1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.566 2009/10/14 11:26:31 nm4 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added patch by Johannes Berg that expands the main option
10 "spamd_servers" if it starts with a dollar sign.
12 TK/02 Write list of recipients to X-Envelope-Sender header when building
13 the mbox-format spool file for content scanning (suggested by Jakob
16 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
17 (http://www.dcc-servers.net/) support via dccifd. Activated by
18 setting EXPERIMENTAL_DCC=yes in Local/Makefile. Check out
19 experimental_spec.txt for more documentation.
21 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
22 by Mark Daniel Reidel <mr@df.eu>.
24 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
25 When building exim an external PCRE library is now needed -
26 PCRE is a system library on the majority of modern systems.
27 See entry on PCRE_LIBS in EDITME file.
29 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
30 conversation. Added nologin parameter to request.
31 Patch contributed by Kirill Miazine
33 TF/01 Do not log submission mode rewrites if they do not change the address.
35 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
37 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
38 log files in place. Contributed by Roberto Lima
40 NM/04 Bugzilla 667: close socket used by dovecot authenticator
42 TF/03 Bugzilla 615: When checking the local_parts router precondition
43 after a local_part_suffix or local_part_prefix option, Exim now
44 does not use the address's named list lookup cache, since this
45 contains cached lookups for the whole local part.
47 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
48 Robert Millan. Documentation is in experimental-spec.txt
50 TF/04 Bugzilla 668: Fix parallel build (make -j).
52 NM/05 Bugzilla 437: Prevent Maildix aux files being created with mode 000
54 NM/05 Bugzilla 598: Improvement to Dovecot authenticator handling.
55 Patch provided by Jan Srzednicki
57 TF/05 Leading white space used to be stripped from $spam_report which
58 wrecked the formatting. Now it is preserved.
60 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
61 that they are available at delivery time.
63 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
65 TF/08 TLS error reporting now respects the incoming_interface and
66 incoming_port log selectors.
68 TF/09 Produce a more useful error message if an SMTP transport's hosts
69 setting expands to an empty string.
71 NM/06 Bugzilla 744: EXPN did not work under TLS.
72 Patch provided by Phil Pennock
74 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
75 Patch provided by Richard Godbee
77 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
78 acl_smtp_notquit, added index entry.
80 NM/09 Bugzilla 787: Potential buffer overflow in string_format
81 Patch provided by Eugene Bujak
83 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to accept()
84 Patch provided by Maxim Dounin
86 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
87 Patch provided by Phil Pennock
89 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
91 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
92 Patch provided by Brad "anomie" Jorsch
94 NM/14 Bugzilla 614: Added timeout setting to transport filter.
95 Patch provided by Dean Brooks
97 TK/05 Add native DKIM support (does not depend on external libraries).
99 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
100 Patch provided by Graeme Fowler
102 NM/16 Bugzilla 851: Documentation example syntax fix.
104 NM/17 Changed NOTICE file to remove references to embedded PCRE.
110 TK/01 Add preliminary DKIM support. Currently requires a forked version of
111 ALT-N's libdkim that I have put here:
112 http://duncanthrax.net/exim-experimental/
114 Note to Michael Haardt: I had to rename some vars in sieve.c. They
115 were called 'true' and it seems that C99 defines that as a reserved
116 keyword to be used with 'bool' variable types. That means you could
117 not include C99-style headers which use bools without triggering
118 build errors in sieve.c.
120 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
121 as mailq or other aliases. Changed the --help handling significantly
122 to do whats expected. exim_usage() emits usage/help information.
124 SC/01 Added the -bylocaldomain option to eximstats.
126 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
128 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
130 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
136 PH/01 Another patch from the Sieve maintainer.
138 PH/02 When an IPv6 address is converted to a string for single-key lookup
139 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
140 dots are used instead of colons so that keys in lsearch files need not
141 contain colons. This was done some time before quoting was made available
142 in lsearch files. However, iplsearch files do require colons in IPv6 keys
143 (notated using the quote facility) so as to distinguish them from IPv4
144 keys. This meant that lookups for IP addresses in host lists did not work
145 for iplsearch lookups.
147 This has been fixed by arranging for IPv6 addresses to be expressed with
148 colons if the lookup type is iplsearch. This is not incompatible, because
149 previously such lookups could never work.
151 The situation is now rather anomolous, since one *can* have colons in
152 ordinary lsearch keys. However, making the change in all cases is
153 incompatible and would probably break a number of configurations.
155 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
158 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
159 conversion specification without a maximum field width, thereby enabling
160 a rogue spamd server to cause a buffer overflow. While nobody in their
161 right mind would setup Exim to query an untrusted spamd server, an
162 attacker that gains access to a server running spamd could potentially
163 exploit this vulnerability to run arbitrary code as the Exim user.
165 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
166 $primary_hostname instead of what libspf2 thinks the hosts name is.
168 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
169 a directory entry by the name of the lookup key. Previously, if a
170 symlink pointed to a non-existing file or a file in a directory that
171 Exim lacked permissions to read, a lookup for a key matching that
172 symlink would fail. Now it is enough that a matching directory entry
173 exists, symlink or not. (Bugzilla 503.)
175 PH/03 The body_linecount and body_zerocount variables are now exported in the
178 PH/04 Added the $dnslist_matched variable.
180 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
181 This means they are set thereafter only if the connection becomes
184 PH/06 Added the client_condition to authenticators so that some can be skipped
185 by clients under certain conditions.
187 PH/07 The error message for a badly-placed control=no_multiline_responses left
188 "_responses" off the end of the name.
190 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
192 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
193 (without spaces) instead of just copying the configuration text.
195 PH/10 Added the /noupdate option to the ratelimit ACL condition.
197 PH/11 Added $max_received_linelength.
199 PH/12 Added +ignore_defer and +include_defer to host lists.
201 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
202 way in which PCRE > 7.0 is built.
204 PH/14 Implemented queue_only_load_latch.
206 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
207 MAIL command. The effect was to mangle the value on 64-bit systems.
209 PH/16 Another patch from the Sieve maintainer.
211 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
213 PH/18 If a system quota error occurred while trying to create the file for
214 a maildir delivery, the message "Mailbox is full" was not appended to the
215 bounce if the delivery eventually timed out. Change 4.67/27 below applied
216 only to a quota excession during the actual writing of the file.
218 PH/19 It seems that peer DN values may contain newlines (and other non-printing
219 characters?) which causes problems in log lines. The DN values are now
220 passed through string_printing() before being added to log lines.
222 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
223 and InterBase are left for another time.)
225 PH/21 Added message_body_newlines option.
227 PH/22 Guard against possible overflow in moan_check_errorcopy().
229 PH/23 POSIX allows open() to be a macro; guard against that.
231 PH/24 If the recipient of an error message contained an @ in the local part
232 (suitably quoted, of course), incorrect values were put in $domain and
233 $local_part during the evaluation of errors_copy.
239 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
240 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
243 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
244 issue a MAIL command.
246 PH/02 In an ACL statement such as
248 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
250 if a client was not listed at all, or was listed with a value other than
251 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
252 the condition was not true (as it should be), so access was not denied.
253 The bug was that the ! inversion was incorrectly passed on to the second
254 item. This has been fixed.
256 PH/03 Added additional dnslists conditions == and =& which are different from
257 = and & when the dns lookup returns more than one IP address.
259 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
260 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
262 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
263 FSYNC, which compiles an option called disable_fsync that allows for
264 bypassing fsync(). The documentation is heavily laced with warnings.
266 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
268 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
269 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
270 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
271 including adding "make clean"; (3) Added -fPIC when compiling the test
272 dynamically loaded module, to get rid of a warning.
274 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
275 message fails, move_frozen_messages = true and ignore_bounce_errors_after
276 = 0s. The bug is otherwise harmless.
278 PH/07 There was a bug in the dovecot authenticator such that the value of
279 $auth1 could be overwritten, and so not correctly preserved, after a
280 successful authentication. This usually meant that the value preserved by
281 the server_setid option was incorrect.
283 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
285 PH/09 Installed PCRE release 7.0.
287 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
288 run for batched SMTP input. It is now run at the start of every message
289 in the batch. While fixing this I discovered that the process information
290 (output by running exiwhat) was not always getting set for -bs and -bS
291 input. This is fixed, and it now also says "batched" for BSMTP.
293 PH/11 Added control=no_pipelining.
295 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
296 patch, slightly modified), and move the expansion of helo_data till after
297 the connection is made in the smtp transport (so it can use these
300 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
302 PH/14 Added log_selector = +pid.
304 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
306 PH/16 Add ${if forany and ${if forall.
308 PH/17 Added dsn_from option to vary the From: line in DSNs.
310 PH/18 Flush SMTP output before performing a callout, unless control =
311 no_callout_flush is set.
313 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
314 was true (the default) a successful delivery failed to delete the retry
315 item, thus causing premature timeout of the address. The bug is now
318 PH/20 Added hosts_avoid_pipelining to the smtp transport.
320 PH/21 Long custom messages for fakedefer and fakereject are now split up
321 into multiline reponses in the same way that messages for "deny" and
322 other ACL rejections are.
324 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
325 with slight modification.
327 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
328 draft, changing the syntax and factoring some duplicate code".
330 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
331 for deliveries of the second and subsequent messages over the same SMTP
334 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
335 ${reduce, with only minor "tidies".
337 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
339 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
340 expansion side effects.
342 PH/27 When a message times out after an over-quota error from an Exim-imposed
343 quota, the bounce message says "mailbox is full". This message was not
344 being given when it was a system quota that was exceeded. It now should
347 MH/03 Made $recipients available in local_scan(). local_scan() already has
348 better access to the recipient list through recipients_list[], but
349 $recipients can be useful in postmaster-provided expansion strings.
351 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
352 in the case of a MAIL command with additional options following the
353 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
354 were accidentally chopped off.
356 PH/29 SMTP synchronization checks are implemented when a command is read -
357 there is a check that no more input is waiting when there shouldn't be
358 any. However, for some commands, a delay in an ACL can mean that it is
359 some time before the response is written. In this time, more input might
360 arrive, invalidly. So now there are extra checks after an ACL has run for
361 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
362 pipelining has not been advertised.
364 PH/30 MH's patch to allow iscntrl() characters to be list separators.
366 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
367 returned in the SMTP response when smtp_return_error_details was false.
370 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
371 instead of the C I/O that was originally supplied, because problems were
374 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
375 Exim which did not show up earlier: it was assuming that a call to
376 SSL_CTX_set_info_callback() might give an error value. In fact, there is
377 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
378 was a macro that became an assignment, so it seemed to work. This has
379 changed to a proper function call with a void return, hence the compile
380 error. Exim's code has been fixed.
382 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
385 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
387 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
389 PH/37 If a message is not accepted after it has had an id assigned (e.g.
390 because it turns out to be too big or there is a timeout) there is no
391 "Completed" line in the log. When some messages of this type were
392 selected by exigrep, they were listed as "not completed". Others were
393 picked up by some special patterns. I have improved the selection
394 criteria to be more general.
396 PH/38 The host_find_failed option in the manualroute router can now be set
397 to "ignore", to completely ignore a host whose IP address cannot be
398 found. If all hosts are ignored, the behaviour is controlled by the new
399 host_all_ignored option.
401 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
402 homing or because of multiple MX records with /mx) generated more than
403 one IP address, and the following item turned out to be the local host,
404 all the secondary addresses of the first item were incorrectly removed
405 from the list, along with the local host and any following hosts (which
406 is what is supposed to happen).
408 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
409 whoever called Exim into the -H file. In the case of the daemon it was
410 behaving confusingly. When first started, it used values for whoever
411 started the daemon, but after a SIGHUP it used the Exim user (because it
412 calls itself on a restart). I have changed the code so that it now always
415 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
416 message are rejected with the same error (e.g. no authentication or bad
417 sender address), and a DATA command is nevertheless sent (as can happen
418 with PIPELINING or a stupid MUA), the error message that was given to the
419 RCPT commands is included in the rejection of the DATA command. This is
420 intended to be helpful for MUAs that show only the final error to their
423 PH/42 Another patch from the Sieve maintainer.
425 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
426 Eximstats - Fixed some broken HTML links and added missing column headers
428 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
429 columns for Rejects, Temp Rejects, Ham, and Spam rows.
