number, so I've had to invent a new
2124 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2125 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2126 radiusclient), but it was contributed by a Radius user.
2128 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2129 files or directories, for OpenSSL.
2131 PH/19 When an Exim process that is running as root has to create an Exim log
2132 file, it does so in a subprocess that runs as exim:exim so as to get the
2133 ownership right at creation (otherwise, other Exim processes might see
2134 the file with the wrong ownership). There was no test for failure of this
2135 fork() call, which would lead to the process getting stuck as it waited
2136 for a non-existent subprocess. Forks do occasionally fail when resources
2137 run out. I reviewed all the other calls to fork(); they all seem to check
2140 PH/20 When checking for unexpected SMTP input at connect time (before writing
2141 the banner), Exim was not dealing correctly with a non-positive return
2142 from the read() function. If the client had disconnected by this time,
2143 the result was a log entry for a synchronization error with an empty
2144 string after "input=" when read() returned zero. If read() returned -1
2145 (an event I could not check), uninitialized data bytes were printed.
2146 There were reports of junk text (parts of files, etc) appearing after
2149 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2151 PH/22 Added support for macro redefinition, and (re)definition in between
2152 driver and ACL definitions.
2154 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2155 forgetting to use the resulting value; it was using the unexpanded value.
2157 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2158 hadn't been configured. The fix is from Juergen Kreileder, who
2159 understands it better than I do:
2161 "Here's what I see happening with three configured cyrus_sasl
2162 authenticators configured (plain, login, cram-md5):
2164 On startup auth_cyrus_sasl_init() gets called for each of these.
2165 This means three calls to sasl_listmech() without a specified mech_list.
2166 => SASL tests which mechs of all available mechs actually work
2167 => three warnings about OTP not working
2168 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2170 With the patch, sasl_listmech() also gets called three times. But now
2171 SASL's mech_list option is set to the server_mech specified in the the
2172 authenticator. Or in other words, the answer from sasl_listmech()
2173 gets limited to just the mech you're testing for (which is different
2175 => the return list contains just 'plain' or 'login', 'cram-md5' or
2176 nothing depending on the value of ob->server_mech.
2178 I've just tested the patch: Authentication still works fine,
2179 unavailable mechs specified in the exim configuration are still
2180 caught, and the auth.log warnings about OTP are gone."
2182 PH/25 When debugging is enabled, the contents of the command line are added
2183 to the debugging output, even when log_selector=+arguments is not
2186 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2187 answer is "GNU", and only if the return is "GNU/something" is the answer
2190 PH/27 $acl_verify_message is now set immediately after the failure of a
2191 verification in an ACL, and so is available in subsequent modifiers. In
2192 particular, the message can be preserved by coding like this:
2194 warn !verify = sender
2195 set acl_m0 = $acl_verify_message
2197 Previously, $acl_verify_message was set only while expanding "message"
2198 and "log_message" when a very denied access.
2200 PH/28 Modified OS/os.c-Linux with
2202 -#ifndef OS_LOAD_AVERAGE
2203 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2205 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2206 nomenclature these days.)
2208 PH/29 Installed patch from the Sieve maintainer that adds the options
2209 sieve_useraddress and sieve_subaddress to the redirect router.
2211 PH/30 In these circumstances:
2212 . Two addresses routed to the same list of hosts;
2213 . First host does not offer TLS;
2214 . First host accepts first address;
2215 . First host gives temporary error to second address;
2216 . Second host offers TLS and a TLS session is established;
2217 . Second host accepts second address.
2218 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2219 and peerdn, if requested) that were in fact used only for the second
2222 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2223 attention to any local part prefix or suffix that was matched by the
2224 router that accepted the address. It now behaves in the same way as it
2225 does for delivery: the affixes are removed from the local part unless
2226 rcpt_include_affixes is set on the transport.
2228 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2229 timeout during the DATA phase of an incoming message.
2231 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2232 applied a patch sanctioned by the Sieve maintainer.
2234 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2235 the uid or gid is negative. A case of a negative gid caused this to be
2236 noticed. The fix allows for either to be negative.
2238 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2239 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2242 PH/36 The change PH/12 above was broken. Fixed it.
2244 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2245 the grounds that routing the same address twice would always produce the
2246 same answer. This might have been true once, but it is certainly no
2247 longer true now. Routing a child address may depend on the previous
2248 routing that produced that child. Some complicated redirection strategies
2249 went wrong when messages had multiple recipients, and made Exim's
2250 behaviour dependent on the order in which the addresses were given.
2252 I have moved the duplicate checking until after the routing is complete.
2253 Exim scans the addresses that are assigned to local and remote
2254 transports, and removes any duplicates. This means that more work will be
2255 done, as duplicates will always all be routed, but duplicates are
2256 presumably rare, so I don't expect this is of any significance.
2258 For deliveries to pipes, files, and autoreplies, the duplicate checking
2259 still happens during the routing process, since they are not going to be
2262 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2263 It corrects a timeout issue with spamd. This is Ian's comment: "The
2264 background is that sometimes spamd either never reads data from a
2265 connection it has accepted, or it never writes response data. The exiscan
2266 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2267 blindly assumes that writes won't block so it may never time out."
2269 PH/39 Allow G after quota size as well as K and M.
2271 PH/40 The value set for $authenticated_id in an authenticator may not contain
2272 binary zeroes or newlines because the value is written to log lines and
2273 to spool files. There was no check on this. Now the value is run through
2274 the string_printing() function so that such characters are converted to
2275 printable escape sequences.
2277 PH/41 $message_linecount is a new variable that contains the total number of
2278 lines in the message. Compare $body_linecount, which is the count for the
2281 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2282 bounce and delay warning messages, except in certain special cases, which
2285 (a) An SMTP error message from a remote host;
2286 (b) A message specified in a :fail: redirection;
2287 (c) A message specified in a "fail" command in a system filter;
2288 (d) A message specified in a FAIL return from the queryprogram router;
2289 (e) A message specified by the cannot_route_message router option.
2291 In these cases only, Exim does include the error details in bounce and
2292 warning messages. There are also a few cases where bland messages such
2293 as "unrouteable address" or "local delivery error" are given.
2295 PH/43 $value is now also set for the "else" part of a ${run expansion.
2297 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2298 being worked on, but at least Exim now implements the latest version to
2301 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2302 process to complete was treated as a delivery failure, a timeout while
2303 writing the message to the pipe was logged, but erroneously treated as a
2304 successful delivery. Such timeouts include transport filter timeouts. For
2305 consistency with the overall process timeout, these timeouts are now
2306 treated as errors, giving rise to delivery failures by default. However,
2307 there is now a new Boolean option for the pipe transport called
2308 timeout_defer, which, if set TRUE, converts the failures into defers for
2309 both kinds of timeout. A transport filter timeout is now identified in
2312 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2313 systems where "make" and "gmake" are different, calling "gmake" at top
2314 level broke things. I've arranged for the value of $(MAKE) to be passed
2315 from the Makefile to this script so that it can call the same version of
2319 A note about Exim versions 4.44 and 4.50
2320 ----------------------------------------
2322 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2323 changes of various kinds. As a consequence, a big documentation update was
2324 needed. This delayed the release for rather longer than seemed good, especially
2325 in the light of a couple of (minor) security issues. Therefore, the changes
2326 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2327 release. So 4.44 and 4.50 are in effect two different branches that both start
2330 I have left the 4.50 change log unchanged; it contains all the changes since
2331 4.43. The change log for 4.44 is below; many of its items are identical to
2332 those for 4.50. This seems to be the most sensible way to preserve the
2333 historical information.
2339 1. Minor wording change to the doc/README.SIEVE file.
2341 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2342 computation of the current number of files was incorrect.
2344 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2345 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2348 4. Give more explanation in the error message when the command for a transport
2349 filter fails to execute.
2351 5. There are several places where Exim runs a non-Exim command in a
2352 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2353 was being done only for the command run by the queryprogram router. It is
2354 now done for all such subprocesses. The other cases are: ${run, transport
2355 filters, and the commands run by the lmtp and pipe transports.
2357 6. Added CONFIGURE_GROUP build-time option.
2359 7. Some older OS have a limit of 256 on the maximum number of file
2360 descriptors. Exim was using setrlimit() to set 1000 as a large value
2361 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2362 systems. I've change it so that if it can't get 1000, it tries for 256.
2364 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2365 was an oversight, and furthermore, ever since the addition of extra
2366 controls (e.g. 4.43/32), the checks on when to allow different forms of
2367 "control" were broken. There should now be diagnostics for all cases when a
2368 control that does not make sense is encountered.
2370 9. Added the /retain_sender option to "control=submission".
2372 10. $recipients is now available in the predata ACL (oversight).
2374 11. Tidy the search cache before the fork to do a delivery from a message
2375 received from the command line. Otherwise the child will trigger a lookup
2376 failure and thereby defer the delivery if it tries to use (for example) a
2377 cached ldap connection that the parent has called unbind on.
2379 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2380 of $address_data from the recipient verification was clobbered by the
2381 sender verification.
2383 13. The value of address_data from a sender verification is now available in
2384 $sender_address_data in subsequent conditions in the ACL statement.
2386 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2388 15. Added a new option "connect=<time>" to callout options, to set a different
2391 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2392 was its contents. (It was OK if the option was not defined at all.)
2394 17. A "Completed" log line is now written for messages that are removed from
2395 the spool by the -Mrm option.
2397 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2398 information about exactly what failed.
2400 19. Added -dd to debug only the daemon process.
2402 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2403 handles timeouts, both on the server side and network timeouts. Renamed the
2404 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2406 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2407 It is now set to "smtps".
2409 22. $host_address is now set to the target address during the checking of
2410 ignore_target_hosts.
2412 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2413 being passed; this would have caused $sender_host_name to have been used if
2414 matching the list had actually called for a host name (not very likely,
2415 since this list is usually IP addresses). A host name is now passed as
2418 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2419 code that specifies a non-restarting handler (typically sigaction() in
2420 modern systems) in an attempt to fix a rare and obscure crash bug.
2422 25. Narrowed the window for a race in the daemon that could cause it to ignore
2423 SIGCHLD signals. This is not a major problem, because they are used only to
2424 wake it up if nothing else does.
2426 26. A malformed maildirsize file could cause Exim to calculate negative values
2427 for the mailbox size or file count. Odd effects could occur as a result.
2428 The maildirsize information is now recalculated if the size or filecount
2431 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2432 support for a long time. Removed HAVE_SYS_VFS_H.
2434 28. Installed the latest version of exipick from John Jetmore.
2436 29. In an address list, if the pattern was not a regular expression, an empty
2437 subject address (from a bounce message) matched only if the pattern was an
2438 empty string. Non-empty patterns were not even tested. This was the wrong
2439 because it is perfectly reasonable to use an empty address as part of a
2440 database query. An empty address is now tested by patterns that are
2441 lookups. However, all the other forms of pattern expect the subject to
2442 contain a local part and a domain, and therefore, for them, an empty
2443 address still always fails if the pattern is not itself empty.
2445 30. Exim went into a mad DNS loop when attempting to do a callout where the
2446 host was specified on an smtp transport, and looking it up yielded more
2447 than one IP address.
2449 31. Re-factored the code for checking spool and log partition space into a
2450 function that finds that data and another that does the check. The former
2451 is then used to implement four new variables: $spool_space, $log_space,
2452 $spool_inodes, and $log_inodes.
2454 32. The RFC2047 encoding function was originally intended for short strings
2455 such as real names; it was not keeping to the 75-character limit for
2456 encoded words that the RFC imposes. It now respects the limit, and
2457 generates multiple encoded words if necessary. To be on the safe side, I
2458 have increased the buffer size for the ${rfc2047: expansion operator from
2461 33. It is now permitted to omit both strings after an "if" condition; if the
2462 condition is true, the result is "true". As before, when the second string
2463 is omitted, a false condition yields an empty string. This makes it less
2464 cumbersome to write custom ACL and router conditions.
2466 34. Failure to deliver a bounce message always caused it to be frozen, even if
2467 there was an errors_to setting on the router. The errors_to setting is now
2470 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2471 canonical form (fully expanded) before being placed in
2472 $sender_host_address.