431 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
433 PH/43 Yet another patch from the Sieve maintainer.
435 PH/44 I found a way to check for a TCP/IP connection going away before sending
436 the response to the final '.' that terminates a message, but only in the
437 case where the client has not sent further data following the '.'
438 (unfortunately, this is allowed). However, in many cases there won't be
439 any further data because there won't be any more messages to send. A call
440 to select() can be used: if it shows that the input is "ready", there is
441 either input waiting, or the socket has been closed. An attempt to read
442 the next input character can distinguish the two cases. Previously, Exim
443 would have sent an OK response which the client would never have see.
444 This could lead to message repetition. This fix should cure that, at
445 least in a lot of common cases.
447 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
448 advertised in response to EHLO.
454 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
455 fixed by 4.65/MH/01 (is this a record?) are fixed:
457 (i) An empty string was always treated as zero by the numeric comparison
458 operators. This behaviour has been restored.
460 (ii) It is documented that the numeric comparison operators always treat
461 their arguments as decimal numbers. This was broken in that numbers
462 starting with 0 were being interpreted as octal.
464 While fixing these problems I realized that there was another issue that
465 hadn't been noticed. Values of message_size_limit (both the global option
466 and the transport option) were treated as octal if they started with 0.
467 The documentation was vague. These values are now always treated as
468 decimal, and I will make that clear in the documentation.
474 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
475 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
478 MH/01 Don't check that the operands of numeric comparison operators are
479 integers when their expansion is in "skipping" mode (fixes bug
480 introduced by 4.64-PH/07).
482 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
483 child addresses, Exim now panics and dies. Previously, because the count
484 is held in a short int, deliveries were likely to be lost. As such a
485 large number of recipients for a single message is ridiculous
486 (performance will be very, very poor), I have chosen to impose a limit
487 rather than extend the field.
493 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
494 leftover -K file (the existence of which was triggered by #402).
495 While we were at it, introduced process PID as part of the -K
496 filename. This should rule out race conditions when creating
499 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
500 processing considerably. Previous code took too long for large mails,
501 triggering a timeout which in turn triggers #401.
503 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
504 in the DK code in transports.c. sendfile() is not really portable,
505 hence the _LINUX specificness.
507 TF/01 In the add_headers option to the mail command in an Exim filter,
508 there was a bug that Exim would claim a syntax error in any
509 header after the first one which had an odd number of characters
512 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
513 callout verification, Exim cached a "reject" for the entire domain. This
514 is correct for most verifications, but it is not correct for a recipient
515 verification with use_sender or use_postmaster set, because in that case
516 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
517 case of MAIL FROM:<> rejection from other early rejections (e.g.
518 rejection of HELO). When verifying a recipient using a non-null MAIL
519 address, the cache is ignored if it shows MAIL FROM:<> rejection.
520 Whatever the result of the callout, the value of the domain cache is
521 left unchanged (for any other kind of callout, getting as far as trying
522 RCPT means that the domain itself is ok).
524 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
527 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
528 manifest itself as EPIPE rather than ECONNECT. When tidying away a
529 session, the daemon ignores ECONNECT errors and logs others; it now
530 ignores EPIPE as well.
532 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
533 (quoted-printable decoding).
535 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
536 later the small subsequent patch to fix an introduced bug.
538 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
540 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
542 PH/08 An error is now given if message_size_limit is specified negative.
544 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
545 to be given (somewhat) arbitrary names.
547 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
550 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
551 miscellaneous code fixes
553 PH/10 Added the log_reject_target ACL modifier to specify where to log
556 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
557 hostname. This is wrong, because it relates to the incoming message (and
558 probably the interface on which it is arriving) and not to the outgoing
559 callout (which could be using a different interface). This has been
560 changed to use the value of the helo_data option from the smtp transport
561 instead - this is what is used when a message is actually being sent. If
562 there is no remote transport (possible with a router that sets up host
563 addresses), $smtp_active_hostname is used.
565 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
566 tweaks were necessary in order to get it to work (see also 21 below):
567 (a) The code assumed that strncpy() returns a negative number on buffer
568 overflow, which isn't the case. Replaced with Exim's string_format()
570 (b) There were several signed/unsigned issues. I just did the minimum
571 hacking in of casts. There is scope for a larger refactoring.
572 (c) The code used strcasecmp() which is not a standard C function.
573 Replaced with Exim's strcmpic() function.
574 (d) The code set only $1; it now sets $auth1 as well.
575 (e) A simple test gave the error "authentication client didn't specify
576 service in request". It would seem that Dovecot has changed its
577 interface. Fortunately there's a specification; I followed it and
578 changed what the client sends and it appears to be working now.
580 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
583 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
584 address A is aliased to B and C, where B exists and C does not. Without
585 -v the output is "A verified" because verification stops after a
586 successful redirection if more than one address is generated. However,
587 with -v the child addresses are also verified. Exim was outputting "A
588 failed to verify" and then showing the successful verification for C,
589 with its parentage. It now outputs "B failed to verify", showing B's
590 parentage before showing the successful verification of C.
592 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
593 look up a TXT record in a specific list after matching in a combined
596 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
597 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
598 they consult the DNS. I had assumed they would set it the way they
599 wanted; and indeed my experiments on Linux seem to show that in some
600 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
601 To be on the safe side, however, I have now made the interface to
602 host_find_byname() similar to host_find_bydns(), with an argument
603 containing the DNS resolver options. The host_find_byname() function now
604 sets these options at its start, just as host_find_bydns() does. The smtp
605 transport options dns_qualify_single and dns_search_parents are passed to
606 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
607 of host_find_byname() use the default settings of RES_DEFNAMES
608 (qualify_single) but not RES_DNSRCH (search_parents).
610 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
611 spool_read_header() do less string testing, by means of a preliminary
612 switch on the second character of optional "-foo" lines. (This is
613 overdue, caused by the large number of possibilities that now exist.
614 Originally there were few.) While I was there, I also converted the
615 str(n)cmp tests so they don't re-test the leading "-" and the first
616 character, in the hope this might squeeze out yet more improvement.
618 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
619 flag allowing group syntax was set by the header_syntax check but not
620 turned off, possible causing trouble later; (2) The flag was not being
621 set at all for the header_verify test, causing "group"-style headers to
622 be rejected. I have now set it in this case, and also caused header_
623 verify to ignore an empty address taken from a group. While doing this, I
624 came across some other cases where the code for allowing group syntax
625 while scanning a header line wasn't quite right (mostly, not resetting
626 the flag correctly in the right place). These bugs could have caused
627 trouble for malformed header lines. I hope it is now all correct.
629 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
630 with the "reply" argument non-NULL. The code, however (which originally
631 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
632 but it didn't always do it. This confused somebody who was copying the
633 code for some other use. I have removed all the tests.
635 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
636 feature that was used to support insecure browsers during the U.S. crypto
637 embargo. It requires special client support, and Exim is probably the
638 only MTA that supported it -- and would never use it because real RSA is
639 always available. This code has been removed, because it had the bad
640 effect of slowing Exim down by computing (never used) parameters for the
641 RSA_EXPORT functionality.
643 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
644 authenticator to fail if there's a tab character in the incoming data
645 (there should never be unless someone is messing about, as it's supposed
646 to be base64-encoded). Also added, on Timo's advice, the "secured" option
647 if the connection is using TLS or if the remote IP is the same as the
648 local IP, and the "valid-client-cert option" if a client certificate has
651 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
652 authenticators. This can be used for authorization after authentication
653 succeeds. (In the case of plaintext, it servers for both authentication
656 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
657 if any retry times were supplied.
659 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
660 connection, where there is no client IP address to check. In this
661 situation, the verify now always succeeds.
663 PH/25 Applied John Jetmore's -Mset patch.
665 PH/26 Added -bem to be like -Mset, but loading a message from a file.
667 PH/27 In a string expansion for a processed (not raw) header when multiple
668 headers of the same name were present, leading whitespace was being
669 removed from all of them, but trailing whitespace was being removed only
670 from the last one. Now trailing whitespace is removed from each header
671 before concatenation. Completely empty headers in a concatenation (as
674 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
675 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
677 PH/29 [Removed. This was a change that I later backed out, and forgot to
678 correct the ChangeLog entry (that I had efficiently created) before
679 committing the later change.]
681 PH/30 Exim was sometimes attempting to deliver messages that had suffered
682 address errors (4xx response to RCPT) over the same connection as other
683 messages routed to the same hosts. Such deliveries are always "forced",
684 so retry times are not inspected. This resulted in far too many retries
685 for the affected addresses. The effect occurred only when there were more
686 hosts than the hosts_max_try setting in the smtp transport when it had
687 the 4xx errors. Those hosts that it had tried were not added to the list
688 of hosts for which the message was waiting, so if all were tried, there
689 was no problem. Two fixes have been applied:
691 (i) If there are any address or message errors in an SMTP delivery, none
692 of the hosts (tried or untried) are now added to the list of hosts
693 for which the message is waiting, so the message should not be a
694 candidate for sending over the same connection that was used for a
695 successful delivery of some other message. This seems entirely
696 reasonable: after all the message is NOT "waiting for some host".
697 This is so "obvious" that I'm not sure why it wasn't done
698 previously. Hope I haven't missed anything, but it can't do any
699 harm, as the worst effect is to miss an optimization.
701 (ii) If, despite (i), such a delivery is accidentally attempted, the
702 routing retry time is respected, so at least it doesn't keep
703 hammering the server.
705 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
706 in ${readsocket because some servers need this prod.
708 PH/32 Added some extra debug output when updating a wait-xxx database.
710 PH/33 The hint "could be header name not terminated by colon", which has been
711 given for certain expansion errors for a long time, was not being given
712 for the ${if def:h_colon_omitted{... case.
714 PH/34 The spec says: "With one important exception, whenever a domain list is
715 being scanned, $domain contains the subject domain." There was at least
716 one case where this was not true.
718 PH/35 The error "getsockname() failed: connection reset by peer" was being
719 written to the panic log as well as the main log, but it isn't really
720 panic-worthy as it just means the connection died rather early on. I have
721 removed the panic log writing for the ECONNRESET error when getsockname()
724 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
725 runs only) independently of the message's sender address. This meant
726 that, if the 4xx error was in fact related to the sender, a different
727 message to the same recipient with a different sender could confuse
728 things. In particualar, this can happen when sending to a greylisting
729 server, but other circumstances could also provoke similar problems.
730 I have changed the default so that the retry time for these errors is now
731 based a combination of the sender and recipient addresses. This change
732 can be overridden by setting address_retry_include_sender=false in the
735 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
736 remote server are returned as part of bounce messages. This was not
737 happening for LMTP over a pipe (the lmtp transport), but now it is the
738 same for both kinds of LMTP.
740 PH/38 Despite being documented as not happening, Exim was rewriting addresses
741 in header lines that were in fact CNAMEs. This is no longer the case.
743 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
744 and queue runs started by the daemon processed all messages. This has
745 been fixed so that -R and -S can now usefully be given with -q<time>.
747 PH/40 Import PCRE release 6.7 (fixes some bugs).
749 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
751 PH/42 Give an error if -q is specified more than once.
753 PH/43 Renamed the variables $interface_address and $interface_port as
754 $received_ip_address and $received_port, to make it clear that these
755 values apply to message reception, and not to the outgoing interface when
756 a message is delivered. (The old names remain recognized, of course.)
758 PH/44 There was no timeout on the connect() call when using a Unix domain
759 socket in the ${readsocket expansion. There now is.
761 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
762 be meaningful with "accept".
764 SC/01 Eximstats V1.43
765 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
767 SC/02 Eximstats V1.44
768 Use a glob alias rather than an array ref in the generated
769 parser. This improves both readability and performance.
771 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
772 Collect SpamAssassin and rejection statistics.
773 Don't display local sender or destination tables unless
774 there is data to show.
775 Added average volumes into the top table text output.
777 SC/04 Eximstats V1.46
778 Collect data on the number of addresses (recipients)
779 as well as the number of messages.
781 SC/05 Eximstats V1.47
782 Added 'Message too big' to the list of mail rejection
783 reasons (thanks to Marco Gaiarin).
785 SC/06 Eximstats V1.48
786 Mainlog lines which have GMT offsets and are too short to
787 have a flag are now skipped.
789 SC/07 Eximstats V1.49 (Alain Williams)
790 Added the -emptyok flag.
792 SC/08 Eximstats V1.50
793 Fixes for obtaining the IP address from reject messages.
795 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
796 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
797 whitesspace changes from 4.64-PH/27
799 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
802 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
803 are found, allow negative numbers in numeric criteria)
805 JJ/06 exipick.20061117.2, added new $message_body_missing variable
807 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
808 to match changes made in 4.64-PH/43
810 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
812 PH/47 Put in an explicit test for a DNS lookup of an address record where the
813 "domain" is actually an IP address, and force a failure. This locks out
814 those revolvers/nameservers that support "A-for-A" lookups, in
815 contravention of the specifications.
817 PH/48 When a host name was looked up from an IP address, and the subsequent
818 forward lookup of the name timed out, the host name was left in
819 $sender_host_name, contrary to the specification.
821 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
822 restricted to single-key lookups, Exim was not diagnosing an error if
823 * or *@ was used with a query-style lookup.
825 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
827 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
828 long ago, but noone interested enough thought of it. Let's just say that
829 the "1.1" means that there are some new functions that weren't there at
830 some point in the past.
832 PH/51 Error processing for expansion failure of helo_data from an smtp
833 transport during callout processing was broken.
835 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
836 tested/used via the -bh/-bhc/-bs options.
838 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
839 bug, fixed in subsequent PCRE releases).
841 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
842 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
844 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
850 SC/01 Use a glob alias rather than an array ref in eximstats generated
851 parser. This improves both readability and performance.
853 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
854 Don't display local sender or destination tables in eximstats unless
855 there is data to show.
856 Added average volumes into the eximstats top table text output.
858 SC/03 Collect data on the number of addresses (recipients) as well
859 as the number of messages in eximstats.
861 TF/01 Correct an error in the documentation for the redirect router. Exim
862 does (usually) call initgroups() when daemonizing.
864 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
865 with consistent privilege compared to when running as a daemon.
867 TF/03 Note in the spec that $authenticated_id is not set for local
868 submissions from trusted users.
870 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
871 Thanks to Dean Brooks <dean@iglou.com> for the patch.
873 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
874 by adding some example configuration directives to the default
875 configuration file. A little bit of work is required to uncomment the
876 directives and define how usernames and passwords are checked, but
877 there is now a framework to start from.
879 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
880 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
881 without this. I don't know how relevant this is to other LDAP libraries.
883 PH/02 Add the verb name to the "unknown ACL verb" error.
885 PH/03 Magnus Holmgren's patch for filter_prepend_home.
887 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
889 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
890 directory not expanded when it should be if an expanded home directory
891 was set for the address (which is overridden by the transport).
893 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
896 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
897 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
898 because it is too late at that time, and has no effect.
900 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
901 security issue with \' (bugzilla #107). I could not use the
902 PQescapeStringConn() function, because it needs a PGconn value as one of
905 PH/08 When testing addresses using -bt, indicate those final addresses that
906 are duplicates that would not cause an additional delivery. At least one
907 person was confused, thinking that -bt output corresponded to deliveries.
908 (Suppressing duplicates isn't a good idea as you lose the information
909 about possibly different redirections that led to the duplicates.)
911 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
912 systems where poll() doesn't work, in particular OS X.
914 PH/10 Added more information to debugging output for retry time not reached.
916 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
917 operations in malware.c.
919 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
922 PH/13 If write_rejectlog was set false when logging was sent to syslog with
923 syslog_duplication set false, log lines that would normally be written
924 both the the main log and to the reject log were not written to syslog at
927 PH/14 In the default configuration, change the use of "message" in ACL warn
928 statements to "add_header".
930 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
931 not followed by a command (e.g. "seen endif").
933 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
934 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
937 PH/17 Added extra conditions to the default value of delay_warning_condition
941 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
942 { match{$h_precedence:}{(?i)bulk|list|junk} } \
943 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
946 The Auto-Submitted: and various List- headers are standardised, whereas I
947 don't think Precedence: ever was.
949 PH/18 Refactored debugging code in route_finduser() to show more information,
950 in particular, the error code if getpwnam() issues one.
952 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
953 This is apparently needed in addition to the PH/07 change above to avoid
954 any possible encoding problems.
956 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
957 but not after initializing Perl.
959 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
960 output them only if debugging. By default they are written stderr,
961 apparently, which is not desirable.
963 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
966 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
969 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
971 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
972 authenticated or an ident call has been made. Suppress the default
973 values for $authenticated_id and $authenticated_sender (but permit -oMai
974 and -oMas) when testing with -bh.
976 PH/24 Re-jigged the order of the tests in the default configuration so that the
977 tests for valid domains and recipients precede the DNS black list and CSA
978 tests, on the grounds that those ones are more expensive.
980 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
981 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
982 This bug exists in every version of Exim that I still have, right back to
985 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
986 However, an attempt to turn on case-sensitivity in a regex key by
987 including (?-i) didn't work because the subject string was already
988 lowercased, and the effects were non-intuitive. It turns out that a
989 one-line patch can be used to allow (?-i) to work as expected.
995 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
996 other effects) broke the use of negated acl sub-conditions.
998 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1001 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1002 "Deny" causes Exim to reject the incoming connection with a 554 error.
1003 Unfortunately, if there is a major crisis, such as a disk failure,
1004 tcp-wrappers gives "deny", whereas what one would like would be some
1005 kind of temporary error. A kludge has been added to help with this.
1006 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1007 554 error is used if errno is still zero or contains ENOENT (which occurs
1008 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1011 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1013 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1014 errors. Otherwise a message that provokes a temporary error (when other
1015 messages do not) can cause a whole host to time out.
1017 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1018 addresses were routed directly to files or pipes from a redirect router.
1019 File deliveries just didn't batch; pipe deliveries might have suffered
1022 PH/06 A failure to get a lock for a hints database would erroneously always say
1023 "Failed to get write lock", even when it was really a read lock.
1025 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1026 of 0600. This has been changed to use the value of the lockfile_mode
1027 option (which defaults to 0600).
1029 PH/08 Applied small patch from the Sieve maintainer.
1031 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1032 folder from quota calculations, a direct delivery into this folder messed
1033 up the contents of the maildirsize file. This was because the regex was
1034 used only to exclude .Trash (or whatever) when the size of the mailbox
1035 was calculated. There was no check that a delivery was happening into an
1036 excluded directory. This bug has been fixed by ignoring all quota
1037 processing for deliveries into excluded directories.
1039 PH/10 Added the maildirfolder_create_regex option to appendfile.
1045 PH/01 The code for finding all the local interface addresses on a FreeBSD
1046 system running IPv6 was broken. This may well have applied to all BSD
1047 systems, as well as to others that have similar system calls. The broken
1048 code found IPv4 interfaces correctly, but gave incorrect values for the
1049 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1050 that it would not match correctly against @[] and not recognize the IPv6
1053 PH/02 The ipliteral router was not recognizing addresses of the form user@
1054 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1056 PH/03 Added disable_ipv6.
1058 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1059 decoded form, because it is most often used to construct To: headers
1060 lines in autoreplies, and the decoded form may well be syntactically
1061 invalid. However, $reply_address has leading white space removed, and all
1062 newlines turned into spaces so that the autoreply transport does not
1065 PH/05 If group was specified without a user on a router, and no group or user
1066 was specified on a transport, the group from the router was ignored.
1068 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1069 for visible compile-time settings that can be used to change these
1070 numbers, for those that want even more. Backwards compatibility with old
1071 spool files has been maintained. However, going back to a previous Exim
1072 release will lost any variables that are in spool files.
1074 PH/07 Two small changes when running in the test harness: increase delay when
1075 passing a TCP/IP connection to a new process, in case the original
1076 process has to generate a bounce, and remove special handling of
1077 127.0.0.2 (sic), which is no longer necessary.
1079 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1080 be the same on different OS.
1082 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1085 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1086 whether --show-vars was specified or not
1088 JJ/02 exipick: Added support for new ACL variable spool format introduced
1091 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1092 syntactically invalid From: or Reply-to: line, and a filter used this to
1093 generate an autoreply, and therefore failed to obtain an address for the
1094 autoreply, Exim could try to deliver to a non-existent relative file
1095 name, causing unrelated and misleading errors. What now happens is that
1096 it logs this as a hard delivery error, but does not attempt to create a
1099 PH/11 The exinext utility has a -C option for testing purposes, but although
1100 the given file was scanned by exinext itself; it wasn't being passed on
1101 when Exim was called.
1103 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1104 an end-of-file indication when reading a command response.
1106 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1107 compiled. In many other places in Exim, IPv6 addresses are always
1108 recognized, so I have changed this. It also means that IPv4 domain
1109 literals of the form [IPV4:n.n.n.n] are now always recognized.
1111 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1112 used if the router is not running as root, for example, when verifying at
1113 ACL time, or when using -bh. The debugging output from this situation was
1114 non-existent - all you got was a failure to exec. I have made two
1117 (a) Failures to set uid/gid, the current directory, or a process leader
1118 in a subprocess such as that created by queryprogram now generate
1119 suitable debugging ouput when -d is set.
1121 (b) The queryprogram router detects when it is not running as root,
1122 outputs suitable debugging information if -d is set, and then runs
1123 the subprocess without attempting to change uid/gid.
1125 PH/15 Minor change to Makefile for building test_host (undocumented testing
1128 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1129 additional section of a DNS packet that returns MX or SRV records.
1130 Instead, it always explicitly searches for A/AAAA records. This avoids
1131 major problems that occur when a DNS server includes only records of one
1132 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1133 fixed another bug: if SRV records were looked up and the corresponding
1134 address records were *not* found in the additional section, the port
1135 values from the SRV records were lost.
1137 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1138 using the correct key (the original address) when searching the retry
1139 rules in order to find which one to use for generating the retry hint.
1141 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1142 adding the default one. Similarly, if it contains a Reply-To: header, the
1143 errors_reply_to option, if set, is not used.
1145 PH/19 When calculating a retry time, Exim used to measure the "time since
1146 failure" by looking at the "first failed" field in the retry record. Now
1147 it does not use this if it is later than than the arrival time of the
1148 message. Instead it uses the arrival time. This makes for better
1149 behaviour in cases where some deliveries succeed, thus re-setting the
1150 "first failed" field. An example is a quota failure for a huge message
1151 when small messages continue to be delivered. Without this change, the
1152 "time since failure" will always be short, possible causing more frequent
1153 delivery attempts for the huge message than are intended.
1154 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1156 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1157 $1, $2, $3) because the numerical variables can be reset during some
1158 expansion items (e.g. "match"), thereby losing the authentication data.
1160 PH/21 Make -bV show the size of off_t variables so that the test suite can
1161 decide whether to run tests for quotas > 2G.
1163 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1164 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1165 filecount value is greater than 2G or if a quota value is greater than 2G
1166 on a system where the size of off_t is not greater than 4, a panic error
1169 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1170 never match. The debug and -bh output now contains an explicit error
1171 message indicating a malformed IPv4 address or mask.
1173 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1174 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1175 PH/23 above applies.
1177 PH/25 Do not write to syslog when running in the test harness. The only
1178 occasion when this arises is a failure to open the main or panic logs
1179 (for which there is an explicit test).
1181 PH/26 Added the /no_tell option to "control=freeze".
1183 PH/27 If a host name lookup failed very early in a connection, for example, if
1184 the IP address matched host_lookup and the reverse lookup yielded a name
1185 that did not have a forward lookup, an error message of the form "no IP
1186 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1187 could be logged. Now it outputs the IP address instead of "NULL".