2474 36. The table in the code that translates DNS record types into text (T_A to
2475 "A" for instance) was missing entries for NS and CNAME. It is just possible
2476 that this could have caused confusion if both these types were looked up
2477 for the same domain, because the text type is used as part of Exim's
2478 per-process caching. But the chance of anyone hitting this buglet seems
2481 37. The dnsdb lookup has been extended in a number of ways.
2483 (1) There is a new type, "zns", which walks up the domain tree until it
2484 finds some nameserver records. It should be used with care.
2486 (2) There is a new type, "mxh", which is like "mx" except that it returns
2487 just the host names, not the priorities.
2489 (3) It is now possible to give a list of domains (or IP addresses) to be
2490 looked up. The behaviour when one of the lookups defers can be
2491 controlled by a keyword.
2493 (4) It is now possible to specify the separator character for use when
2494 multiple records are returned.
2496 38. The dnslists ACL condition has been extended: it is now possible to supply
2497 a list of IP addresses and/or domains to be looked up in a particular DNS
2500 39. Added log_selector=+queue_time_overall.
2502 40. When running the queue in the test harness, wait just a tad after forking a
2503 delivery process, to get repeatability of debugging output.
2505 41. Include certificate and key file names in error message when GnuTLS fails
2506 to set them up, because the GnuTLS error message doesn't include the name
2507 of the failing file when there is a problem reading it.
2509 42. Allow both -bf and -bF in the same test run.
2511 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2513 44. The "Exiscan patch" is now merged into the mainline Exim source.
2515 45. Sometimes the final signoff response after QUIT could fail to get
2516 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2517 < 0 before doing a fflush(). This bug looks as though it goes back to the
2518 introduction of TLS in release 3.20, but "sometimes" must have been rare
2519 because the tests only now provoked it.
2521 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2522 (this can affect the format of dates).
2524 47. exim_tidydb, when checking for the continued existence of a message for
2525 which it has found a message-specific retry record, was not finding
2526 messages that were in split spool directories. Consequently, it was
2527 deleting retry records that should have stayed in existence.
2529 48. Steve fixed some bugs in eximstats.
2531 49. The SPA authentication driver was not abandoning authentication and moving
2532 on to the next authenticator when an expansion was forced to fail,
2533 contradicting the general specification for all authenticators. Instead it
2534 was generating a temporary error. It now behaves as specified.
2536 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2537 (the order specifies the preference for clients). The order is now AES256,
2538 AES128, 3DES, ARCFOUR128.
2540 51. Small patch to Sieve code - explicitly set From: when generating an
2543 52. Exim crashed if a remote delivery caused a very long error message to be
2544 recorded - for instance if somebody sent an entire SpamAssassin report back
2545 as a large number of 550 error lines. This bug was coincidentally fixed by
2546 increasing the size of one of Exim's internal buffers (big_buffer) that
2547 happened as part of the Exiscan merge. However, to be on the safe side, I
2548 have made the code more robust (and fixed the comments that describe what
2551 53. Now that there can be additional text after "Completed" in log lines (if
2552 the queue_time_overall log selector is set), a one-byte patch to exigrep
2553 was needed to allow it to recognize "Completed" as not the last thing in
2556 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2557 patch that reportedly fixes this has been added. I am not expert enough to
2558 create a test for it. This is what the patch creator wrote:
2560 "I found a little strange behaviour of ldap code when working with
2561 Windows 2003 AD Domain, where users was placed in more than one
2562 Organization Units. When I tried to give exim partial DN, the exit code
2563 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2564 But simultaneously result of request was absolutely normal ldap result,
2565 so I produce this patch..."
2567 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2568 REFERENCE, so I have modified the code to exclude the patch when that macro
2571 55. Some experimental protocols are using DNS PTR records for new purposes. The
2572 keys for these records are domain names, not reversed IP addresses. The
2573 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2574 leaves it alone. Component reversal etc. now happens only for IP addresses.
2577 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2579 57. Double the size of the debug message buffer (to 2048) so that more of very
2580 long debug lines gets shown.
2582 58. The exicyclog utility now does better if the number of log files to keep
2583 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2585 59. Two changes related to the smtp_active_hostname option:
2587 (1) $smtp_active_hostname is now available as a variable.
2588 (2) The default for smtp_banner uses $smtp_active_hostname instead
2589 of $primary_hostname.
2591 60. The host_aton() function is supposed to be passed a string that is known
2592 to be a valid IP address. However, in the case of IPv6 addresses, it was
2593 not checking this. This is a hostage to fortune. Exim now panics and dies
2594 if the condition is not met. A case was found where this could be provoked
2595 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2596 components; fortuitously, this particular loophole had already been fixed
2597 by change 4.50/55 above.
2599 If there are any other similar loopholes, the new check in host_aton()
2600 itself should stop them being exploited. The report I received stated that
2601 data on the command line could provoke the exploit when Exim was running as
2602 exim, but did not say which command line option was involved. All I could
2603 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2604 running as the user.
2607 61. There was a buffer overflow vulnerability in the SPA authentication code
2608 (which came originally from the Samba project). I have added a test to the
2609 spa_base64_to_bits() function which I hope fixes it.
2612 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2613 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2615 63. The daemon start-up calls getloadavg() while still root for those OS that
2616 need the first call to be done as root, but it missed one case: when
2617 deliver_queue_load_max is set with deliver_drop_privilege. This is
2618 necessary for the benefit of the queue runner, because there is no re-exec
2619 when deliver_drop_privilege is set.
2621 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2622 This has been fixed.
2624 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2625 was in use, was not putting the data itself into the right store pool;
2626 consequently, it could be overwritten for a subsequent message in the same
2627 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2630 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2632 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2633 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2634 the function were treating the return as a boolean value, which happened to
2635 work because 0=false and not-0=true, but is not correct code.
2637 68. The host_aton() function was not handling scoped IPv6 addresses (those
2638 with, for example, "%eth0" on the end) correctly.
2640 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2641 negated items (that is, ~something) in unsigned ints. Some compilers
2642 apparently mutter when there is no cast.
2644 70. If an address verification called from an ACL failed, and did not produce a
2645 user-specific message (i.e. there was only a "system" message), nothing was
2646 put in $acl_verify_message. In this situation, it now puts the system
2649 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2650 session; change 4.31/43 added the unwanted input to the log line - except
2651 that it did not do this in the start of session case. It now does.
2653 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2654 This is wrong and can cause the other end to generate a synchronization
2655 error if it is another Exim or anything else that does the synchronization
2656 check. A QUIT command is no longer sent after a timeout.
2658 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2659 during host lookups.
2661 74. The defer_ok option of callout verification was not working if it was used
2662 when verifying addresses in header lines, that is, for this case:
2664 verify = header_sender/callout=defer_ok
2666 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2667 those file descriptors could be used for SMTP connections. If anything
2668 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2669 could be sent to the SMTP client, causing chaos. The daemon now opens
2670 stdin, stdout, and stderr to /dev/null when it puts itself into the
2673 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2674 log by default. The user can override this with suitable Perl magic.
2676 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2677 the log message when discard triggers, was not working for the DATA ACL or
2678 for the non-SMTP ACL.
2680 78. Error message wording change in sieve.c.
2682 79. If smtp_accept_max_per_host was set, the number of connections could be
2683 restricted to fewer than expected, because the daemon was trying to set up
2684 a new connection before checking whether the processes handling previous
2685 connections had finished. The check for completed processes is now done
2686 earlier. On busy systems, this bug wouldn't be noticed because something
2687 else would have woken the daemon, and it would have reaped the completed
2690 80. If a message was submitted locally by a user whose login name contained one
2691 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2692 It caused a spool format error. I have fixed the spool reading code. A
2693 related problem was that the "from" clause in the Received: line became
2694 illegal because of the space(s). It is now covered by ${quote_local_part.
2696 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2699 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2701 83. Patch from Sieve maintainer to fix unterminated string problem in
2702 "vacation" handling.
2704 84. Some minor changes to the Linux configuration files to help with other
2705 OS variants using glibc.
2707 85. One more patch for Sieve to update vacation handling to latest spec.
2710 ----------------------------------------------------
2711 See the note above about the 4.44 and 4.50 releases.
2712 ----------------------------------------------------
2718 1. Change 4.43/35 introduced a bug that caused file counts to be
2719 incorrectly computed when quota_filecount was set in an appendfile
2722 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2723 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2726 3. Give more explanation in the error message when the command for a transport
2727 filter fails to execute.
2729 4. There are several places where Exim runs a non-Exim command in a
2730 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2731 was being done only for the command run by the queryprogram router. It is
2732 now done for all such subprocesses. The other cases are: ${run, transport
2733 filters, and the commands run by the lmtp and pipe transports.
2735 5. Some older OS have a limit of 256 on the maximum number of file
2736 descriptors. Exim was using setrlimit() to set 1000 as a large value
2737 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2738 systems. I've change it so that if it can't get 1000, it tries for 256.
2740 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2741 was an oversight, and furthermore, ever since the addition of extra
2742 controls (e.g. 4.43/32), the checks on when to allow different forms of
2743 "control" were broken. There should now be diagnostics for all cases when a
2744 control that does not make sense is encountered.
2746 7. $recipients is now available in the predata ACL (oversight).
2748 8. Tidy the search cache before the fork to do a delivery from a message
2749 received from the command line. Otherwise the child will trigger a lookup
2750 failure and thereby defer the delivery if it tries to use (for example) a
2751 cached ldap connection that the parent has called unbind on.
2753 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2754 of $address_data from the recipient verification was clobbered by the
2755 sender verification.
2757 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2758 was its contents. (It was OK if the option was not defined at all.)
2760 11. A "Completed" log line is now written for messages that are removed from
2761 the spool by the -Mrm option.
2763 12. $host_address is now set to the target address during the checking of
2764 ignore_target_hosts.
2766 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2767 being passed; this would have caused $sender_host_name to have been used if
2768 matching the list had actually called for a host name (not very likely,
2769 since this list is usually IP addresses). A host name is now passed as
2772 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2773 code that specifies a non-restarting handler (typically sigaction() in
2774 modern systems) in an attempt to fix a rare and obscure crash bug.
2776 15. Narrowed the window for a race in the daemon that could cause it to ignore
2777 SIGCHLD signals. This is not a major problem, because they are used only to
2778 wake it up if nothing else does.
2780 16. A malformed maildirsize file could cause Exim to calculate negative values
2781 for the mailbox size or file count. Odd effects could occur as a result.
2782 The maildirsize information is now recalculated if the size or filecount
2785 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2786 support for a long time. Removed HAVE_SYS_VFS_H.
2788 18. Updated exipick to current release from John Jetmore.
2790 19. Allow an empty sender to be matched against a lookup in an address list.
2791 Previously the only cases considered were a regular expression, or an
2794 20. Exim went into a mad DNS lookup loop when doing a callout where the
2795 host was specified on the transport, if the DNS lookup yielded more than
2798 21. The RFC2047 encoding function was originally intended for short strings
2799 such as real names; it was not keeping to the 75-character limit for
2800 encoded words that the RFC imposes. It now respects the limit, and
2801 generates multiple encoded words if necessary. To be on the safe side, I
2802 have increased the buffer size for the ${rfc2047: expansion operator from
2805 22. Failure to deliver a bounce message always caused it to be frozen, even if
2806 there was an errors_to setting on the router. The errors_to setting is now
2809 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2810 canonical form (fully expanded) before being placed in
2811 $sender_host_address.
2813 24. Updated eximstats to version 1.33
2815 25. Include certificate and key file names in error message when GnuTLS fails
2816 to set them up, because the GnuTLS error message doesn't include the name
2817 of the failing file when there is a problem reading it.
2819 26. Expand error message when OpenSSL has problems setting up cert/key files.
2822 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2823 (this can affect the format of dates).
2825 28. exim_tidydb, when checking for the continued existence of a message for
2826 which it has found a message-specific retry record, was not finding
2827 messages that were in split spool directories. Consequently, it was
2828 deleting retry records that should have stayed in existence.
2830 29. eximstats updated to version 1.35
2831 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2832 1.35 - bugfix such that pie charts by volume are generated correctly
2834 30. The SPA authentication driver was not abandoning authentication and moving
2835 on to the next authenticator when an expansion was forced to fail,
2836 contradicting the general specification for all authenticators. Instead it
2837 was generating a temporary error. It now behaves as specified.