1189 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1190 allows the sender and the authenticated sender to be set when
1191 submitting a message from within Exim. Since child_open_exim() is
1192 documented for local_scan(), the new function should be too.
1194 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1195 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1196 results in an empty string is now treated as unset.
1198 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1200 PH/31 Added sender_verify_fail logging option.
1202 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1203 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1204 tidied the source and removed it altogether.
1206 PH/33 When a queue run was abandoned because the load average was too high, a
1207 log line was always written; now it is written only if the queue_run log
1208 selector is set. In addition, the log line for abandonment now contains
1209 information about the queue run such as the pid. This is always present
1210 in "start" and "stop" lines but was omitted from the "abandon" line.
1212 PH/34 Omit spaces between a header name and the colon in the error message that
1213 is given when verify = headers_syntax fails (if there are lots of them,
1214 the message gets confusing).
1216 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1217 names, as there are now some PTR records that contain slashes. This check
1218 is only to protect against broken name servers that fall over on strange
1219 characters, so the fact that it applies to all lookups doesn't matter.
1221 PH/36 Now that the new test suite is complete, we can remove some of the
1222 special code in Exim that was needed for the old test suite. For example,
1223 sorting DNS records because real resolvers return them in an arbitrary
1224 order. The new test suite's fake resolver always returns records in the
1227 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1228 bounces) except when queue_only is set, to avoid logging races between
1229 the different processes.
1231 PH/38 Panic-die if .include specifies a non-absolute path.
1233 PH/39 A tweak to the "H" retry rule from its user.
1235 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1236 a label. They prevented compilation on older perls.
1238 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1239 a warning to be raised on newish perls.
1241 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1242 on queue. Changes to match documented behaviour of showing count of
1243 messages matching specified criteria.
1245 PH/40 Changed the default ident timeout from 30s to 5s.
1247 PH/41 Added support for the use of login_cap features, on those BSD systems
1248 that have them, for controlling the resources used by pipe deliveries.
1250 PH/42 The content-scanning code uses fopen() to create files in which to put
1251 message data. Previously it was not paying any attention to the mode of
1252 the files. Exim runs with umask(0) because the rest of the code creates
1253 files with open(), and sets the required mode explicitly. Thus, these
1254 files were ending up world-writeable. This was not a big issue, because,
1255 being within the spool directory, they were not world-accessible. I have
1256 created a function called modefopen, which takes an additional mode
1257 argument. It sets umask(777), creates the file, chmods it to the required
1258 mode, then resets the umask. All the relevant calls to fopen() in the
1259 content scanning code have been changed to use this function.
1261 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1262 to 24 hours. This avoids potential overflow problems when processing G
1263 and H retry rules. I suspect nobody ever tinkers with this value.
1265 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1267 PH/45 When the plaintext authenticator is running as a client, the server's
1268 challenges are checked to ensure they are valid base64 strings. By
1269 default, the authentication attempt is cancelled if an invalid string is
1270 received. Setting client_ignore_invalid_base64 true ignores these errors.
1271 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1272 they are received. Thus, the responses can be made to depend on the
1273 challenges. If an invalid string is ignored, an empty string is placed in
1276 PH/46 Messages that are created by the autoreply transport now contains a
1277 References: header, in accordance with RFCs 2822 and 3834.
1279 PH/47 Added authenticated_sender_force to the smtp transport.
1281 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1283 PH/49 Installed latest patch from the Sieve maintainer.
1285 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1286 was also set, the appendfile transport was unnecessarily scanning a
1287 directory of message files (e.g. for maildir delivery) to find the count
1288 of files (along with the size), even though it did not need this
1289 information. It now does the scan only if it needs to find either the
1290 size of the count of files.
1292 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1294 PH/52 Two bugs concerned with error handling when the smtp transport is
1297 (i) Exim was not creating retry information for temporary errors given
1298 for individual recipients after the DATA command when the smtp transport
1299 was used in LMTP mode. This meant that they could be retried too
1300 frequently, and not timed out correctly.
1302 (ii) Exim was setting the flag that allows error details to be returned
1303 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1304 recipients that were returned after the DATA command.
1306 PH/53 This is related to PH/52, but is more general: for any failing address,
1307 when detailed error information was permitted to be returned to the
1308 sender, but the error was temporary, then after the final timeout, only
1309 "retry timeout exceeded" was returned. Now it returns the full error as
1310 well as "retry timeout exceeded".
1312 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1313 do this, and (what is worse) MTAs that accept it.
1315 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1316 will now be deprecated.
1318 PH/56 New os.c-cygwin from the Cygwin maintainer.
1320 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1321 formats (previously only available in exim formats via -bpr, -bpru,
1322 and -bpra. Now also available in native and exiqgrep formats)
1324 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1325 with very large, slow to parse queues
1327 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1329 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1331 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1332 responses to authentication challenges, though it was showing the
1333 challenges; (ii) I've removed the CR characters from the debug output for
1336 PH/58 Allow for the insertion of a newline as well as a space when a string
1337 is turned into more than one encoded-word during RFC 2047 encoding. The
1338 Sieve code now uses this.
1340 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1341 data_4xx, lost_connection, tls_required.
1343 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1344 message was being sent as an SMTP response.
1346 PH/61 Add -l and -k options to exicyclog.
1348 PH/62 When verifying, if an address was redirected to one new address, so that
1349 verification continued, and the new address failed or deferred after
1350 having set something in $address_data, the value of $address_data was not
1351 passed back to the ACL. This was different to the case when no
1352 redirection occurred. The value is now passed back in both cases.
1354 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1355 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1356 use login_cap.h, so on its own it isn't the distinguishing feature. The
1357 new name refers directly to the setclassresources() function.
1359 PH/65 Added configuration files for NetBSD3.
1361 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1363 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1364 is preferred over IPv4.
1366 PH/68 The bounce_return_message and bounce_return_body options were not being
1367 honoured for bounces generated during the reception of non-SMTP messages.
1368 In particular, this applied to messages rejected by the ACL. This bug has
1369 been fixed. However, if bounce_return_message is true and bounce_return_
1370 body is false, the headers that are returned for a non-SMTP message
1371 include only those that have been read before the error was detected.
1372 (In the case of an ACL rejection, they have all been read.)
1374 PH/69 The HTML version of the specification is now built in a directory called
1375 spec_html instead of spec.html, because the latter looks like a path with
1376 a MIME-type, and this confuses some software.
1378 PH/70 Catch two compiler warnings in sieve.c.
1380 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1381 function verify_get_ident() calls ip_connect() to connect a socket, but
1382 if the "connect()" function timed out, ip_connect() used to close the
1383 socket. However, verify_get_ident() also closes the socket later, and in
1384 between Exim writes to the log, which may get opened at this point. When
1385 the socket was closed in ip_connect(), the log could get the same file
1386 descriptor number as the socket. This naturally causes chaos. The fix is
1387 not to close the socket in ip_connect(); the socket should be closed by
1388 the function that creates it. There was only one place in the code where
1389 this was missing, in the iplookup router, which I don't think anybody now
1390 uses, but I've fixed it anyway.
1392 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1393 well as to direct DNS lookups. Otherwise the handling of names in host
1394 lists is inconsistent and therefore confusing.
1400 PH/01 Two changes to the default runtime configuration:
1402 (1) Move the checks for relay_from_hosts and authenticated clients from
1403 after to before the (commented out) DNS black list checks.
1405 (2) Add control=submission to the relay_from_hosts and authenticated
1406 clients checks, on the grounds that messages accepted by these
1407 statements are most likely to be submissions.
1409 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1411 (1) Generate an error if the third argument for the ${prvs expansion is
1414 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1417 (3) Reset the variables that are obtained from the first argument of
1418 ${prvscheck and used in the second argument before leaving the code,
1419 because their memory is reclaimed, so using them afterwards may do
1422 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1423 one (it's much easier than Tom thought :-).
1425 (5) Because of (4), we can now allow for the use of $prvscheck_result
1426 inside the third argument.
1428 PH/03 For some reason, the default setting of PATH when running a command from
1429 a pipe transport was just "/usr/bin". I have changed it to
1432 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1433 anything to be listed in the output from -bV.
1435 PH/05 When a filter generated an autoreply, the entire To: header line was
1436 quoted in the delivery log line, like this:
1438 => >A.N.Other <ano@some.domain> <original@ddress> ...
1440 This has been changed so that it extracts the operative address. There
1441 may be more than one such address. If so, they are comma-separated, like
1444 => >ano@some.domain,ona@other.domain <original@ddress> ...
1446 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1447 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1448 not being looked up in the rDNS to get a host name, Exim was showing the
1449 IP address twice in Received: lines, even though the IP addresses were
1450 identical. For example:
1452 Received: from [1.2.3.4] (helo=[1.2.3.4])
1454 However, if the real host name was known, it was omitting the HELO data
1455 if it matched the actual IP address. This has been tidied up so that it
1456 doesn't show the same IP address twice.
1458 PH/07 When both +timestamp and +memory debugging was on, the value given by
1459 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1460 called by the debug printing, thereby overwriting the timestamp buffer.
1461 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1463 PH/08 When the original message was included in an autoreply transport, it
1464 always said "this is a copy of the message, including all the headers",
1465 even if body_only or headers_only was set. It now gives an appropriate
1468 PH/09 Applied a patch from the Sieve maintainer which:
1470 o fixes some comments
1471 o adds the (disabled) notify extension core
1472 o adds some debug output for the result of if/elsif tests
1473 o points to the current vacation draft in the documentation
1474 and documents the missing references header update
1478 o fixes a bug in processing the envelope test (when testing
1479 multiple envelope elements, the last element determinted the
1482 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1483 Electronic Mail") by including:
1485 Auto-submitted: auto-generated
1487 in the messages that it generates (bounce messages and others, such as
1488 warnings). In the case of bounce messages for non-SMTP mesages, there was
1489 also a typo: it was using "Auto_submitted" (underscore instead of
1490 hyphen). Since every message generated by Exim is necessarily in response
1491 to another message, thes have all been changed to:
1493 Auto-Submitted: auto-replied
1495 in accordance with these statements in the RFC:
1497 The auto-replied keyword:
1499 - SHOULD be used on messages sent in direct response to another
1500 message by an automatic process,
1502 - MUST NOT be used on manually-generated messages,
1504 - MAY be used on Delivery Status Notifications (DSNs) and Message
1505 Disposition Notifications (MDNs),
1507 - MUST NOT be used on messages generated by automatic or periodic
1508 processes, except for messages which are automatic responses to
1511 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1512 to the default Received: header definition.
1514 PH/12 Added log selector acl_warn_skipped (default on).
1516 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1517 variables because (a) they are in the wrong storage pool and (b) even if
1518 they were copied, it wouldn't work properly because of the caching.
1520 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1521 checking when decoding. Apparently there are clients that generate
1522 overlong encoded strings. Why am I not surprised?
1524 PH/15 If the first argument of "${if match_address" was not empty, but did not
1525 contain an "@" character, Exim crashed. Now it writes a panic log message
1526 and treats the condition as false.
1528 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1530 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1531 extension "envelope-auth". The code is finished and in agreement with
1532 other implementations, but there is no documentation so far and in fact,
1533 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1534 not changing the active code.
1536 Print executed "if" and "elsif" statements when debugging is used. This
1537 helps a great deal to understand what a filter does.
1539 Document more things not specified clearly in RFC3028. I had all this
1540 sorted out, when out of a sudden new issues came to my mind. Oops."
1542 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1545 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1546 canonical form (as documented). However, after a host name lookup from
1547 the IP address, check_host() was doing a simple string comparison with
1548 addresses acquired from the DNS when checking that the found name did
1549 have the original IP as one of its addresses. Since any found IPv6
1550 addresses are likely to be in abbreviated form, the comparison could
1551 fail. Luckily, there already exists a function for doing the comparison
1552 by converting both addresses to binary, so now that is used instead of
1553 the text comparison.