2839 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2840 (the order specifies the preference for clients). The order is now AES256,
2841 AES128, 3DES, ARCFOUR128.
2843 31. Small patch to Sieve code - explicitly set From: when generating an
2846 32. Exim crashed if a remote delivery caused a very long error message to be
2847 recorded - for instance if somebody sent an entire SpamAssassin report back
2848 as a large number of 550 error lines. This bug was coincidentally fixed by
2849 increasing the size of one of Exim's internal buffers (big_buffer) that
2850 happened as part of the Exiscan merge. However, to be on the safe side, I
2851 have made the code more robust (and fixed the comments that describe what
2854 33. Some experimental protocols are using DNS PTR records for new purposes. The
2855 keys for these records are domain names, not reversed IP addresses. The
2856 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2857 leaves it alone. Component reversal etc. now happens only for IP addresses.
2860 34. The host_aton() function is supposed to be passed a string that is known
2861 to be a valid IP address. However, in the case of IPv6 addresses, it was
2862 not checking this. This is a hostage to fortune. Exim now panics and dies
2863 if the condition is not met. A case was found where this could be provoked
2864 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2865 components; fortuitously, this particular loophole had already been fixed
2866 by change 4.50/55 or 4.44/33 above.
2868 If there are any other similar loopholes, the new check in host_aton()
2869 itself should stop them being exploited. The report I received stated that
2870 data on the command line could provoke the exploit when Exim was running as
2871 exim, but did not say which command line option was involved. All I could
2872 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2873 running as the user.
2876 35. There was a buffer overflow vulnerability in the SPA authentication code
2877 (which came originally from the Samba project). I have added a test to the
2878 spa_base64_to_bits() function which I hope fixes it.
2881 36. The daemon start-up calls getloadavg() while still root for those OS that
2882 need the first call to be done as root, but it missed one case: when
2883 deliver_queue_load_max is set with deliver_drop_privilege. This is
2884 necessary for the benefit of the queue runner, because there is no re-exec
2885 when deliver_drop_privilege is set.
2887 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2888 was in use, was not putting the data itself into the right store pool;
2889 consequently, it could be overwritten for a subsequent message in the same
2890 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2893 38. Sometimes the final signoff response after QUIT could fail to get
2894 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2895 < 0 before doing a fflush(). This bug looks as though it goes back to the
2896 introduction of TLS in release 3.20, but "sometimes" must have been rare
2897 because the tests only now provoked it.
2903 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2904 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2905 Now it returns an integer. A number of calls were still expecting a T/F
2906 return. Fortuitously, in all cases, the tests worked in OK situations,
2907 which is the norm. However, things would have gone wrong on any write
2908 failures on the smtp file descriptor. This function is used when sending
2909 messages over SMTP and also when doing verify callouts.
2911 2. When Exim is called to do synchronous delivery of a locally submitted
2912 message (the -odf or -odi options), it no longer closes stderr before doing
2915 3. Implemented the mua_wrapper option.
2917 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2919 5. Implemented the functions header_remove(), header_testname(),
2920 header_add_at_position(), and receive_remove_recipient(), and exported them
2923 6. If an ACL "warn" statement specified the addition of headers, Exim already
2924 inserted X-ACL-Warn: at the start if there was no header name. However, it
2925 was not making this test for the second and subsequent header lines if
2926 there were newlines in the string. This meant that an invalid header could
2927 be inserted if Exim was badly configured.
2929 7. Allow an ACL "warn" statement to add header lines at the start or after all
2930 the Received: headers, as well as at the end.
2932 8. Added the rcpt_4xx retry error code.
2934 9. Added postmaster_mailfrom=xxx to callout verification option.
2936 10. Added mailfrom=xxxx to the callout verification option, for verify=
2939 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2940 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2942 12. Inserted some casts to stop certain compilers warning when using pointer
2943 differences as field lengths or precisions in printf-type calls (mostly
2944 affecting debugging statements).
2946 13. Added optional readline() support for -be (dynamically loaded).
2948 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2949 same clock tick as a message's arrival, so that its received time was the
2950 same as the "first fail" time on the retry record, and that message
2951 remained on the queue past the ultimate address timeout, every queue runner
2952 would try a delivery (because it was past the ultimate address timeout) but
2953 after another failure, the ultimate address timeout, which should have then
2954 bounced the address, did not kick in. This was a "< instead of <=" error;
2955 in most cases the first failure would have been in the next clock tick
2956 after the received time, and all would be well.
2958 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2959 being recognized when the domain list was tested by the match_domain
2960 condition in an expansion string.
2962 16. Added the ${str2b64: operator.
2964 17. Exim was always calling setrlimit() to set a large limit for the number of
2965 processes, without checking whether the existing limit was already
2966 adequate. (It did check for the limit on file descriptors.) Furthermore,
2967 errors from getrlimit() and setrlimit() were being ignored. Now they are
2968 logged to the main and panic logs, but Exim does carry on, to try to do its
2969 job under whatever limits there are.
2971 18. Imported PCRE 5.0.
2973 19. Trivial typo in log message " temporarily refused connection" (the leading
2976 20. If the log selector return_path_on_delivery was set and an address was
2977 redirected to /dev/null, the delivery process crashed because it assumed
2978 that a return path would always be set for a "successful" delivery. In this
2979 case, the whole delivery is bypassed as an optimization, and therefore no
2982 21. Internal re-arrangement: the function for sending a challenge and reading
2983 a response while authentication was assuming a zero-terminated challenge
2984 string. It's now changed to take a pointer and a length, to allow for
2985 binary data in such strings.
2987 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2989 23. Exim was not respecting finduser_retries when seeking the login of the
2990 uid under which it was called; it was always trying 10 times. (The default
2991 setting of finduser_retries is zero.) Also, it was sleeping after the final
2992 failure, which is pointless.
2994 24. Implemented tls_on_connect_ports.
2996 25. Implemented acl_smtp_predata.
2998 26. If the domain in control=submission is set empty, Exim assumes that the
2999 authenticated id is a complete email address when it generates From: or
3000 Sender: header lines.
3002 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3003 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3004 chown and chgrp in /bin and hostname in /usr/bin.
3006 28. Exim was keeping the "process log" file open after each use, just as it
3007 does for the main log. This opens the possibility of it remaining open for
3008 long periods when the USR1 signal hits a daemon. Occasional processlog
3009 errors were reported, that could have been caused by this. Anyway, it seems
3010 much more sensible not to leave this file open at all, so that is what now
3013 29. The long-running daemon process does not normally write to the log once it
3014 has entered its main loop, and it closes the log before doing so. This is
3015 so that log files can straightforwardly be renamed and moved. However,
3016 there are a couple of unusual error situations where the daemon does write
3017 log entries, and I had neglected to close the log afterwards.
3019 30. The text of an SMTP error response that was received during a remote
3020 delivery was being truncated at 512 bytes. This is too short for some of
3021 the long messages that one sometimes sees. I've increased the limit to
3024 31. It is now possible to make retry rules that apply only when a message has a
3025 specific sender, in particular, an empty sender.
3027 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3028 it possible to be selective about when SMTP synchronization is enforced.
3030 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3032 32. Implemented hosts_connection_nolog.
3034 33. Added an ACL for QUIT.
3036 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3039 35. Added mailbox_size and mailbox_filecount to appendfile.
3041 36. Added control = no_multiline_responses to ACLs.
3043 37. There was a bug in the logic of the code that waits for the clock to tick
3044 in the case where the clock went backwards by a substantial amount such
3045 that the microsecond fraction of "now" was more than the microsecond
3046 fraction of "then" (but the whole seconds number was less).
3048 38. Added support for the libradius Radius client library this is found on
3049 FreeBSD (previously only the radiusclient library was supported).
3055 1. When certain lookups returned multiple values in the form name=value, the
3056 quoting of the values was not always being done properly. Specifically:
3057 (a) If the value started with a double quote, but contained no whitespace,
3059 (b) If the value contained whitespace other than a space character (i.e.
3060 tabs or newlines or carriage returns) it was not quoted.
3061 This fix has been applied to the mysql and pgsql lookups by writing a
3062 separate quoting function and calling it from the lookup code. The fix
3063 should probably also be applied to nisplus, ibase and oracle lookups, but
3064 since I cannot test any of those, I have not disturbed their existing code.
3066 2. A hit in the callout cache for a specific address caused a log line with no
3067 reason for rejecting RCPT. Now it says "Previous (cached) callout
3068 verification failure".
3070 3. There was an off-by-one bug in the queryprogram router. An over-long
3071 return line was truncated at 256 instead of 255 characters, thereby
3072 overflowing its buffer with the terminating zero. As well as fixing this, I
3073 have increased the buffer size to 1024 (and made a note to document this).
3075 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3076 when Exim is waiting for an SMTP response from a remote server, Exim
3077 restarts its select() call on the socket, thereby resetting its timeout.
3078 This is not a problem when such interrupts are rare. Somebody set up a cron
3079 job to run exiwhat every 2 minutes, which is less than the normal select()
3080 timeout (5 or 10 minutes). This meant that the select() timeout never
3081 kicked in because it was always reset. I have fixed this by comparing the
3082 time when an interrupt arrives with the time at the start of the first call
3083 to select(). If more time than the timeout has elapsed, the interrupt is
3084 treated as a timeout.
3086 5. Some internal re-factoring in preparation for the addition of Sieve
3087 extensions (by MH). In particular, the "personal" test is moved to a
3088 separate function, and given an option for scanning Cc: and Bcc: (which is
3089 not set for Exim filters).
3091 6. When Exim created an email address using the login of the caller as the
3092 local part (e.g. when creating a From: or Sender: header line), it was not
3093 quoting the local part when it contained special characters such as @.
3095 7. Installed new OpenBSD configuration files.
3097 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3098 try to make them clearer.
3100 9. Callout options, other than the timeout value, were being ignored when
3101 verifying sender addresses in header lines. For example, when using
3103 verify = header_sender/callout=no_cache
3105 the cache was (incorrectly) being used.
3107 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3108 only the Cygwin environment.
3110 11. When return_path_on_delivery was set as a log selector, if different remote
3111 addresses in the same message used different return paths and parallel
3112 remote delivery occurred, the wrong values would sometimes be logged.
3113 (Whenever a remote delivery process finished, the return path value from
3114 the most recently started remote delivery process was logged.)
3116 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3117 lines when AUTH and/or TLS are in use. This is the "received protocol"
3118 field. Exim used to use "asmtp" for authenticated SMTP, without any
3119 indication (in the protocol name) for TLS use. Now it follows the RFC and
3120 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3121 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3122 names appear in log lines as well as in Received: header lines.
3124 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3125 extensions, and comparison tests, and to fix some bugs.
3127 14. Changes to the "personal" filter test:
3129 (1) The test was buggy in that it was just doing the equivalent of
3130 "contains" tests on header lines. For example, if a user's address was
3131 anne@some.where, the "personal" test would incorrectly be true for
3133 To: susanne@some.where
3135 This test is now done by extracting each address from the header in turn,
3136 and checking the entire address. Other tests that are part of "personal"
3137 are now done using regular expressions (for example, to check local parts
3138 of addresses in From: header lines).
3140 (2) The list of non-personal local parts in From: addresses has been
3141 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3142 taken from the Sieve specification recommendations.
3144 (3) If the message contains any header line starting with "List-" it is
3145 treated as non-personal.
3147 (4) The test for "circular" in the Subject: header line has been removed
3148 because it now seems ill-conceived.
3150 15. Minor typos in src/EDITME comments corrected.
3152 16. Installed latest exipick from John Jetmore.
3154 17. If headers_add on a router specified a text string that was too long for
3155 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3156 of string_sprintf() is now avoided.
3158 18. $message_body_size was not set (it was always zero) when running the DATA
3159 ACL and the local_scan() function.
3161 19. For the "mail" command in an Exim filter, no default was being set for
3162 the once_repeat time, causing a random time value to be used if "once" was
3163 specified. (If the value happened to be <= 0, no repeat happened.) The
3164 default is now 0s, meaning "never repeat". The "vacation" command was OK
3165 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3166 (I found it when inspecting the code).
3168 20. There is now an overall timeout for performing a callout verification. It
3169 defaults to 4 times the callout timeout, which applies to individual SMTP
3170 commands during the callout. The overall timeout applies when there is more
3171 than one host that can be tried. The timeout is checked before trying the
3172 next host. This prevents very long delays if there are a large number of
3173 hosts and all are timing out (e.g. when the network connections are timing
3174 out). The value of the overall timeout can be changed by specifying an
3175 additional sub-option for "callout", called "maxwait". For example:
3177 verify = sender/callout=5s,maxwait=20s
3179 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3180 to the end before writing, but this should make it even safer).
3182 22. Exim was forgetting that it had advertised PIPELINING for the second and
3183 subsequent messages on an SMTP connection. It was also not resetting its
3184 memory on STARTTLS and an internal HELO.
3186 23. When Exim logs an SMTP synchronization error within a session, it now
3187 records whether PIPELINING has been advertised or not.
3189 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3190 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3191 rather than long int.
3193 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3195 26. Added the never_mail option to autoreply.
3201 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3202 crash if the getsockname() call failed; this can happen if a connection is
3203 closed very soon after it is established. The problem was simply in the
3204 order in which certain operations were done, causing Exim to try to write
3205 to the SMTP stream before it had set up the file descriptor. The bug has
3206 been fixed by making things happen in the correct order.
3212 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3213 before the connection was closed, thus losing the rejection response.
3215 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3216 some early Solaris releases, but causes trouble in current releases where
3217 socklen_t is defined.
3219 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3222 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3225 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3226 that releasing the top of it at the end releases what was used for sub-
3227 expansions (unless the block got too big). However, discard this block if
3228 the first thing is a variable or header, so that we can use its block when
3229 it is dynamic (useful for very large $message_headers, for example).
3231 6. Lookups now cache *every* query, not just the most recent. A new, separate
3232 store pool is used for this. It can be recovered when all lookup caches are
3233 flushed. Lookups now release memory at the end of their result strings.
3234 This has involved some general refactoring of the lookup sources.
3236 7. Some code has been added to the store_xxx() functions to reduce the amount
3237 of flapping under certain conditions.
3239 8. log_incoming_interface used to affect only the <= reception log lines. Now
3240 it causes the local interface and port to be added to several more SMTP log
3241 lines, for example "SMTP connection from", and rejection lines.
3243 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3245 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3247 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3248 could be overwritten at the end of the current message (or the start of a
3249 new message if it was set in a HELO ACL). The value is now preserved for
3250 the duration of the SMTP connection.
3252 12. If a transport had a headers_rewrite setting, and a matching header line
3253 contained an unqualified address, that address was qualified, even if it
3254 did not match any rewriting rules. The underlying bug was that the values
3255 of the flags that permit the existence of unqualified sender and recipient
3256 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3257 non-local messages, and by -bnq for local messages) were not being
3258 preserved with the message after it was received.
3260 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3261 "next input=" as part of the text comprising the host identity instead of
3262 the correct text. The code was using the same buffer for two different
3263 strings. However, depending on which order the printing function evaluated
3264 its arguments, the bug did not always show up. Under Linux, for example, my
3265 test suite worked just fine.
3267 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3268 doesn't work with some older versions of Perl. It has been changed to "my",
3269 which in any case is probably the better facility to use.
3271 15. A really picky compiler found some instances of statements for creating
3272 error messages that either had too many or two few arguments for the format
3275 16. The size of the buffer for calls to the DNS resolver has been increased
3276 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3277 for addresses that have a lot of PTR records. This alleviates a problem; it
3278 does not fully solve it.
3280 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3281 buffer now truncates the list and logs the incident, which is the same
3282 action as happens when Exim is looking up a host name and its aliases.
3283 Previously in this situation something unpredictable would happen;
3284 sometimes it was "internal error: store_reset failed".
3286 18. If a server dropped the connection unexpectedly when an Exim client was
3287 using GnuTLS and trying to read a response, the client delivery process
3288 crashed while trying to generate an error log message.
3290 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3291 string, for example:
3293 warn message = H1: something\nH2: something
3295 the text was added as a single header line from Exim's point of view
3296 though it ended up OK in the delivered message. However, searching for the
3297 second and subsequent header lines using $h_h2: did not work. This has been
3298 fixed. Similarly, if a system filter added multiple headers in this way,
3299 the routers could not see them.
3301 20. Expanded the error message when iplsearch is called with an invalid key to
3302 suggest using net-iplsearch in a host list.
3304 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3305 ACLs are no longer actually imposed (and a message to that effect is
3308 22. If a "gecos" field in a passwd entry contained escaped characters, in
3309 particular, if it contained a \" sequence, Exim got it wrong when building
3310 a From: or a Sender: header from that name. A second bug also caused
3311 incorrect handling when an unquoted " was present following a character
3312 that needed quoting.
3314 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3315 was not being matched caselessly.
3317 24. Arranged for all hyphens in the exim.8 source to be escaped with
3320 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3321 properly. Recipient callout cache records were still being keyed to include
3322 the sender, even when use_sender was set false. This led to far more
3323 callouts that were necessary. The sender is no longer included in the key
3324 when use_sender is false.
3326 26. Added "control = submission" modifier to ACLs.
3328 27. Added the ${base62d: operator to decode base 62 numbers.
3330 28. dnsdb lookups can now access SRV records.
3332 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3333 the configuration file.
3335 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3336 (-v) mode. This makes the output for a verbose queue run more intelligible.
3338 31. Added a use_postmaster feature to recipient callouts.
3340 32. Added the $body_zerocount variable, containing the number of binary zero
3341 bytes in the message body.
3343 33. The time of last modification of the "new" subdirectory is now used as the
3344 "mailbox time last read" when there is a quota error for a maildir
3347 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3349 35. Added +ignore_unknown as a special item in host lists.
3351 36. Code for decoding IPv6 addresses in host lists is now included, even if
3352 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3353 address was recognized as an IP address, but was then not correctly decoded
3354 into binary, causing unexpected and incorrect effects when compared with
3361 1. Very minor rewording of debugging text in manualroute to say "list of
3362 hosts" instead of "hostlist".
3364 2. If verify=header_syntax was set, and a header line with an unqualified
3365 address (no domain) and a large number of spaces between the end of the
3366 name and the colon was received, the reception process suffered a buffer
3367 overflow, and (when I tested it) crashed. This was caused by some obsolete
3368 code that should have been removed. The fix is to remove it!
3370 3. When running in the test harness, delay a bit after writing a bounce
3371 message to get a bit more predictability in the log output.
3373 4. Added a call to search_tidyup() just before forking a reception process. In
3374 theory, someone could use a lookup in the expansion of smtp_accept_max_
3375 per_host which, without the tidyup, could leave open a database connection.
3377 5. Added the variables $recipient_data and $sender_data which get set from a
3378 lookup success in an ACL "recipients" or "senders" condition, or a router
3379 "senders" option, similar to $domain_data and $local_part_data.
3381 6. Moved the writing of debug_print from before to after the "senders" test
3384 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3385 problems for message scanning, either using a data ACL, or using
3386 local_scan() because the Received: header was not generated till after they
3387 were called (in order to set the time as the time of reception completion).
3388 I have revised the way this works. The header is now generated after the
3389 body is received, but before the ACL or local_scan() are called. After they
3390 are run, the timestamp in the header is updated.
3396 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3397 before starting a queue runner without re-exec. This happened only when
3398 deliver_drop_privilege was set or when the Exim user was set to root. The
3399 effect of the bug was that timeouts during subsequent deliveries caused
3400 crashes instead of being properly handled. The handler is now left at its
3401 default (and expected) setting.
3403 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3404 message, again when deliver_drop_privilege is set or Exim is run as root.
3405 The bug described in (1) was not present in this case, but the tidying up
3406 of the other signals was missing. I have made the two cases consistent.
3408 3. The ignore_target_hosts setting on a manualroute router was being ignored
3409 for hosts that were looked up using the /MX notation.
3411 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3414 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3415 operated on the sender address. After changing the $sender_address to <>
3416 for the sender address verify, Exim was re-instated it as the original
3417 (before rewriting) address, but remembering that it had rewritten it, so it
3418 wasn't rewriting it again. This bug also had the effect of breaking the
3419 sender address verification caching when the sender address was rewritten.
3421 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3422 This has been changed so that if the ip literal address matches
3423 ignore_target_hosts, the router declines.
3425 7. Added expansion conditions match_domain, match_address, and match_local_
3426 part (NOT match_host).
3428 8. The placeholder for the Received: header didn't have a length field set.
3430 9. Added code to Exim itself and to exim_lock to test for a specific race
3431 condition that could lead to file corruption when using MBX delivery. The
3432 issue is with the lockfile that is created in /tmp. If this file is removed
3433 after a process has opened it but before that process has acquired a lock,
3434 there is the potential for a second process to recreate the file and also
3435 acquire a lock. This could lead to two Exim processes writing to the file
3436 at the same time. The added code performs the same test as UW imapd; it
3437 checks after acquiring the lock that its file descriptor still refers to
3438 the same named file.
3440 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3441 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3444 11. Added the smtp_active_hostname option. If used, this will typically be made
3445 to depend on the incoming interface address. Because $interface_address is
3446 not set up until the daemon has forked a reception process, error responses
3447 that can happen earlier (such as "too many connections") no longer contain
3450 12. If an expansion in a condition on a "warn" statement fails because a lookup
3451 defers, the "warn" statement is abandoned, and the next ACL statement is
3452 processed. Previously this caused the whole ACL to be aborted.
3454 13. Added the iplsearch lookup type.
3456 14. Added ident_timeout as a log selector.
3458 15. Added tls_certificate_verified as a log selector.
3460 16. Added a global option tls_require_ciphers (compare the smtp transport
3461 option of the same name). This controls incoming TLS connections.
3463 17. I finally figured out how to make tls_require_ciphers do a similar thing
3464 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3465 before starting the TLS session.
3467 18. Tabs are now shown as \t in -bP output.
3469 19. If the log selector return_path_on_delivery was set, Exim crashed when
3470 bouncing a message because it had too many Received: header lines.
3472 20. If two routers both had headers_remove settings, and the first one included
3473 a superfluous trailing colon, the final name in the first list and the
3474 first name in the second list were incorrectly joined into one item (with a
3475 colon in the middle).
3481 1. Added -C and -D options to the exinext utility, mainly to make it easier
3482 to include in the automated testing, but these could be helpful when
3483 multiple configurations are in use.
3485 2. The exinext utility was not formatting the output nicely when there was
3486 an alternate port involved in the retry record key, nor when there was a
3487 message id as well (for retries that were specific to a specific message
3488 and a specific host). It was also confused by IPv6 addresses, because of
3489 the additional colons they contain. I have fixed the IPv4 problem, and
3490 patched it up to do a reasonable job for IPv6.
3492 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3493 delivery, the log line now contains "pipelined" if PIPELINING was used.
3495 4. An SMTP transport process used to panic and die if the bind() call to set
3496 an explicit outgoing interface failed. This has been changed; it is now
3497 treated in the same way as a connect() failure.
3499 5. A reference to $sender_host_name in the part of a conditional expansion
3500 that was being skipped was still causing a DNS lookup. This no longer
3503 6. The def: expansion condition was not recognizing references to header lines
3504 that used bh_ and bheader_.
3506 7. Added the _cache feature to named lists.
3508 8. The code for checking quota_filecount in the appendfile transport was
3509 allowing one more file than it should have been.
3511 9. For compatibility with Sendmail, the command line option
3519 and sets the incoming protocol and host name (for trusted callers). The
3520 host name and its colon can be omitted when only the protocol is to be set.
3521 Note the Exim already has two private options, -pd and -ps, that refer to
3522 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3523 "s", but I don't think that's a major issue.
3525 10. A number of refactoring changes to the code, none of which should affect
3528 (a) The number of logging options was getting close to filling up the
3529 32-bit word that was used as a bit map. I have split them into two classes:
3530 those that are passed in the argument to log_write(), and those that are
3531 only ever tested independently outside of that function. These are now in
3532 separate 32-bit words, so there is plenty of room for expansion again.