1555 PH/20 There was another similar case to PH/19, when a complete host name was
1556 given in a host list; looking up its IP address could give an abbreviated
1557 form, whereas the current host's name might or might not be abbreviated.
1558 The same fix has been applied.
1564 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1565 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1568 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1569 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1571 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1573 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1574 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1575 Also, for "Auto-Submitted", treat anything other than "no" as
1576 non-personal, in accordance with RFC 3834. (Previously it treated
1577 anything starting "auto-" as non-personal.)
1579 TF/01 The control=submission/name=... option had a problem with syntax
1580 errors if the name included a slash character. The /name= option
1581 now slurps the rest of the string, so it can include any characters
1582 but it must come last in the list of options (after /sender_retain
1585 PH/05 Some modifications to the interface to the fake nameserver for the new
1593 TK/01 Added the "success_on_redirect" address verification option. See
1594 NewStuff for rationale and an example.
1596 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1598 PH/02 Patch to exigrep to allow it to work on syslog lines.
1600 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1601 fread() to skip over the body file's header line, because in Cygwin the
1602 header line is locked and is inaccessible.
1604 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1605 co-exist for some time) to make it clear that it is the Exim ID that is
1606 referenced, not the Message-ID: header line.
1608 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1609 string_format() function, because snprintf() does not exist on all
1612 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1613 ${stat: expansion item.
1615 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1616 protocol synchronization error", to keep the pedants happy.
1618 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1619 well as for IRIX systems, when gcc is being used. See the host.c source
1622 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1624 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1627 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1628 transport and to the smtp transport in LMTP mode.
1630 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1632 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1633 run a filter in a subprocess. This could lead to confusion in subsequent
1634 lookups in the parent process. There should also be a search_tidyup() at
1635 the end of the subprocess.
1637 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1638 only if the host matched helo_try_verify_hosts, which caused the
1639 verification to occur when the EHLO/HELO command was issued. The ACL just
1640 tested the remembered result. Now, if a previous verification attempt has
1641 not happened, "verify = helo" does it there and then.
1643 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1645 TK/03 Fix log output including CR from clamd.
1647 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1648 exist provoked a memory error which could cause a segfault.
1650 PH/15 Installed PCRE 6.2
1652 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1654 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1655 of the problem. Specifically, suggested +O2 rather than +O1 for the
1658 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1660 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1661 record, the port was not correctly logged when the outgoing_port log
1662 selector was set (it logged the transort's default port).
1664 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1665 fallback_hosts, and "hosts" in the smtp transport.
1667 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1668 host errors such as "Connection refused".
1670 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1671 authentication with radiusclient 0.4.9:
1673 - Error returned from rc_read_config was caught wrongly
1674 - Username/password not passed on to radius server due to wrong length.
1676 The presumption is that some radiusclient API changes for 4.51/PH/17
1677 were not taken care of correctly. The code is still untested by me (my
1678 Linux distribution still has 0.3.2 of radiusclient), but it was
1679 contributed by a Radius user.
1681 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1682 expanding the "port" option of the smtp transport.
1684 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1685 while reading a MIME header. Thanks to Tom Hughes for a patch.
1687 PH/24 Include config.h inside local_scan.h so that configuration settings are
1690 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1691 that in an ACL for RCPT (for example), you can examine exactly what was
1694 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1695 commands, but it was not correctly comparing the address with the actual
1696 client host address. Thus, it would show the EHLO address in Received:
1697 header lines when this was not necessary.
1699 PH/27 Added the % operator to ${eval:}.
1701 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1702 it should be ignoring failures (because with -C, for example, it has lost
1703 privilege). It wasn't ignoring creation failures other than "already
1706 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1709 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1710 because an input line was too long, either on its own, or by virtue of
1711 too many continuations, the temporary file was not being removed, and the
1712 return code was incorrect.
1714 PH/31 Missing "BOOL" in function definition in filtertest.c.
1716 PH/32 Applied Sieve patches from the maintainer.
1718 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1720 PH/33 Added "verify = not_blind".
1722 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1723 Local/Makefile (with some defaults set). These are used in built scripts
1724 such as exicyclog, but they have never been used in the exim_install
1725 script (though there are many overriding facilities there). I have
1726 arranged that the exim_install script now takes note of these two
1729 PH/35 Installed configuration files for Dragonfly.
1731 PH/36 When a locally submitted message by a trusted user did not contain a
1732 From: header, and the sender address was obtained from -f or from an SMTP
1733 MAIL command, and the trusted user did not use -F to supply a sender
1734 name, $originator_name was incorrectly used when constructing a From:
1735 header. Furthermore, $originator_name was used for submission mode
1736 messages from external hosts without From: headers in a similar way,
1737 which is clearly wrong.
1739 PH/37 Added control=suppress_local_fixups.
1741 PH/38 When log_selector = +received_sender was set, and the addition of the
1742 sender made the log line's construction buffer exactly full, or one byte
1743 less than full, an overflow happened when the terminating "\n" was
1746 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1747 when the result of a list match is failure because a DNS lookup failed.
1749 PH/40 RM_COMMAND is now used in the building process.
1751 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1752 the "build-* directories that it finds.
1754 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1755 address in a domain literal was a prefix of an interface address.
1757 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1758 when verifying a sender address, unless rewrite_headers is false.
1760 PH/44 Wrote a long comment about why errors_to addresses are verified as
1761 recipients, not senders.
1763 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1764 the ratelimit ACL was added.
1766 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1768 PH/46 Added extra information about PostgreSQL errors to the error string.
1770 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1771 avoiding the need to install special zones in a real server. This is
1772 backwards compatible; if it can't find the fake resolver, it drops back.
1773 Thus, both old and new test suites can be run.
1775 TF/02 Added util/ratelimit.pl
1777 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1778 clock is set back in time.
1780 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1781 Candler <B.Candler@pobox.com>.
1783 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1784 OK for addresses that are the result of redirections.
1786 PH/48 A number of further additions for the benefit of the new test suite,
1787 including a fake gethostbyname() that interfaces to the fake DNS resolver
1790 TF/06 The fix for widen_domains has also been applied to qualify_single and
1791 search_parents which are the other dnslookup options that can cause
1794 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1797 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1799 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1800 service" if it encountered an MX record with an empty target hostname.
1801 The message is now "an MX or SRV record indicated no SMTP service".
1803 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1804 if the DNS of the sending site is misconfigured. This is quite a
1805 common situation. This change restores the behaviour of treating a
1806 helo verification defer as a failure.
1808 PH/51 If self=fail was set on a router, the bounce message did not include the
1809 actual error message.
1815 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1817 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1818 command itself ended in error, the underlying message about the transport
1819 filter timeout was being overwritten with the pipe command error. Now the
1820 underlying error message should be appended to the second error message.
1822 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1824 PH/02 Reduce the amount of output that "make" produces by default. Full output
1825 can still be requested.
1827 PH/03 The warning log line about a condition test deferring for a "warn" verb
1828 was being output only once per connection, rather than after each
1829 occurrence (because it was using the same function as for successful
1830 "warn" verbs). This seems wrong, so I have changed it.
1832 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1833 it should not have, which might have caused a crash in the right
1834 circumstances, but probably never did.
1836 PH/04 Installed a modified version of Tony Finch's patch to make submission
1837 mode fix the return path as well as the Sender: header line, and to
1838 add a /name= option so that you can make the user's friendly name appear
1841 TF/03 Added the control = fakedefer ACL modifier.
1843 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1844 Mark Lowes for thorough testing.
1846 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1848 TK/03 Merged latest SRS patch from Miles Wilton.
1850 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1851 with the definition in sysexits.h (which is #included earlier).
1852 Fortunately, Exim does not actually use EX_OK. The code used to try to
1853 preserve the sysexits.h value, by assumimg that macro definitions were
1854 scanned for macro replacements. I have been disabused of this notion,
1855 so now the code just undefines EX_OK before #including unistd.h.
1857 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1858 in the smtp transport. When a block could not be written in a single
1859 write() function, the timeout was being re-applied to each part-write.
1860 This seems wrong - if the receiver was accepting one byte at a time it
1861 would take for ever. The timeout is now adjusted when this happens. It
1862 doesn't have to be particularly precise.
1864 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1865 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1867 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1868 without a domain if the check to <postmaster@domain> fails.
1870 SC/01 Eximstats: added -xls and the ability to specify output files
1871 (patch written by Frank Heydlauf).
1873 SC/02 Eximstats: use FileHandles for outputing results.
1875 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1877 SC/04 Eximstats: fixed display of large numbers with -nvr option
1879 SC/05 Eximstats: fixed merging of reports with empty tables.
1881 SC/06 Eximstats: added the -include_original_destination flag
1883 SC/07 Eximstats: removed tabs and trailing whitespace.
1885 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1887 TK/06 MBOX spool code: Add real "From " MBOX separator line
1888 so the .eml file is really in mbox format (even though
1889 most programs do not really care). Patch from Alex Miller.
1891 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1892 The latter is generated from $received_to and is only set if the
1893 message has one envelope recipient. SA can use these headers,
1894 obviously out-of-the-box. Patch from Alex Miller.
1896 PH/08 The ${def test on a variable was returning false if the variable's
1897 value was "0", contrary to what the specification has always said!
1898 The result should be true unless the variable is empty.
1900 PH/09 The syntax error of a character other than { following "${if
1901 def:variable_name" (after optional whitespace) was not being diagnosed.
1902 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1903 accidental colon was present, for example, could give incorrect results.
1905 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1906 result is used (not including appendfile, where other changes are about
1909 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1910 This involved changing a lot of size variables from int to off_t. It
1911 should work with maildirs and everything.
1913 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1914 spamd dying while we are connected to it.
1916 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1919 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1920 function for BDB 4.3.
1922 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1924 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1925 the sources for PCRE have been re-arranged and more files are now
1928 PH/15 The code I had for printing potentially long long variables in PH/11
1929 above was not the best (it lost precision). The length of off_t variables
1930 is now inspected at build time, and an appropriate printing format (%ld
1931 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1932 to be "long long int" or "long int". This is needed for the internal
1933 formatting function string_vformat().
1935 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1936 the configuration file to be ":syslog", then the script "guesses" where
1937 the logs files are, rather than using the compiled in default. In our
1938 case the guess is not the same as the compiled default, so the script
1939 suddenly stopped working when I started to use syslog. The patch checks
1940 to see if log_file_path is "". If so, it attempts to read it from exim
1941 with no configuration file to get the compiled in version, before it
1942 falls back to the previous guessing code."
1944 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1945 implementing BATV in an Exim configuration. See NewStuff for the gory
1948 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1949 Makefile that are specific to HP-UX.
1951 PH/18 If the "use_postmaster" option was set for a recipient callout together
1952 with the "random" option, the postmaster address was used as the MAIL
1953 FROM address for the random test, but not for the subsequent recipient
1954 test. It is now used for both.
1956 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1957 patch removes a few documentation additions to RFC 3028, because the
1958 latest draft now contains them. It adds the new en;ascii-case comparator
1959 and a new error check for 8bit text in MIME parts. Comparator and
1960 require names are now matched exactly. I enabled the subaddress
1961 extension, but it is not well tested yet (read: it works for me)."
1963 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1964 rework some of the code of TK/09 above to avoid the hardwired use of
1965 "%lld" and "long long". Replaced the call to snprintf() with a call to
1968 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1969 records point to non-existent hosts", "retry timeout exceeded", and
1970 "retry time not reached for any host after a long failure period".
1972 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1973 experimental DomainKeys support:
1975 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1976 (2) On an error such as an illegally used "control", the wrong name for
1977 the control was given.
1979 These problems did NOT occur unless DomainKeys support was compiled.
1981 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1983 PH/24 Added ${if match_ip condition.