3533 There is no change in the user interface or the logging behaviour.
3535 (b) When building, for example, log lines, the code previously used a
3536 macro that called string_cat() twice, in order to add two strings. This is
3537 not really sufficiently general. Furthermore, there was one instance where
3538 it was actually wrong because one of the argument was used twice, and in
3539 one call a function was used. (As it happened, calling the function twice
3540 did not affect the overall behaviour.) The macro has been replaced by a
3541 function that can join an arbitrary number of extra strings onto a growing
3544 (c) The code for expansion conditions now uses a table and a binary chop
3545 instead of a serial search (which was left over from when there were very
3546 few conditions). Also, it now recognizes conditions like "pam" even when
3547 the relevant support is not compiled in: a suitably worded error message is
3548 given if an attempt is made to use such a condition.
3550 11. Added ${time_interval:xxxxx}.
3552 12. A bug was causing one of the ddress fields not to be passed back correctly
3553 from remote delivery subprocesses. The field in question was not being
3554 subsequently used, so this caused to problems in practice.
3556 13. Added new log selectors queue_time and deliver_time.
3558 14. Might have fixed a bug in maildirsizefile handling that threw up
3559 "unexpected character" debug warnings, and recalculated the data
3560 unnecessarily. In any case, I expanded the warning message to give more
3563 15. Added the message "Restricted characters in address" to the statements in
3564 the default ACL that block characters like @ and % in local parts.
3566 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3567 Three changes have been made:
3569 (a) There was a serious bug; a negative response to MAIL caused the whole
3570 recipient domain to be cached as invalid, thereby blocking all messages
3571 to all local parts at the same domain, from all senders. This bug has
3572 been fixed. The domain is no longer cached after a negative response to
3573 MAIL if the sender used is not empty.
3575 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3578 (c) A new callout option, "use_sender" has been added for people who want
3579 the modified behaviour.
3585 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3588 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3589 indeed breaks things for older releases.
3591 3. Added additional logging to the case where there is a problem reading data
3592 from a filter that is running in a subprocess using a pipe, in order to
3593 try to track down a specific problem.
3595 4. Testing facility fudge: when running in the test harness and attempting
3596 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3597 getting "No route to host". Convert this to a timeout.
3599 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3602 6. Some OS don't have socklen_t but use size_t instead. This affects the
3603 fifth argument of getsockopt() amongst other things. This is now
3604 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3605 can be set for individual OS. I have set it for SunOS5, OSF1, and
3606 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3607 some earlier ones do not.
3609 7. Change 4.30/15 was not doing the test caselessly.
3611 8. The standard form for an IPv6 address literal was being rejected by address
3612 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3613 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3614 this, as well as the form without the "IPv6" on the front (but only when
3615 address literals are enabled, of course).
3617 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3619 10. Exim crashed if a message with an empty sender address specified by -f
3620 encountered a router with an errors_to setting. This could be provoked only
3621 by a command such as
3625 where an empty string was supplied; "<>" did not hit this bug.
3627 11. Installed PCRE release 4.5.
3629 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3630 remained set. It is now erased.
3632 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3633 times from message ids (which are base 36 rather than the normal 62).
3635 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3636 were being counted as actual protocol errors, and logged if the log
3637 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3638 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3639 and DATA following a set of rejected RCPTs do not count as protocol errors.
3640 In other words, Exim assumes they were pipelined, though this may not
3641 actually be the case. Of course, in all cases the client gets an
3642 appropriate error code.
3644 15. If a lookup fails in an ACL condition, a message about the failure may
3645 be available; it is used if testing the ACL cannot continue, because most
3646 such messages specify what the cause of the deferral is. However, some
3647 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3648 that caused an old message to be retained and used if a later statement
3649 caused a defer, replacing the real cause of the deferral.
3651 16. If an IP address had so many PTR records that the DNS lookup buffer
3652 was not large enough to hold them, Exim could crash while trying to process
3653 the truncated data. It now detects and logs this case.
3655 17. Further to 4.21/58, another change has been made: if (and only if) the
3656 first line of a message (the first header line) ends with CRLF, a bare LF
3657 in a subsequent header line has a space inserted after it, so as not to
3658 terminate the header.
3660 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3661 unnecessarily, used atoi() instead of strtol(), and didn't check the
3662 termination when getting file sizes from file names by regex.
3664 19. Completely re-implemented the support for maildirsize files, in the light
3665 of a number of problems with the previous contributed implementation
3666 (4.30/29). In particular:
3668 . If the quota is zero, the maildirsize file is maintained, but no quota is
3671 . If the maildir directory does not exist, it is created before any attempt
3672 to write a maildirsize file.
3674 . The quota value in the file is just a cache; if the quota is changed in
3675 the transport, the new value overrides.
3677 . A regular expression is available for excluding directories from the
3680 20. The autoreply transport checks the characters in options that define the
3681 message's headers; it allows continued headers, but it was checking with
3682 isspace() after an embedded newline instead of explicitly looking for a
3685 21. If all the "regular" hosts to which an address was routed had passed their
3686 expiry times, and had not reached their retry times, the address was
3687 bounced, even if fallback hosts were defined. Now Exim should go on to try
3690 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3691 equivalent code in the SMTP transport. Some hosts send humungous responses
3692 to HELO/EHLO, more than 1024 it seems.
3694 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3695 gives compiler warnings in some environments. I've now done it "properly",
3698 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3699 (because of problems with the built-in one) was declared to return uschar *
3700 instead of char *, causing compiler failure.
3702 25. Fixed a file descriptor leak when processing alias/forward files.
3704 26. Fixed a minor format string issue in dbfn.c.
3706 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3708 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3709 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3712 29. When a sender address is verified, it is cached, to save repeating the test
3713 when there is more than one recipient in a message. However, when the
3714 verification involves a callout, it is possible for different callout
3715 options to be set for different recipients. It is too complicated to keep
3716 track of this in the cache, so now Exim always runs a verification when a
3717 callout is required, relying on the callout cache for the optimization.
3718 The overhead is duplication of the address routing, but this should not be
3721 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3722 to be verified with callout=postmaster, and the main callout worked but the
3723 postmaster check failed, the verification correctly failed. However, if a
3724 subsequent RCPT command asked for sender verification *without* the
3725 postmaster check, incorrect caching caused this verification also to fail,
3728 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3729 it was not caching the DNS options (qualify_single, search_parents) that
3730 were used when the lookup failed. A subsequent lookup with different
3731 options therefore always gave the same answer, though there were cases
3732 where it should not have. (Example: a "domains = !$mx_any" option on a
3733 dnslookup router: the "domains" option is always processed without any
3734 widening, but the router might have qualify_single set.) Now Exim uses the
3735 cached value only when the same options are set.
3737 32. Added John Jetmore's "exipick" utility to the distribution.
3739 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3740 than a timeout (e.g. a certificate is required, and is not provided), an
3741 Exim server now closes the connection immediately. Previously it waited for
3742 the client to close - but if the client is SSL, it seems that they each
3743 wait for each other, leading to a delay before one of them times out.
3745 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3746 maintained 0.8.x compatibility because I don't think many are using it, and
3747 it is clearly obsolete.
3749 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3752 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3753 client certificate was expired. A simple patch fixes this, though I don't
3754 understand the full logic of why the verify callback is called multiple
3757 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3758 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3759 which causes problems with some clients (such as the Certicom SSL Plus
3760 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3761 disables the coutermeasure allowing Eudora to connect."
3763 38. Exim was not checking that a write() to a log file succeeded. This could
3764 lead to Bad Things if a log got too big, in particular if it hit a file
3765 size limit. Exim now panics and dies if it cannot write to a log file, just
3766 as it does if it cannot open a log file.
3768 39. Modified OS/Makefile-Linux so that it now contains
3770 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3772 The two -D definitions ensure that Exim is compiled with large file
3773 support, which makes it possible to handle log files that are bigger than
3776 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3777 instance) a domain was checked against a named list that involved a lookup,
3778 causing $domain_data to be set, then another domain was checked against the
3779 same list, then the first domain was re-checked, the value of $domain_data
3780 after the final check could be wrong. In particular, if the second check
3781 failed, it could be set empty. This bug probably also applied to
3784 41. The strip_trailing_dot option was not being applied to the address given
3785 with the -f command-line option.
3787 42. The code for reading a message's header from the spool was incrementing
3788 $received_count, but never initializing it. This meant that the value was
3789 incorrect (doubled) while delivering a message in the same process in which
3790 it was received. In the most common configuration of Exim, this never
3791 happens - a fresh exec is done - but it can happen when
3792 deliver_drop_privilege is set.
3794 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3795 it now includes up to 150 characters of the unexpected data in the log
3798 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3799 and building data strings. The size of both of these buffers was 10 000
3800 bytes - far larger than anybody would *ever* want, thought I. Needless to
3801 say, somebody hit the limit. I have increased the maximum line length to
3802 20 000 and the maximum data length of concatenated lines to 100 000. I have
3803 also fixed two bugs, because there was no checking on these buffers. Tsk,
3804 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3805 buffer is too small.
3807 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3808 lsearch lookups. Now it does.
3810 46. When parsing a route_list item in a manualroute router, a fixed-length
3811 buffer was used for the list of hosts. I made this 1024 bytes long,
3812 thinking that nobody would ever have a list of hosts that long. Wrong.
3813 Somebody had a whole pile of complicated expansion conditions, and the
3814 string was silently truncated, leading to an expansion error. It turns out
3815 that it is easier to change to an unlimited length (owing to other changes
3816 that have happened since this code was originally written) than to build
3817 structure for giving a limitation error. The length of the item that
3818 expands into the list of hosts is now unlimited.
3820 47. The lsearch lookup could not handle data where the length of text line was
3821 more than 4095 characters. Such lines were truncated, leading to shortened
3822 data being returned. It should now handle lines of any length.
3824 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3825 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3828 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3829 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3830 no longer happens in comment lines. A list of replacements is now placed
3831 at the head of all of the source files, except those whose only change is
3832 to replace PERL_COMMAND in the very first #! line.
3834 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3835 messages on the queue, with a bottom-up merge sort, using code contributed
3836 by Michael Haardt. This should make operations like -bp somewhat faster on
3837 large queues. It won't affect queue runners, except when queue_run_in_order
3840 51. Installed eximstats 1.31 in the distribution.
3842 52. Added support for SRV lookups to the dnslookup router.
3844 53. If an ACL referred to $message_body or $message_body_end, the value was not
3845 reset for any messages that followed in the same SMTP session.
3847 54. The store-handling optimization for building very long strings was not
3848 differentiating between the different store pools. I don't think this
3849 actually made any difference in practice, but I've tidied it.
3851 55. While running the routers to verify a sender address, $sender_address
3852 was still set to the sender address. This is wrong, because when routing to
3853 send a bounce to the sender, it would be empty. Therefore, I have changed
3854 it so that, while verifying a sender address, $sender_address is set to <>.
3855 (There is no change to what happens when verifying a recipient address.)
3857 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3858 target A or AAAA records (if not already returned) without resetting the
3859 qualify_single or search_parents options of the DNS resolver. These are
3860 inappropriate in this case because the targets of MX and SRV records must
3861 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3862 target that, when qualified, matched something in the local domain. These
3863 two options are now turned off when doing these lookups.
3865 57. It seems that at least some releases of Reiserfs (which does not have the
3866 concept of a fixed number of inodes) returns zero and not -1 for the
3867 number of available inodes. This interacted badly with check_spool_inodes,
3868 which assumed that -1 was the "no such thing" setting. What I have done is
3869 to check that the total number of inodes is greater than zero before doing
3870 the test of how many are available.
3872 58. When a "warn" ACL statement has a log_message modifier, the message is
3873 remembered, and not repeated. This is to avoid a lot of repetition when a
3874 message has many recipients that cause the same warning to be written.
3875 Howewer, Exim was preserving the list of already written lines for an
3876 entire SMTP session, which doesn't seem right. The memory is now reset if a
3877 new message is started.
3879 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3880 debugging output unless log_rewrite was also set.
3882 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3883 of (int)(handle) when we know that handle contains (void *)(-1).
3885 61. The Exim daemon panic-logs an error return when it closes the incoming
3886 connection. However "connection reset by peer" seems to be common, and
3887 isn't really an error worthy of noting specially, so that particular error
3890 62. When Exim is trying to find all the local interfaces, it used to panic and
3891 die if the ioctl to get the interface flags failed. However, it seems that
3892 on at least one OS (Solaris 9) it is possible to have an interface that is
3893 included in the list of interfaces, but for which you get a failure error
3894 for this call. This happens when the interface is not "plumbed" into a
3895 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3896 failure of the "get flags" call assumes that the interface is down.
3898 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3899 makes life easier for people who are doing arithmetic on fields extracted
3900 from dates, where you often get leading zeros that should not be
3901 interpreted as octal.
3903 64. Added qualify_domain to the redirect router, to override the global
3906 65. If a pathologically long header line contained very many addresses (the
3907 report of this problem mentioned 10 000) and each of them was rewritten,
3908 Exim could use up a very large amount of memory. (It kept on making new
3909 copies of the header line as it rewrote, and never released the old ones.)
3910 At the expense of a bit more processing, the header rewriting function has
3911 been changed so that it no longer eats memory in this way.
3913 66. The generation of the Received: header has been moved from the time that a
3914 message starts to be received, to the time that it finishes. The timestamp
3915 in the Received: header should now be very close to that of the <= log
3916 line. There are two side-effects of this change:
3918 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3919 logged header lines no longer include the local Received: line, because
3920 it has not yet been created. The same applies to a copy of the message
3921 that is returned to a non-SMTP sender when a message is rejected.
3923 (b) When a filter file is tested using -bf, no additional Received: header
3924 is added to the test message. After some thought, I decided that this
3927 This change does not affect the value of $received_for. It is still set
3928 after address rewriting, but before local_scan() is called.
3930 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3932 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3933 gave an unhelpful panic error message, and a defer error. I have managed to
3934 change this behaviour so that it now rejects any supplied certificate,
3935 which seems right, as the list of acceptable certificates is empty.
3937 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3938 gave an unhelpful defer error. I have not managed to make this reject any
3939 supplied certificates, but the error message it gives is "no certificate
3940 supplied", which is not helpful.
3942 70. exigrep's output now also includes lines that are not associated with any
3943 message, but which match the given pattern. Implemented by a patch from
3944 Martin Sluka, which also tidied up the Perl a bit.
3946 71. Recipient callout verification, like sender verification, was using <> in
3947 the MAIL FROM command. This isn't really the right thing, since the actual
3948 sender may affect whether the remote host accepts the recipient or not. I
3949 have changed it to use the actual sender in the callout; this means that
3950 the cache record is now keyed on a recipient/sender pair, not just the
3951 recipient address. There doesn't seem to be a real danger of callout loops,
3952 since a callout by the remote host to check the sender would use <>.
3953 [SEE ABOVE: changed after hitting problems.]
3955 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3956 temporary errors. However, in the case of such a code being given after
3957 the end of a data transmission (i.e. after ".") Exim was failing to write
3958 a retry record for the message. (Yes, there was some broken host that was
3959 actually sending 8xx at this point.)
3961 73. An unknown lookup type in a host list could cause Exim to panic-die when
3962 the list was checked. (An example that provoked this was putting <; in the
3963 middle of a list instead of at the start.) If this happened during a DATA
3964 ACL check, a -D file could be left lying around. This kind of configuration
3965 error no longer causes Exim to die; instead it causes a defer errror. The
3966 incident is still logged to the main and panic logs.
3968 74. Buglet left over from Exim 3 conversion. The message "too many messages
3969 in one connection" was written to the rejectlog but not the mainlog, except
3970 when address rewriting (yes!) was being logged.
3972 75. Added write_rejectlog option.
3974 76. When a system filter was run not as root (that is, when system_filter_user
3975 was set), the values of the $n variables were not being returned to the
3976 main process; thus, they were not subsequently available in the $sn
3979 77. Added +return_path_on_delivery log selector.
3981 78. A connection timeout was being treated differently from recipients deferred
3982 when testing hosts_max_try with a message that was older than the host's
3983 retry timeout. (The host should not be counted, thus allowing all hosts to
3984 be tried at least once before bouncing.) This may have been the cause of an
3985 occasionally reported bug whereby a message would remain on the queue
3986 longer than the retry timeout, but would be bounced if a delivery was
3987 forced. I say "may" because I never totally pinned down the problem;
3988 setting up timeout/retry tests is difficult. See also the next item.
3990 79. The ultimate address timeout was not being applied to errors that involved
3991 a combination of host plus message (for example, a timeout on a MAIL
3992 command). When an address resolved to a number of possible hosts, and they
3993 were not all tried for each delivery (e.g. because of hosts_max_try), a
3994 message could remain on the queue longer than the retry timeout.
3996 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3999 81. Fixed an obscure SMTP outgoing bug which required at least the following
4000 conditions: (a) there was another message waiting for the same server;
4001 (b) the server returned 5xx to all RCPT commands in the first message so
4002 that the message was not completed; (c) the server dropped the connection
4003 or gave a negative response to the RSET that Exim sends to abort the
4004 transaction. The observed case was a dropped connection after DATA that had
4005 been sent in pipelining mode. That is, the server had advertised PIPELINING
4006 but was not implementing it correctly. The effect of the bug was incorrect
4007 behaviour, such as trying another host, and this could lead to a crash.
4013 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4014 and daemon.c were passed as pointers to ints; they should have been
4015 pointers to socklen_t variables (which are typically unsigned ints).
4017 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4020 3. Fixed a really odd bug that affected only the testing scheme; patching a
4021 certain fixed string in the binary changed the value of another string that
4022 happened to be identical to the end of the original first string.
4024 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4025 name", it returns that address as the IP address. On some operating
4026 systems (e.g. Solaris), it also passes back the IP address string as the
4027 "host name". However, on others (e.g. Linux), it passes back an empty
4028 string. Exim wasn't checking for this, and was changing the host name to an
4029 empty string, assuming it had been canonicized.
4031 5. Although rare, it is permitted to have more than one PTR record for a given
4032 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4033 all the names associated with an address, because they do in Solaris.
4034 However, it seems that they do not in Linux for data that comes from the
4035 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4036 I found this out when I moved to a new Linux workstation and tried to run
4037 the Exim test suite.
4039 To get round this problem I have changed the code so that it now does its
4040 own call to the DNS to look up PTR records when searching for a host name.
4041 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4042 addresses that are only in /etc/hosts are still found.
4044 This behaviour is, however, controlled by an option called host_lookup_
4045 order, which defaults to "bydns:byaddr". If people want to use the other
4046 order, or indeed, just use one or the other means of lookup, they can
4047 specify it in this variable.
4049 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4050 some operating systems, this comes back from gethostbyaddr() as an empty
4051 string, and this is what Exim used to test for. However, it seems that in
4052 other systems, "." is yielded. Exim now tests for this case too.
4054 7. The values of check_spool_space and check_log_space are now held internally
4055 as a number of kilobytes instead of an absolute number of bytes. If a
4056 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4057 kilobyte. This means that much larger values can be stored.
4059 8. Exim monitor: an attempt to get the action menu when not actually pointing
4060 at a message produces an empty menu entitled "No message selected". This
4061 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4062 no entries in it ("Shell widget menu has zero width and/or height"). So I
4063 have added a single, blank menu entry in this case.
4065 9. Added ${quote_local_part.
4067 10. MIME decoding is now applied to the contents of Subject: header lines when
4070 11. Now that a reference to $sender_host_address automatically causes a reverse
4071 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4072 host lookup before query-style lookups in lists that might use this
4073 variable. This has therefore been abolished, and the "net-" prefix is no
4074 longer necessary for query-style lookups.
4076 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4077 have been a typo for LFLAGS, so it has been changed.
4079 13. The install script calls Exim with "-C /dev/null" in order to find the
4080 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4081 to be output. Howeve, since Exim outputs its version number before the
4082 error, it didn't break the script. It just looked ugly. I fixed this by
4083 always allowing "-C /dev/null" if the caller is root.
4085 14. Ignore overlarge ACL variable number when reading spool file - insurance
4086 against a later release with more variables having written the file.
4088 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4089 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4090 this, as well as the form without the "IPv6" on the front.
4092 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4093 OS/Makefile-Darwin file.
4095 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4096 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4098 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4099 variable. However, some OS use ints for inodes. I've added cast to long int
4100 to get rid of the compiler warning.
4102 19. I had forgotten to lock out "/../" in configuration file names when
4103 ALT_CONFIG_PREFIX was set.
4105 20. Routers used for verification do not need to specify transports. However,
4106 if such a router generated a host list, and callout was configured, Exim
4107 crashed, because it could not find a port number from the (non-existent)
4108 transport. It now assumes port 25 in this circumstance.
4110 21. Added the -t option to exigrep.
4112 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4113 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4114 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4115 them is set, LOOKUP_LSEARCH is forced.
4117 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4119 24. Added sender and host information to the "rejected by local_scan()" log
4120 line; previously there was no indication of these.
4122 25. Added .include_if_exists.
4124 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4125 Linux. It turns out that not all file systems support this. Apparently some
4126 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4127 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4128 not supported on the file descriptor, is now ignored when Exim is trying to
4129 sync a directory. This applies only to Linux.
4131 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4133 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4134 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4137 29. Added support for maildirsize files from supplied patch (modified a bit).
4139 30. The use of :fail: followed by an empty string could lead Exim to respond to
4140 sender verification failures with (e.g.):
4142 550 Verification failed for <xxx>
4143 550 Sender verify failed
4145 where the first response line was missing the '-' that indicates it is not
4146 the final line of the response.
4148 31. The loop for finding the name of the user that called Exim had a hardwired
4149 limit of 10; it now uses the value of finduser_retries, which is used for
4150 all other user lookups.
4152 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4155 33. Exim was neglecting to zero errno before one call of strtol() when
4156 expanding a string and expecting an integer value. On some systems this
4157 resulted in spurious "integer overflow" errors. Also, it was casting the
4158 result into an int without checking.
4160 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4161 did not work. The code looks as if it has *never* worked, though it appears
4162 to have been documented since at least releast 1.62. I have made it work.
4164 35. The "timeout_DNS" error in retry rules, also documented since at least
4165 1.62, also never worked. As it isn't clear exactly what this means, and
4166 clearly it isn't a major issue, I have abolished the feature by treating it
4167 as "timeout", and writing a warning to the main and panic logs.
4169 36. The display of retry rules for -brt wasn't always showing the error code
4172 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4173 timeout_connect_A, timeout_connect_MX.
4175 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4176 to the empty sender.
4178 39. The daemon was not analysing the content of -oX till after it had closed
4179 stderr and disconnected from the controlling terminal. This meant that any
4180 syntax errors were only noted on the panic log, and the return code from
4181 the command was 0. By re-arranging the code a little, I've made the
4182 decoding happen first, so such errors now appear on stderr, and the return
4183 code is 1. However, the actual setting up of the sockets still happens in
4184 the disconnected process, so errors there are still only recorded on the
4187 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4188 connections (as happens on some IP stacks) was logged at start up time as
4189 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4190 it from "IPv6 and IPv4", which means that two separate sockets are being
4193 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4194 says whether AF_INET or AF_INET6 was passed as an argument.
4196 42. Exiwhat output was messed up when time zones were included in log
4199 43. Exiwhat now gives more information about the daemon's listening ports,
4200 and whether -tls-on-connect was used.
4202 44. The "port" option of the smtp transport is now expanded.
4204 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4205 silently ignored. Now an error message is written to the main and panic
4208 46. There's a new ACL modifier called "logwrite" which writes to a log file
4209 as soon as it is encountered.
4211 47. Added $local_user_uid and $local_user_gid at routing time.
4213 48. Exim crashed when trying to verify a sender address that was being
4216 49. Exim was recognizing only a space character after ".include". It now also
4217 recognizes a tab character.
4219 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4220 extracting the relevant information from the specification. The man page no
4221 longer contains scrambled data for the -d option, and I've added a section
4222 at the front about calling Exim under different names.
4224 51. Added "extra_headers" argument to the "mail" command in filter files.
4226 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4229 53. Installed eximstats 1.29.
4231 54. Added transport_filter_timeout as a generic transport option.
4233 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4234 Cc: header lines. This was required by RFC 822, but it not required by RFC
4237 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4238 incoming messages that did not have them. Now it does so only if the
4239 message originates locally, that is, if there is no associated remote host
4240 address. When Resent- header lines are present, this applies to the Resent-
4241 lines rather than the non-Resent- lines.