1985 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1986 databases so that it will be absolutely obvious if a crash occurs in the
1987 DB library. This is a regular occurrence (often caused by mis-matched
1990 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1991 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1992 picked up on a user's system that detects such things. There doesn't seem
1993 to be a gcc warning option for this - only an attribute that has to be
1994 put on the function's prototype. It seems that in Fedora Core 4 they have
1995 set this on a number of new functions. No doubt there will be more in due
1998 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1999 specify a transport. However, if an address that was verified by such a
2000 router was the subject of a callout, Exim crashed because it tried to
2001 read the rcpt_include_affixes from the non-existent transport. Now it
2002 just assumes that the setting of that option is false. This bug was
2003 introduced by 4.51/PH/31.
2005 PH/28 Changed -d+all to exclude +memory, because that information is very
2006 rarely of interest, but it makes the output a lot bigger. People tend to
2007 do -d+all out of habit.
2009 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2010 code in os-type was giving problems when libc.so lives in lib64, like on
2013 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2014 aren't the modern standard, and it seems that some systems' include files
2015 don't always have them. Exim was already checking for some of the newer
2016 ones like T_AAAA, and defining it itself. I've added checks for all the
2017 record types that Exim uses.
2019 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2020 not automatically generating a new one, as it is supposed to. This
2021 prevented TLS from working. If the file did exist, but contained invalid
2022 data, a new version was generated, as expected. It was only the case of a
2023 non-existent file that was broken.
2025 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2026 with a change in libdomainkeys > 0.64.
2028 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2029 from DNS. If the selector record carries the flag, it now has
2030 precedence over the domain-wide flag.
2032 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2034 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2035 the use of a port name that isn't defined in /etc/services) occurred, the
2036 message was deferred as in a normal delivery, and thus remained on the
2037 spool, instead of being failed because of the mua_wrapper setting. This
2038 is now fixed, and I tidied up some of the mua_wrapper messages at the
2041 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2042 the messages in a hash of arrays rather than using individual hashes.
2043 This is a bit cleaner and results in dramatic memory savings, albeit
2044 at a slight CPU cost.
2046 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2047 as requested by Marc Sherman.
2049 SC/10 Eximstats: added histograms for user specified patterns as requested
2052 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2054 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2055 fopen() in the content-scanning modules that did not already have it.
2061 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2062 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2064 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2066 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2068 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2069 to test Sieve filters that use "vacation".
2071 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2072 that changes the way the GnuTLS parameters are stored in the cache file.
2073 The new format can be generated externally. For backward compatibility,
2074 if the data in the cache doesn't make sense, Exim assumes it has read an
2075 old-format file, and it generates new data and writes a new file. This
2076 means that you can't go back to an older release without removing the
2079 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2080 work if there are any delivery delays because "one_time" forces the
2081 parent to be marked "delivered", so its unseen clone is never tried
2082 again. For this reason, Exim now forbids the simultaneous setting of
2085 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2086 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2087 entry if you want to know the details. The fix, however, neglected to
2088 consider the case when local delivery batching is involved. The test for
2089 "previously delivered" was not happening when checking to see if an
2090 address could be batched with a previous (undelivered) one; under
2091 certain circumstances this could lead to multiple deliveries to the same
2094 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2095 in its include files, and this causes problems building Exim.
2097 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2098 header_syntax) but Exim was just ignoring anything given after a slash.
2099 In particular, this caused confusion with an attempt to use "verify =
2100 reverse_host_lookup/defer_ok". An error is now given when options are
2101 supplied for verify items that do not have them. (Maybe reverse_host_
2102 lookup should have a defer_ok option, but that's a different point.)
2104 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2105 defined by RFC 821) to 2048, because there were problems with some AUTH
2106 commands, and RFC 1869 says the size should be increased for extended
2107 SMTP commands that take arguments.
2109 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2112 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2113 "unknown" error; now it says that the functionality isn't in the binary.
2115 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2116 an address' error message when a string expansion fails (syntax or
2117 whatever). Otherwise the password may appear in the log. Following change
2118 PH/42 below, there is no longer a chance of it appearing in a bounce
2121 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2123 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2124 of its addresses was ever tried. (Bugzilla bug #2.)
2126 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2127 the result incorrectly in the debug output. (It correctly added a newline
2128 to what was transported.)
2130 TF/01 Added $received_time.
2132 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2133 commented out examples of how to interface to a virus scanner and to
2134 SpamAssassin. Also added commented examples of av_scanner and
2135 spamd_address settings.
2137 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2138 and controls are allowed in which ACLs. There were a couple of minor
2139 errors. Some of the entries in the conditions table (which is a table of
2140 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2141 negation of where the condition IS allowed.
2143 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2145 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2146 header file does not have a version number, so I've had to invent a new
2147 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2148 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2149 radiusclient), but it was contributed by a Radius user.
2151 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2152 files or directories, for OpenSSL.
2154 PH/19 When an Exim process that is running as root has to create an Exim log
2155 file, it does so in a subprocess that runs as exim:exim so as to get the
2156 ownership right at creation (otherwise, other Exim processes might see
2157 the file with the wrong ownership). There was no test for failure of this
2158 fork() call, which would lead to the process getting stuck as it waited
2159 for a non-existent subprocess. Forks do occasionally fail when resources
2160 run out. I reviewed all the other calls to fork(); they all seem to check
2163 PH/20 When checking for unexpected SMTP input at connect time (before writing
2164 the banner), Exim was not dealing correctly with a non-positive return
2165 from the read() function. If the client had disconnected by this time,
2166 the result was a log entry for a synchronization error with an empty
2167 string after "input=" when read() returned zero. If read() returned -1
2168 (an event I could not check), uninitialized data bytes were printed.
2169 There were reports of junk text (parts of files, etc) appearing after
2172 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2174 PH/22 Added support for macro redefinition, and (re)definition in between
2175 driver and ACL definitions.
2177 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2178 forgetting to use the resulting value; it was using the unexpanded value.
2180 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2181 hadn't been configured. The fix is from Juergen Kreileder, who
2182 understands it better than I do:
2184 "Here's what I see happening with three configured cyrus_sasl
2185 authenticators configured (plain, login, cram-md5):
2187 On startup auth_cyrus_sasl_init() gets called for each of these.
2188 This means three calls to sasl_listmech() without a specified mech_list.
2189 => SASL tests which mechs of all available mechs actually work
2190 => three warnings about OTP not working
2191 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2193 With the patch, sasl_listmech() also gets called three times. But now
2194 SASL's mech_list option is set to the server_mech specified in the the
2195 authenticator. Or in other words, the answer from sasl_listmech()
2196 gets limited to just the mech you're testing for (which is different
2198 => the return list contains just 'plain' or 'login', 'cram-md5' or
2199 nothing depending on the value of ob->server_mech.
2201 I've just tested the patch: Authentication still works fine,
2202 unavailable mechs specified in the exim configuration are still
2203 caught, and the auth.log warnings about OTP are gone."
2205 PH/25 When debugging is enabled, the contents of the command line are added
2206 to the debugging output, even when log_selector=+arguments is not
2209 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2210 answer is "GNU", and only if the return is "GNU/something" is the answer
2213 PH/27 $acl_verify_message is now set immediately after the failure of a
2214 verification in an ACL, and so is available in subsequent modifiers. In
2215 particular, the message can be preserved by coding like this:
2217 warn !verify = sender
2218 set acl_m0 = $acl_verify_message
2220 Previously, $acl_verify_message was set only while expanding "message"
2221 and "log_message" when a very denied access.
2223 PH/28 Modified OS/os.c-Linux with
2225 -#ifndef OS_LOAD_AVERAGE
2226 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2228 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2229 nomenclature these days.)
2231 PH/29 Installed patch from the Sieve maintainer that adds the options
2232 sieve_useraddress and sieve_subaddress to the redirect router.
2234 PH/30 In these circumstances:
2235 . Two addresses routed to the same list of hosts;
2236 . First host does not offer TLS;
2237 . First host accepts first address;
2238 . First host gives temporary error to second address;
2239 . Second host offers TLS and a TLS session is established;
2240 . Second host accepts second address.
2241 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2242 and peerdn, if requested) that were in fact used only for the second
2245 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2246 attention to any local part prefix or suffix that was matched by the
2247 router that accepted the address. It now behaves in the same way as it
2248 does for delivery: the affixes are removed from the local part unless
2249 rcpt_include_affixes is set on the transport.
2251 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2252 timeout during the DATA phase of an incoming message.
2254 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2255 applied a patch sanctioned by the Sieve maintainer.
2257 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2258 the uid or gid is negative. A case of a negative gid caused this to be
2259 noticed. The fix allows for either to be negative.
2261 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2262 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2265 PH/36 The change PH/12 above was broken. Fixed it.
2267 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2268 the grounds that routing the same address twice would always produce the
2269 same answer. This might have been true once, but it is certainly no
2270 longer true now. Routing a child address may depend on the previous
2271 routing that produced that child. Some complicated redirection strategies
2272 went wrong when messages had multiple recipients, and made Exim's
2273 behaviour dependent on the order in which the addresses were given.
2275 I have moved the duplicate checking until after the routing is complete.
2276 Exim scans the addresses that are assigned to local and remote
2277 transports, and removes any duplicates. This means that more work will be
2278 done, as duplicates will always all be routed, but duplicates are
2279 presumably rare, so I don't expect this is of any significance.
2281 For deliveries to pipes, files, and autoreplies, the duplicate checking
2282 still happens during the routing process, since they are not going to be
2285 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2286 It corrects a timeout issue with spamd. This is Ian's comment: "The
2287 background is that sometimes spamd either never reads data from a
2288 connection it has accepted, or it never writes response data. The exiscan
2289 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2290 blindly assumes that writes won't block so it may never time out."
2292 PH/39 Allow G after quota size as well as K and M.
2294 PH/40 The value set for $authenticated_id in an authenticator may not contain
2295 binary zeroes or newlines because the value is written to log lines and
2296 to spool files. There was no check on this. Now the value is run through
2297 the string_printing() function so that such characters are converted to
2298 printable escape sequences.
2300 PH/41 $message_linecount is a new variable that contains the total number of
2301 lines in the message. Compare $body_linecount, which is the count for the
2304 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2305 bounce and delay warning messages, except in certain special cases, which
2308 (a) An SMTP error message from a remote host;
2309 (b) A message specified in a :fail: redirection;
2310 (c) A message specified in a "fail" command in a system filter;
2311 (d) A message specified in a FAIL return from the queryprogram router;
2312 (e) A message specified by the cannot_route_message router option.
2314 In these cases only, Exim does include the error details in bounce and
2315 warning messages. There are also a few cases where bland messages such
2316 as "unrouteable address" or "local delivery error" are given.
2318 PH/43 $value is now also set for the "else" part of a ${run expansion.
2320 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2321 being worked on, but at least Exim now implements the latest version to
2324 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2325 process to complete was treated as a delivery failure, a timeout while
2326 writing the message to the pipe was logged, but erroneously treated as a
2327 successful delivery. Such timeouts include transport filter timeouts. For
2328 consistency with the overall process timeout, these timeouts are now
2329 treated as errors, giving rise to delivery failures by default. However,
2330 there is now a new Boolean option for the pipe transport called
2331 timeout_defer, which, if set TRUE, converts the failures into defers for
2332 both kinds of timeout. A transport filter timeout is now identified in
2335 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2336 systems where "make" and "gmake" are different, calling "gmake" at top
2337 level broke things. I've arranged for the value of $(MAKE) to be passed
2338 from the Makefile to this script so that it can call the same version of
2342 A note about Exim versions 4.44 and 4.50
2343 ----------------------------------------
2345 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2346 changes of various kinds. As a consequence, a big documentation update was
2347 needed. This delayed the release for rather longer than seemed good, especially
2348 in the light of a couple of (minor) security issues. Therefore, the changes
2349 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2350 release. So 4.44 and 4.50 are in effect two different branches that both start
2353 I have left the 4.50 change log unchanged; it contains all the changes since
2354 4.43. The change log for 4.44 is below; many of its items are identical to
2355 those for 4.50. This seems to be the most sensible way to preserve the
2356 historical information.
2362 1. Minor wording change to the doc/README.SIEVE file.