4243 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4244 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4246 58. Messages for configuration errors now include the name of the main
4247 configuration file - useful now that there may be more than one file in a
4248 list (.included file names were always shown).
4250 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4251 for those rare installations that do not start the daemon as root or run it
4252 setuid root. I've cut out the call to initgroups() if the daemon is not
4255 60. The Exim user and group can now be bound into the binary as text strings
4256 that are looked up at the start of Exim's processing.
4258 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4260 62. Added $mailstore_basename variable.
4262 63. Installed patch to sieve.c from Michael Haardt.
4264 64. When Exim failed to open the panic log after failing to open the main log,
4265 the original message it was trying to log was written to stderr and debug
4266 output, but if they were not available (the usual case in production), it
4267 was lost. Now it is written to syslog before the two lines that record the
4268 failures to open the logs.
4270 65. Users' Exim filters run in subprocesses under the user's uid. It is
4271 possible for a "deliver" command or an alias in a "personal" command to
4272 provoke an address rewrite. If logging of address rewriting is configured,
4273 this fails because the process is not running as root or exim. There may be
4274 a better way of dealing with this, but for the moment (because 4.30 needs
4275 to be released), I have disabled address rewrite logging when running a
4276 filter in a non-root, non-exim process.
4282 1. The buildconfig auxiliary program wasn't quoting the value set for
4283 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4284 not defined. This bug was masked in 4.22 by the effect that was fixed in
4287 2. Some messages that were rejected after a message id was allocated were
4288 shown as "incomplete" by exigrep. It no longer does this for messages that
4289 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4291 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4292 have allow_domain_literals set, the ID did not get logged in the <= line.
4293 Domain literals are now always recognized in Message-ID: header lines.
4295 4. The first argument for a ${extract expansion item is the key name or field
4296 number. Leading and trailing spaces in this item were not being ignored,
4297 causing some misleading effects.
4299 5. When deliver_drop_privilege was set, single queue runner processes started
4300 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4301 same command in the process it spins off) were not dropping privilege.
4303 6. When the daemon running as "exim" started a queue runner, it always
4304 re-executed Exim in the spun-off process. This is a waste of effort when
4305 deliver_drop_privilege is set. The new process now just calls the
4306 queue-runner function directly.
4312 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4315 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4316 ignored. Though the use of -oP was forcing the writing of a pid file, it
4317 was always written to the default place.
4319 3. If the message "no IP address found for host xxxx" is generated during
4320 incoming verification, it is now followed by identification of the incoming
4321 connection (so you can more easily find what provoked it).
4323 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4325 5. Added some features to "harden" Exim a bit more against certain attacks:
4327 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4328 be put in Local/Makefile. This is like the never_users runtime option,
4329 but it cannot be overridden. The default setting is "root".
4331 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4332 prefix string with which any file named in a -C command line option
4335 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4336 is retained for -C and -D only if the caller of Exim is root. Without
4337 it, the exim user may also use -C and -D and retain privilege.
4339 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4340 command line option is disabled.
4342 6. Macro names set by the -D option must start with an upper case letter, just
4343 like macro names defined in the configuration file.
4345 7. Added "dereference=" facility to LDAP.
4347 8. Two instances of the typo "uknown" in the source files are fixed.
4349 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4350 the Configure-Makefile script screwed up while processing it.
4352 10. Incorporated PCRE 4.4.
4354 11. The SMTP synchronization check was not operating right at the start of an
4355 SMTP session. For example, it could not catch a HELO sent before the client
4356 waited for the greeting. There is now a check for outstanding input at the
4357 point when the greeting is written. Because of the duplex, asynchronous
4358 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4359 way, but not yet received, when the check is performed.
4361 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4362 on TCP/IP sockets, because this apparently causes some broken clients to
4365 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4366 unchanged) from the Cygwin maintainer.
4368 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4369 was supported instead of testing for mbx. Effectively a typo.
4371 15. The spa authenticator server code was not checking that the input it
4372 received was valid base64.
4374 16. The debug output line for the "set" modifier in ACLs was not showing the
4375 name of the variable that was being set.
4377 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4379 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4380 lookup on its own. Something else was needed to trigger the lookup. For
4381 example, a match in host_lookup or the need for a host name in a host list.
4382 Now, if $sender_host_name is referenced and the host name has not yet been
4383 looked up, a lookup is performed. If the lookup fails, the variable remains
4384 empty, and $host_lookup_failed is set to "1".
4386 19. Added "eqi" as a case-independent comparison operator.
4388 20. The saslauthd authentication condition could segfault if neither service
4389 nor realm was specified.
4391 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4392 error message that was logged was misleading, and incoming SMTP
4393 connections were dropped. The message is now more accurate, and temporary
4394 errors are given to SMTP connections.
4396 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4397 (except RSET) with a 421 error, until QUIT is received. However, it was
4398 failing to send a response to QUIT.
4400 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4401 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4402 ACL is now run after the helo_try_verify_hosts code.
4404 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4405 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4406 case-independent, so other case variants are also recognized). Apparently
4407 some systems use these upper case variants.
4409 25. If more than two messages were waiting for the same host, and a transport
4410 filter was specified for the transport, Exim sent two messages over the
4411 same TCP/IP connection, and then failed with "socket operation on non-
4412 socket" when it tried to send the third.
4414 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4416 27. The extern definition of crypt16() in expand.c was not being excluded when
4417 the OS had its own crypt16() function.
4419 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4420 as a preferred synonym for return_size_limit, both as an option and as an
4423 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4425 30. Changed the default configuration ACL to relax the local part checking rule
4426 for addresses that are not in any local domains. For these addresses,
4427 slashes and pipe symbols are allowed within local parts, but the sequence
4428 /../ is explicitly forbidden.
4430 31. SPA server authentication was not clearing the challenge buffer before
4433 32. log_message in a "warn" ACL statement was writing to the reject log as
4434 well as to the main log, which contradicts the documentation and doesn't
4435 seem right (because no rejection is happening). So I have stopped it.
4437 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4438 However, I am unable to do any testing of this.
4440 34. Fixed an infelicity in the appendfile transport. When checking directories
4441 for a mailbox, to see if any needed to be created, it was accidentally
4442 using path names with one or more superfluous leading slashes; tracing
4443 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4445 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4446 contents are added to the log line that is written for every discarded
4447 recipient. (Previously a log_message setting was ignored.)
4449 36. The ${quote: operator now quotes the string if it is empty.
4451 37. The install script runs exim in order to find its version number. If for
4452 some reason other than non-existence or emptiness, which it checks, it
4453 could not run './exim', it was installing it with an empty version number,
4454 i.e. as "exim-". This error state is now caught, and the installation is
4457 38. An argument was missing from the function that creates an error message
4458 when Exim fails to connect to the socket for saslauthd authentication.
4459 This could cause Exim to crash, or give a corrupted message.
4461 39. Added isip, isip4, and isip6 to ${if conditions.
4463 40. The ACL variables $acl_xx are now saved with the message, and can be
4464 accessed later in routers, transports, and filters.
4466 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4467 strings in the file are not string-expanded.
4469 42. If a MAIL command specified a SIZE value that was too large to fit into an
4470 int variable, the check against message_size_limit failed. Such values are
4471 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4472 day this will have to be increased, but I don't think I want to be around
4473 when emails are that large.
4480 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4481 iconv() is not standard in FreeBSD.
4483 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4484 IPv6 enabled. The observed symptom was a segmentation fault on return from
4485 the function os_common_find_running_interfaces() in src/os.c.
4487 3. In the check_special_case() function in daemon.c I had used "errno" as an
4488 argument name, which causes warnings on some systems. This was basically a
4489 typo, since it was named "eno" in the comments!
4491 4. The code that waits for the clock to tick (at a resolution of some fraction
4492 of a second) so as to ensure message-id uniqueness was always waiting for
4493 at least one whole tick, when it could have waited for less. [This is
4494 almost certainly not relevant at current processor speeds, where it is
4495 unlikely to ever wait at all. But we try to future-proof.]
4497 5. The function that sleeps for a time interval that includes fractions of a
4498 second contained a race. It did not block SIGALRM between setting the
4499 timer, and suspending (a couple of lines later). If the interval was short
4500 and the sigsuspend() was delayed until after it had expired, the suspension
4501 never ended. On busy systems this could lead to processes getting stuck for
4504 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4505 process, before it forks any delivery processes. The open lookup caching
4506 mechanism meant that the open file or database connection was passed into
4507 the delivery process. The problem was that delivery processes always tidy
4508 up cached lookup data. This could cause a problem for the next delivery
4509 process started by the queue runner, because the external queue runner
4510 process does not know about the closure. So the next delivery process
4511 still has data in the lookup cache. In the case of a file lookup, there was
4512 no problem because closing a file descriptor in a subprocess doesn't affect
4513 the parent. However, if the lookup was caching a connection to a database,
4514 the connection was closed, and the second delivery process was likely to
4515 see errors such as "PGSQL: query failed: server closed the connection
4516 unexpectedly". The problem has been fixed by closing all cached lookups
4517 in a queue runner before running a delivery process.
4519 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4520 seem to have the "const" qualifier which it has on other OS. I've
4523 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4524 *of the same type* (client or server) with the same public name that an
4525 error should be diagnosed.
4527 9. When Exim looked up a host name for an IP address, but failed to find the
4528 original IP address when looking up the host name (a safety check), it
4529 output the message "<ip address> does not match any IP for NULL", which was
4530 confusing, to say the least. The bug was that the host name should have
4531 appeared instead of "NULL".
4533 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4534 user that is built into the binary, and the -C or -D options is used, root
4535 privilege is dropped before the configuration file is read. In addition,
4536 logging is switched to stderr instead of the normal log files. If the
4537 configuration then re-defines the Exim user, the unprivileged environment
4538 is probably not what is expected, so Exim logs a panic warning message (but
4541 However, if deliver_drop_privilege is set, the unprivileged state may well
4542 be exactly what is intended, so the warning has been cut out in that case,
4543 and Exim is allowed to try to write to its normal log files.
4549 1. smtp_return_error_details was not giving details for temporary sender
4550 or receiver verification errors.
4552 2. Diagnose a configuration error if two authenticators have the same public
4555 3. Exim used not to create the message log file for a message until the first
4556 delivery attempt. This could be confusing when incoming messages were held
4557 for policy or load reasons. The message log file is now created at the time
4558 the message is received, and an initial "Received" line is written to it.
4560 4. The automatically generated man page for command line options had a minor
4561 bug that caused no ill effects; however, a more serious problem was that
4562 the procedure for building the man page automatically didn't always
4563 operate. Consequently, release 4.20 contains an out-of-date version. This
4564 shouldn't happen again.
4566 5. When building Exim with embedded Perl support, the script that builds the
4567 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4568 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4570 6. The freeze_tell option was not being used for messages that were frozen on
4571 arrival, either by an ACL or by local_scan().
4573 7. Added the smtp_incomplete_transaction log selector.
4575 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4576 was accepting AUTH without a new EHLO.
4578 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4579 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4582 10. Logging of TCP/IP connections (when configured) now happens in the main
4583 daemon process instead of the child process, so that the TCP/IP connection
4584 count is more accurate (but it can never be perfect).
4586 11. The use of "drop" in a nested ACL was not being handled correctly in the
4587 outer ACL. Now, if condition failure induced by the nested "drop" causes
4588 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4589 or "require"), the connection is dropped.
4591 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4592 that yield "discard" can now be used with an "accept" or a "discard" verb,
4593 but an error is generated for any others (because I can't see a useful way
4594 to define what should happen).
4596 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4597 are now processed in the same way as lines in the Exim configuration file.
4598 In particular, continuation lines are supported.
4600 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4602 15. Added -ti meaning -t -i.
4604 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4605 domains, and warn by logging if others are found.
4607 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4608 returned from GIFCONF when Exim is trying to find the list of interfaces on
4609 a host. The code in os.c has been modified to copy each ifreq to an aligned
4610 structure in all cases.
4612 Also, in some cases, the returned ifreq's were being copied to a 'struct
4613 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4614 means the last couple of bytes of an IPv6 address could be chopped if the
4615 ifreq contained only a normal sockaddr (14 bytes storage).
4617 18. Named domain lists were not supported in the hosts_treat_as_local option.