2364 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2365 computation of the current number of files was incorrect.
2367 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2368 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2371 4. Give more explanation in the error message when the command for a transport
2372 filter fails to execute.
2374 5. There are several places where Exim runs a non-Exim command in a
2375 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2376 was being done only for the command run by the queryprogram router. It is
2377 now done for all such subprocesses. The other cases are: ${run, transport
2378 filters, and the commands run by the lmtp and pipe transports.
2380 6. Added CONFIGURE_GROUP build-time option.
2382 7. Some older OS have a limit of 256 on the maximum number of file
2383 descriptors. Exim was using setrlimit() to set 1000 as a large value
2384 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2385 systems. I've change it so that if it can't get 1000, it tries for 256.
2387 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2388 was an oversight, and furthermore, ever since the addition of extra
2389 controls (e.g. 4.43/32), the checks on when to allow different forms of
2390 "control" were broken. There should now be diagnostics for all cases when a
2391 control that does not make sense is encountered.
2393 9. Added the /retain_sender option to "control=submission".
2395 10. $recipients is now available in the predata ACL (oversight).
2397 11. Tidy the search cache before the fork to do a delivery from a message
2398 received from the command line. Otherwise the child will trigger a lookup
2399 failure and thereby defer the delivery if it tries to use (for example) a
2400 cached ldap connection that the parent has called unbind on.
2402 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2403 of $address_data from the recipient verification was clobbered by the
2404 sender verification.
2406 13. The value of address_data from a sender verification is now available in
2407 $sender_address_data in subsequent conditions in the ACL statement.
2409 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2411 15. Added a new option "connect=<time>" to callout options, to set a different
2414 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2415 was its contents. (It was OK if the option was not defined at all.)
2417 17. A "Completed" log line is now written for messages that are removed from
2418 the spool by the -Mrm option.
2420 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2421 information about exactly what failed.
2423 19. Added -dd to debug only the daemon process.
2425 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2426 handles timeouts, both on the server side and network timeouts. Renamed the
2427 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2429 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2430 It is now set to "smtps".
2432 22. $host_address is now set to the target address during the checking of
2433 ignore_target_hosts.
2435 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2436 being passed; this would have caused $sender_host_name to have been used if
2437 matching the list had actually called for a host name (not very likely,
2438 since this list is usually IP addresses). A host name is now passed as
2441 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2442 code that specifies a non-restarting handler (typically sigaction() in
2443 modern systems) in an attempt to fix a rare and obscure crash bug.
2445 25. Narrowed the window for a race in the daemon that could cause it to ignore
2446 SIGCHLD signals. This is not a major problem, because they are used only to
2447 wake it up if nothing else does.
2449 26. A malformed maildirsize file could cause Exim to calculate negative values
2450 for the mailbox size or file count. Odd effects could occur as a result.
2451 The maildirsize information is now recalculated if the size or filecount
2454 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2455 support for a long time. Removed HAVE_SYS_VFS_H.
2457 28. Installed the latest version of exipick from John Jetmore.
2459 29. In an address list, if the pattern was not a regular expression, an empty
2460 subject address (from a bounce message) matched only if the pattern was an
2461 empty string. Non-empty patterns were not even tested. This was the wrong
2462 because it is perfectly reasonable to use an empty address as part of a
2463 database query. An empty address is now tested by patterns that are
2464 lookups. However, all the other forms of pattern expect the subject to
2465 contain a local part and a domain, and therefore, for them, an empty
2466 address still always fails if the pattern is not itself empty.
2468 30. Exim went into a mad DNS loop when attempting to do a callout where the
2469 host was specified on an smtp transport, and looking it up yielded more
2470 than one IP address.
2472 31. Re-factored the code for checking spool and log partition space into a
2473 function that finds that data and another that does the check. The former
2474 is then used to implement four new variables: $spool_space, $log_space,
2475 $spool_inodes, and $log_inodes.
2477 32. The RFC2047 encoding function was originally intended for short strings
2478 such as real names; it was not keeping to the 75-character limit for
2479 encoded words that the RFC imposes. It now respects the limit, and
2480 generates multiple encoded words if necessary. To be on the safe side, I
2481 have increased the buffer size for the ${rfc2047: expansion operator from
2484 33. It is now permitted to omit both strings after an "if" condition; if the
2485 condition is true, the result is "true". As before, when the second string
2486 is omitted, a false condition yields an empty string. This makes it less
2487 cumbersome to write custom ACL and router conditions.
2489 34. Failure to deliver a bounce message always caused it to be frozen, even if
2490 there was an errors_to setting on the router. The errors_to setting is now
2493 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2494 canonical form (fully expanded) before being placed in
2495 $sender_host_address.
2497 36. The table in the code that translates DNS record types into text (T_A to
2498 "A" for instance) was missing entries for NS and CNAME. It is just possible
2499 that this could have caused confusion if both these types were looked up
2500 for the same domain, because the text type is used as part of Exim's
2501 per-process caching. But the chance of anyone hitting this buglet seems
2504 37. The dnsdb lookup has been extended in a number of ways.
2506 (1) There is a new type, "zns", which walks up the domain tree until it
2507 finds some nameserver records. It should be used with care.
2509 (2) There is a new type, "mxh", which is like "mx" except that it returns
2510 just the host names, not the priorities.
2512 (3) It is now possible to give a list of domains (or IP addresses) to be
2513 looked up. The behaviour when one of the lookups defers can be
2514 controlled by a keyword.
2516 (4) It is now possible to specify the separator character for use when
2517 multiple records are returned.
2519 38. The dnslists ACL condition has been extended: it is now possible to supply
2520 a list of IP addresses and/or domains to be looked up in a particular DNS
2523 39. Added log_selector=+queue_time_overall.
2525 40. When running the queue in the test harness, wait just a tad after forking a
2526 delivery process, to get repeatability of debugging output.
2528 41. Include certificate and key file names in error message when GnuTLS fails
2529 to set them up, because the GnuTLS error message doesn't include the name
2530 of the failing file when there is a problem reading it.
2532 42. Allow both -bf and -bF in the same test run.
2534 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2536 44. The "Exiscan patch" is now merged into the mainline Exim source.
2538 45. Sometimes the final signoff response after QUIT could fail to get
2539 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2540 < 0 before doing a fflush(). This bug looks as though it goes back to the
2541 introduction of TLS in release 3.20, but "sometimes" must have been rare
2542 because the tests only now provoked it.
2544 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2545 (this can affect the format of dates).
2547 47. exim_tidydb, when checking for the continued existence of a message for
2548 which it has found a message-specific retry record, was not finding
2549 messages that were in split spool directories. Consequently, it was
2550 deleting retry records that should have stayed in existence.
2552 48. Steve fixed some bugs in eximstats.
2554 49. The SPA authentication driver was not abandoning authentication and moving
2555 on to the next authenticator when an expansion was forced to fail,
2556 contradicting the general specification for all authenticators. Instead it
2557 was generating a temporary error. It now behaves as specified.
2559 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2560 (the order specifies the preference for clients). The order is now AES256,
2561 AES128, 3DES, ARCFOUR128.
2563 51. Small patch to Sieve code - explicitly set From: when generating an
2566 52. Exim crashed if a remote delivery caused a very long error message to be
2567 recorded - for instance if somebody sent an entire SpamAssassin report back
2568 as a large number of 550 error lines. This bug was coincidentally fixed by
2569 increasing the size of one of Exim's internal buffers (big_buffer) that
2570 happened as part of the Exiscan merge. However, to be on the safe side, I
2571 have made the code more robust (and fixed the comments that describe what
2574 53. Now that there can be additional text after "Completed" in log lines (if
2575 the queue_time_overall log selector is set), a one-byte patch to exigrep
2576 was needed to allow it to recognize "Completed" as not the last thing in
2579 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2580 patch that reportedly fixes this has been added. I am not expert enough to
2581 create a test for it. This is what the patch creator wrote:
2583 "I found a little strange behaviour of ldap code when working with
2584 Windows 2003 AD Domain, where users was placed in more than one
2585 Organization Units. When I tried to give exim partial DN, the exit code
2586 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2587 But simultaneously result of request was absolutely normal ldap result,
2588 so I produce this patch..."
2590 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2591 REFERENCE, so I have modified the code to exclude the patch when that macro
2594 55. Some experimental protocols are using DNS PTR records for new purposes. The
2595 keys for these records are domain names, not reversed IP addresses. The
2596 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2597 leaves it alone. Component reversal etc. now happens only for IP addresses.
2600 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2602 57. Double the size of the debug message buffer (to 2048) so that more of very
2603 long debug lines gets shown.
2605 58. The exicyclog utility now does better if the number of log files to keep
2606 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2608 59. Two changes related to the smtp_active_hostname option:
2610 (1) $smtp_active_hostname is now available as a variable.
2611 (2) The default for smtp_banner uses $smtp_active_hostname instead
2612 of $primary_hostname.
2614 60. The host_aton() function is supposed to be passed a string that is known
2615 to be a valid IP address. However, in the case of IPv6 addresses, it was
2616 not checking this. This is a hostage to fortune. Exim now panics and dies
2617 if the condition is not met. A case was found where this could be provoked
2618 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2619 components; fortuitously, this particular loophole had already been fixed
2620 by change 4.50/55 above.
2622 If there are any other similar loopholes, the new check in host_aton()
2623 itself should stop them being exploited. The report I received stated that
2624 data on the command line could provoke the exploit when Exim was running as
2625 exim, but did not say which command line option was involved. All I could
2626 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2627 running as the user.
2630 61. There was a buffer overflow vulnerability in the SPA authentication code
2631 (which came originally from the Samba project). I have added a test to the
2632 spa_base64_to_bits() function which I hope fixes it.
2635 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2636 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2638 63. The daemon start-up calls getloadavg() while still root for those OS that
2639 need the first call to be done as root, but it missed one case: when
2640 deliver_queue_load_max is set with deliver_drop_privilege. This is
2641 necessary for the benefit of the queue runner, because there is no re-exec
2642 when deliver_drop_privilege is set.
2644 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2645 This has been fixed.
2647 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2648 was in use, was not putting the data itself into the right store pool;
2649 consequently, it could be overwritten for a subsequent message in the same
2650 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2653 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2655 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2656 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2657 the function were treating the return as a boolean value, which happened to
2658 work because 0=false and not-0=true, but is not correct code.
2660 68. The host_aton() function was not handling scoped IPv6 addresses (those
2661 with, for example, "%eth0" on the end) correctly.
2663 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2664 negated items (that is, ~something) in unsigned ints. Some compilers
2665 apparently mutter when there is no cast.
2667 70. If an address verification called from an ACL failed, and did not produce a
2668 user-specific message (i.e. there was only a "system" message), nothing was
2669 put in $acl_verify_message. In this situation, it now puts the system
2672 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2673 session; change 4.31/43 added the unwanted input to the log line - except
2674 that it did not do this in the start of session case. It now does.
2676 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2677 This is wrong and can cause the other end to generate a synchronization
2678 error if it is another Exim or anything else that does the synchronization
2679 check. A QUIT command is no longer sent after a timeout.
2681 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2682 during host lookups.
2684 74. The defer_ok option of callout verification was not working if it was used
2685 when verifying addresses in header lines, that is, for this case:
2687 verify = header_sender/callout=defer_ok
2689 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2690 those file descriptors could be used for SMTP connections. If anything
2691 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2692 could be sent to the SMTP client, causing chaos. The daemon now opens
2693 stdin, stdout, and stderr to /dev/null when it puts itself into the
2696 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2697 log by default. The user can override this with suitable Perl magic.
2699 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2700 the log message when discard triggers, was not working for the DATA ACL or
2701 for the non-SMTP ACL.
2703 78. Error message wording change in sieve.c.
2705 79. If smtp_accept_max_per_host was set, the number of connections could be
2706 restricted to fewer than expected, because the daemon was trying to set up
2707 a new connection before checking whether the processes handling previous
2708 connections had finished. The check for completed processes is now done
2709 earlier. On busy systems, this bug wouldn't be noticed because something
2710 else would have woken the daemon, and it would have reaped the completed
2713 80. If a message was submitted locally by a user whose login name contained one
2714 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2715 It caused a spool format error. I have fixed the spool reading code. A
2716 related problem was that the "from" clause in the Received: line became
2717 illegal because of the space(s). It is now covered by ${quote_local_part.
2719 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2722 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2724 83. Patch from Sieve maintainer to fix unterminated string problem in
2725 "vacation" handling.
2727 84. Some minor changes to the Linux configuration files to help with other
2728 OS variants using glibc.
2730 85. One more patch for Sieve to update vacation handling to latest spec.
2733 ----------------------------------------------------
2734 See the note above about the 4.44 and 4.50 releases.
2735 ----------------------------------------------------
2741 1. Change 4.43/35 introduced a bug that caused file counts to be
2742 incorrectly computed when quota_filecount was set in an appendfile
2745 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2746 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2749 3. Give more explanation in the error message when the command for a transport
2750 filter fails to execute.
2752 4. There are several places where Exim runs a non-Exim command in a
2753 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2754 was being done only for the command run by the queryprogram router. It is
2755 now done for all such subprocesses. The other cases are: ${run, transport
2756 filters, and the commands run by the lmtp and pipe transports.
2758 5. Some older OS have a limit of 256 on the maximum number of file
2759 descriptors. Exim was using setrlimit() to set 1000 as a large value
2760 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2761 systems. I've change it so that if it can't get 1000, it tries for 256.
2763 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2764 was an oversight, and furthermore, ever since the addition of extra
2765 controls (e.g. 4.43/32), the checks on when to allow different forms of
2766 "control" were broken. There should now be diagnostics for all cases when a
2767 control that does not make sense is encountered.
2769 7. $recipients is now available in the predata ACL (oversight).
2771 8. Tidy the search cache before the fork to do a delivery from a message
2772 received from the command line. Otherwise the child will trigger a lookup
2773 failure and thereby defer the delivery if it tries to use (for example) a
2774 cached ldap connection that the parent has called unbind on.
2776 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2777 of $address_data from the recipient verification was clobbered by the
2778 sender verification.
2780 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2781 was its contents. (It was OK if the option was not defined at all.)
2783 11. A "Completed" log line is now written for messages that are removed from
2784 the spool by the -Mrm option.
2786 12. $host_address is now set to the target address during the checking of
2787 ignore_target_hosts.
2789 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2790 being passed; this would have caused $sender_host_name to have been used if
2791 matching the list had actually called for a host name (not very likely,
2792 since this list is usually IP addresses). A host name is now passed as
2795 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2796 code that specifies a non-restarting handler (typically sigaction() in
2797 modern systems) in an attempt to fix a rare and obscure crash bug.
2799 15. Narrowed the window for a race in the daemon that could cause it to ignore
2800 SIGCHLD signals. This is not a major problem, because they are used only to
2801 wake it up if nothing else does.
2803 16. A malformed maildirsize file could cause Exim to calculate negative values
2804 for the mailbox size or file count. Odd effects could occur as a result.
2805 The maildirsize information is now recalculated if the size or filecount
2808 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2809 support for a long time. Removed HAVE_SYS_VFS_H.
2811 18. Updated exipick to current release from John Jetmore.
2813 19. Allow an empty sender to be matched against a lookup in an address list.
2814 Previously the only cases considered were a regular expression, or an
2817 20. Exim went into a mad DNS lookup loop when doing a callout where the
2818 host was specified on the transport, if the DNS lookup yielded more than
2821 21. The RFC2047 encoding function was originally intended for short strings
2822 such as real names; it was not keeping to the 75-character limit for
2823 encoded words that the RFC imposes. It now respects the limit, and
2824 generates multiple encoded words if necessary. To be on the safe side, I
2825 have increased the buffer size for the ${rfc2047: expansion operator from
2828 22. Failure to deliver a bounce message always caused it to be frozen, even if
2829 there was an errors_to setting on the router. The errors_to setting is now
2832 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2833 canonical form (fully expanded) before being placed in
2834 $sender_host_address.
2836 24. Updated eximstats to version 1.33
2838 25. Include certificate and key file names in error message when GnuTLS fails
2839 to set them up, because the GnuTLS error message doesn't include the name
2840 of the failing file when there is a problem reading it.
2842 26. Expand error message when OpenSSL has problems setting up cert/key files.
2845 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2846 (this can affect the format of dates).
2848 28. exim_tidydb, when checking for the continued existence of a message for
2849 which it has found a message-specific retry record, was not finding
2850 messages that were in split spool directories. Consequently, it was
2851 deleting retry records that should have stayed in existence.
2853 29. eximstats updated to version 1.35
2854 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2855 1.35 - bugfix such that pie charts by volume are generated correctly
2857 30. The SPA authentication driver was not abandoning authentication and moving
2858 on to the next authenticator when an expansion was forced to fail,
2859 contradicting the general specification for all authenticators. Instead it
2860 was generating a temporary error. It now behaves as specified.
2862 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2863 (the order specifies the preference for clients). The order is now AES256,
2864 AES128, 3DES, ARCFOUR128.
2866 31. Small patch to Sieve code - explicitly set From: when generating an
2869 32. Exim crashed if a remote delivery caused a very long error message to be
2870 recorded - for instance if somebody sent an entire SpamAssassin report back
2871 as a large number of 550 error lines. This bug was coincidentally fixed by
2872 increasing the size of one of Exim's internal buffers (big_buffer) that
2873 happened as part of the Exiscan merge. However, to be on the safe side, I
2874 have made the code more robust (and fixed the comments that describe what
2877 33. Some experimental protocols are using DNS PTR records for new purposes. The
2878 keys for these records are domain names, not reversed IP addresses. The
2879 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2880 leaves it alone. Component reversal etc. now happens only for IP addresses.
2883 34. The host_aton() function is supposed to be passed a string that is known
2884 to be a valid IP address. However, in the case of IPv6 addresses, it was
2885 not checking this. This is a hostage to fortune. Exim now panics and dies
2886 if the condition is not met. A case was found where this could be provoked
2887 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2888 components; fortuitously, this particular loophole had already been fixed
2889 by change 4.50/55 or 4.44/33 above.
2891 If there are any other similar loopholes, the new check in host_aton()
2892 itself should stop them being exploited. The report I received stated that
2893 data on the command line could provoke the exploit when Exim was running as
2894 exim, but did not say which command line option was involved. All I could
2895 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2896 running as the user.
2899 35. There was a buffer overflow vulnerability in the SPA authentication code
2900 (which came originally from the Samba project). I have added a test to the
2901 spa_base64_to_bits() function which I hope fixes it.
2904 36. The daemon start-up calls getloadavg() while still root for those OS that
2905 need the first call to be done as root, but it missed one case: when
2906 deliver_queue_load_max is set with deliver_drop_privilege. This is
2907 necessary for the benefit of the queue runner, because there is no re-exec
2908 when deliver_drop_privilege is set.
2910 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2911 was in use, was not putting the data itself into the right store pool;
2912 consequently, it could be overwritten for a subsequent message in the same
2913 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2916 38. Sometimes the final signoff response after QUIT could fail to get
2917 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2918 < 0 before doing a fflush(). This bug looks as though it goes back to the
2919 introduction of TLS in release 3.20, but "sometimes" must have been rare
2920 because the tests only now provoked it.
2926 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2927 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2928 Now it returns an integer. A number of calls were still expecting a T/F
2929 return. Fortuitously, in all cases, the tests worked in OK situations,
2930 which is the norm. However, things would have gone wrong on any write
2931 failures on the smtp file descriptor. This function is used when sending
2932 messages over SMTP and also when doing verify callouts.
2934 2. When Exim is called to do synchronous delivery of a locally submitted
2935 message (the -odf or -odi options), it no longer closes stderr before doing
2938 3. Implemented the mua_wrapper option.
2940 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2942 5. Implemented the functions header_remove(), header_testname(),
2943 header_add_at_position(), and receive_remove_recipient(), and exported them
2946 6. If an ACL "warn" statement specified the addition of headers, Exim already
2947 inserted X-ACL-Warn: at the start if there was no header name. However, it
2948 was not making this test for the second and subsequent header lines if
2949 there were newlines in the string. This meant that an invalid header could
2950 be inserted if Exim was badly configured.
2952 7. Allow an ACL "warn" statement to add header lines at the start or after all
2953 the Received: headers, as well as at the end.
2955 8. Added the rcpt_4xx retry error code.
2957 9. Added postmaster_mailfrom=xxx to callout verification option.
2959 10. Added mailfrom=xxxx to the callout verification option, for verify=
2962 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2963 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2965 12. Inserted some casts to stop certain compilers warning when using pointer
2966 differences as field lengths or precisions in printf-type calls (mostly
2967 affecting debugging statements).
2969 13. Added optional readline() support for -be (dynamically loaded).
2971 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2972 same clock tick as a message's arrival, so that its received time was the
2973 same as the "first fail" time on the retry record, and that message
2974 remained on the queue past the ultimate address timeout, every queue runner
2975 would try a delivery (because it was past the ultimate address timeout) but
2976 after another failure, the ultimate address timeout, which should have then
2977 bounced the address, did not kick in. This was a "< instead of <=" error;
2978 in most cases the first failure would have been in the next clock tick
2979 after the received time, and all would be well.
2981 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2982 being recognized when the domain list was tested by the match_domain
2983 condition in an expansion string.
2985 16. Added the ${str2b64: operator.
2987 17. Exim was always calling setrlimit() to set a large limit for the number of
2988 processes, without checking whether the existing limit was already
2989 adequate. (It did check for the limit on file descriptors.) Furthermore,
2990 errors from getrlimit() and setrlimit() were being ignored. Now they are
2991 logged to the main and panic logs, but Exim does carry on, to try to do its
2992 job under whatever limits there are.
2994 18. Imported PCRE 5.0.
2996 19. Trivial typo in log message " temporarily refused connection" (the leading
2999 20. If the log selector return_path_on_delivery was set and an address was
3000 redirected to /dev/null, the delivery process crashed because it assumed
3001 that a return path would always be set for a "successful" delivery. In this
3002 case, the whole delivery is bypassed as an optimization, and therefore no
3005 21. Internal re-arrangement: the function for sending a challenge and reading
3006 a response while authentication was assuming a zero-terminated challenge
3007 string. It's now changed to take a pointer and a length, to allow for
3008 binary data in such strings.
3010 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3012 23. Exim was not respecting finduser_retries when seeking the login of the
3013 uid under which it was called; it was always trying 10 times. (The default
3014 setting of finduser_retries is zero.) Also, it was sleeping after the final
3015 failure, which is pointless.
3017 24. Implemented tls_on_connect_ports.
3019 25. Implemented acl_smtp_predata.
3021 26. If the domain in control=submission is set empty, Exim assumes that the
3022 authenticated id is a complete email address when it generates From: or
3023 Sender: header lines.
3025 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3026 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3027 chown and chgrp in /bin and hostname in /usr/bin.
3029 28. Exim was keeping the "process log" file open after each use, just as it
3030 does for the main log. This opens the possibility of it remaining open for
3031 long periods when the USR1 signal hits a daemon. Occasional processlog
3032 errors were reported, that could have been caused by this. Anyway, it seems
3033 much more sensible not to leave this file open at all, so that is what now
3036 29. The long-running daemon process does not normally write to the log once it
3037 has entered its main loop, and it closes the log before doing so. This is
3038 so that log files can straightforwardly be renamed and moved. However,
3039 there are a couple of unusual error situations where the daemon does write
3040 log entries, and I had neglected to close the log afterwards.
3042 30. The text of an SMTP error response that was received during a remote
3043 delivery was being truncated at 512 bytes. This is too short for some of
3044 the long messages that one sometimes sees. I've increased the limit to
3047 31. It is now possible to make retry rules that apply only when a message has a
3048 specific sender, in particular, an empty sender.
3050 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3051 it possible to be selective about when SMTP