4618 An entry such as +xxxx was not recognized, and was treated as a literal
4621 19. Ensure that header lines added by a DATA ACL are included in the reject log
4622 if the ACL subsequently rejects the message.
4624 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4625 MD5 (which is deprecated).
4627 21. When testing a filter file using -bf, Exim was writing a message when it
4628 took the sender from a "From " line in the message, but it was not doing so
4629 when it took $return_path from a Return-Path: header line. It now does.
4631 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4632 with a valid header line field name (a series of printing characters
4633 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4635 23. Changed "disc" in the source to "disk" to conform to the documentation and
4636 the book and for uniformity.
4638 24. Ignore Sendmail's -Ooption=value command line item.
4640 25. When execve() failed while trying to run a command in a pipe transport,
4641 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4642 could be confused with a return value of 69 from the command itself. This
4643 has been changed to 127, the value the shell returns if it is asked to run
4644 a non-existent command. The wording for the related log line suggests a
4645 non-existent command as the problem.
4647 26. If received_header_text expands to an empty string, do not add a Received:
4648 header line to the message. (Well, it adds a token one on the spool, but
4649 marks it "old" so that it doesn't get used or transmitted.)
4651 27. Installed eximstats 1.28 (addition of -nt option).
4653 28. There was no check for failure on the call to getsockname() in the daemon
4654 code. This can fail if there is a shortage of resources on the system, with
4655 ENOMEM, for example. A temporary error is now given on failure.
4657 29. Contrary to the C standard, it seems that in some environments, the
4658 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4659 program. Exim now does this explicitly; it affects the formatting of
4660 timestamps using strftime().
4662 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4663 complaints. I've now initialized all the variables, to avoid this.
4665 32. Header lines added by a system filter were not being "seen" during
4666 transport-time rewrites.
4668 33. The info_callback() function passed to OpenSSL is set up with type void
4669 (*)(SSL *, int, int), as described somewhere. However, when calling the
4670 function (actually a macro) that sets it up, the type void(*)() is
4671 expected. I've put in a cast to prevent warnings from picky compilers.
4673 34. If a DNS black list lookup found a CNAME record, but there were no A
4674 records associated with the domain it pointed at, Exim crashed.
4676 35. If a DNS black list lookup returned more than one A record, Exim ignored
4677 all but the first. It now scans all returned addresses if a particular IP
4678 value is being sought. In this situation, the contents of the
4679 $dnslist_value variable are a list of all the addresses, separated by a
4682 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4683 to accept a host name and all its aliases if the forward lookup for any of
4684 them yielded the IP address of the incoming connection. Now it accepts only
4685 those names whose forward lookup yields the correct IP address. Any other
4686 names are discarded. This closes a loophole whereby a rogue DNS
4687 administrator could create reverse DNS records to break through a
4688 wildcarded host restriction in an ACL.
4690 37. If a user filter or a system filter that ran in a subprocess used any of
4691 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4692 the wrong values were passed to the pipe command ($thisaddress had the
4693 value of $0, $0 had the value of $1, etc). This bug was introduced by
4694 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4696 38. Improved the line breaking for long SMTP error messages from ACLs.
4697 Previously, if there was no break point between 40 and 75 characters, Exim
4698 left the rest of the message alone. Two changes have been made: (a) I've
4699 reduced the minimum length to 35 characters; (b) if it can't find a break
4700 point between 35 and 75 characters, it looks ahead and uses the first one
4701 that it finds. This may give the occasional overlong line, but at least the
4702 remaining text gets split now.
4704 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4705 file descriptors might be low, and that setting 1000 would always raise it.
4706 It turns out that in some environments, the limit is already over 1000 and
4707 that lowering it causes trouble. So now Exim takes care not to decrease it.
4709 40. When delivering a message, the value of $return_path is set to $sender_
4710 address at the start of routing (routers may change the value). By an
4711 oversight, this default was not being set up when an address was tested by
4712 -bt or -bv, which affected the outcome if any router or filter referred to
4715 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4716 log only when "message" or "log_message" are set. However, if one of the
4717 conditions was an address verification, or a call to a nested ACL, the
4718 messages generated by the underlying test were being passed through. This
4719 no longer happens. The underlying message is available in $acl_verify_
4720 message for both "message" and "log_message" expansions, so it can be
4721 passed through if needed.
4723 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4724 new expansion $bh_ to give the encoded byte string without charset
4725 translation. Translation happens only if iconv() is available; HAVE_ICONV
4726 indicates this at build time. HEADERS_CHARSET gives the charset to
4727 translate to; headers_charset can change it in the configuration, and
4728 "headers charset" can change it in an individual filter file.
4730 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4731 that creates RFC 2047 encoded "words" labels them as that charset instead
4732 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4733 expansion operator; (ii) when Exim creates a From: line for a local
4734 message; (iii) when a header line is rewritten to include a "phrase" part.
4736 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4737 buggy, causing it to skip the first lines of messages whose message ID
4738 ended in 'D'. This would not have bitten before Exim release 4.14, because
4739 message IDs were unlikely to end in 'D' before then. The effect was to have
4740 incorrect size information for certain domains.
4742 45. #include "config.h" was missing at the start of the crypt16.c module. This
4743 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4746 46. If there was a timeout during a "random" callout check, Exim treated it as
4747 a failure of the random address, and carried on sending RSET and the real
4748 address. If the delay was just some slowness somewhere, the response to the
4749 original RCPT would be taken as a response to RSET and so on, causing
4750 mayhem of various kinds.
4752 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4753 when I implemented it. It didn't allow for the fact that some option values
4754 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4755 the right test for positive values.
4757 48. Domain names in DNS records are case-independent. Exim always looks them up
4758 in lower case. Some resolvers return domain names in exactly the case they
4759 appear in the zone file, that is, they may contain uppercase letters. Not
4760 all resolvers do this - some return always lower case. Exim was treating a
4761 change of case by a resolver as a change of domain, similar to a widening
4762 of a domain abbreviation. This triggered its re-routing code and so it was
4763 trying to route what was effectively the same domain again. This normally
4764 caused routing to fail (because the router wouldn't handle the domain
4765 twice). Now Exim checks for this case specially, and just changes the
4766 casing of the domain that it ultimately uses when it transmits the message
4769 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4772 50. If a filter generated a file delivery with a non-absolute name (possible if
4773 no home directory exists for the router), the forbid_file option was not
4776 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4777 the existing equality matching.
4779 52. Exim was using ints instead of ino_t variables in some places where it was
4780 dealing with inode numbers.
4782 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4783 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4784 called TMPDIR, and if it finds it is different, it changes its value.
4786 54. The smtp_printf() function is now made available to local_scan() so
4787 additional output lines can be written before returning. There is also an
4788 smtp_fflush() function to enable the detection of a dropped connection.
4789 The variables smtp_input and smtp_batched_input are exported to
4792 55. Changed the default runtime configuration: the message "Unknown user"
4793 has been removed from the ACL, and instead placed on the localuser router,
4794 using the cannot_route_message feature. This means that any verification
4795 failures that generate their own messages won't get overridden. Similarly,
4796 the "Unrouteable address" message that was in the ACL for unverifiable
4797 relay addresses has also been removed.
4799 56. Added hosts_avoid_esmtp to the smtp transport.
4801 57. The exicyclog script was not checking for the esoteric option
4802 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4803 will work only if exicyclog is run under the appropriate euid.
4805 58. Following a discussion on the list, the rules by which Exim recognises line
4806 endings on incoming messages have been changed. The -dropcr and drop_cr
4807 options are now no-ops, retained only for backwards compatibility. The
4808 following line terminators are recognized: LF CRLF CR. However, special
4809 processing applies to CR:
4811 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4812 nor a local message in the state where . is a terminator.
4814 (ii) If a bare CR is encountered in a header line, an extra space is added
4815 after the line terminator so as not to end the header. The reasoning
4816 behind this is that bare CRs in header lines are most likely either
4817 to be mistakes, or people trying to play silly games.
4819 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4820 was being incorrectly given as 18 bytes larger than it should have been.
4821 This is a VOB (very old bug).
4823 60. This may never have affected anything current, but just in case it has:
4824 When the local host is found other than at the start of a list of hosts,
4825 the local host, those with the same MX, and any that follow, are discarded.
4826 When the list in question was part of a longer list of hosts, the following
4827 hosts (not currently being processed) were also being discarded. This no
4828 longer happens. I'm not sure if this situation could ever has previously
4831 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4834 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4835 header. This is something that is recommended in a new Internet Draft, and
4836 is something that is documented as being done by Sendmail. There are two
4837 possible values. For messages generated by the autoreply transport, Exim
4840 Auto-Submitted: auto-replied
4842 whereas for all other generated messages (e.g. bounces) it adds
4844 Auto-Submitted: auto-generated
4846 63. The "personal" condition in filters now includes a test for the
4847 Auto-Submitted: header. If it contains the string "auto-" the message it
4848 not considered personal.
4850 64. Added rcpt_include_affixes as a generic transport option.
4852 65. Added queue_only_override (default true).
4854 66. Added the syslog_duplication option.
4856 67. If what should have been the first header line of a message consisted of
4857 a space followed by a colon, Exim was mis-interpreting it as a header line.
4858 It isn't of course - it is syntactically invalid and should therefore be
4859 treated as the start of the message body. The misbehaviour could have
4860 caused a number of strange effects, including loss of data in subsequent
4861 header lines, and spool format errors.
4863 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4864 client host had authenticated. This control can now be exercised by an ACL
4865 for more flexibility.
4867 69. By default, callouts do not happen when testing with -bh. There is now a
4868 variant, -bhc, which does actually run the callout code, including
4869 consulting and updating the callout cache.
4871 70. Added support for saslauthd authentication, courtesy of Alexander
4874 71. If statvfs() failed on the spool or log directories while checking their
4875 size for availability, Exim confusingly gave the error "space shortage".
4876 Furthermore, in debugging mode it crashed with a floating point exception.
4877 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4878 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4879 serious problem, Exim now writes to the main and panic logs when this
4880 happens, with details of the failure. It then refuses to accept the
4881 incoming message, giving the message "spool directory problem" or "log
4882 directory problem" with a 421 code for SMTP messages.
4884 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4885 0, 1, and 2 exist, because some OS complain for execs without them (see
4886 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4887 descriptors. However, the code omitted to check that the open succeeded,
4888 causing mysterious errors if for some reason the permissions on /dev/null
4889 got screwed. Now Exim writes a message to the main and panic logs, and
4890 bombs out if it can't open /dev/null.
4892 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4893 interact so that it is all more flexible. It is supposed to remain
4894 backwards compatible. Also added extra_local_interfaces.
4896 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4897 to bomb out with an assertion failure - to the client this appears as a
4898 connection drop. This problem occurs in the part of the code that was taken
4899 from the Samba project. Fortunately, the assertion is in a very simple
4900 function, so I have fixed this by reproducing the function inline in the
4901 one place where it is called, and arranging for authentication to fail
4902 instead of killing the process with assert().
4904 75. The SPA client code was not working when the server requested OEM rather
4905 than Unicode encoding.
4907 76. Added code to make require_files with a specific uid setting more usable in
4908 the case where statting the file as root fails - usually a non-root-mounted
4909 NFS file system. When this happens and the failure is EACCES, Exim now
4910 forks a subprocess and does the per-uid checking as the relevant uid.
4912 77. Added process_log_path.
4914 78. If log_file_path was not explicitly set, a setting of check_log_space or
4915 check_log_inodes was ignored.
4917 79. If a space check for the spool or log partitions fails, the incident is now
4918 logged. Of course, in the latter case the data may get lost...
4920 80. Added the %p formatting code to string_format() so that it can be used to
4921 print addresses in debug_print(). Adjusted all the address printing in the
4922 debugging in store.c to use %p rather than %d.
4924 81. There was a concern that a line of code in smtp_in.c could overflow a
4925 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4926 initially expressed, the concern was not well-founded, because trailing
4927 spaces are removed early. However, if the trailing spaces were followed by
4928 a NULL, they did not get removed, so the overflow was possible. Two fixes
4931 (a) I re-wrote the offending code in a cleaner fashion.
4932 (b) If an incoming SMTP command contains a NULL character, it is rejected
4935 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4936 runs initgroups(), so that if the Exim user is in any additional groups,
4937 they will be used during message reception.
4943 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob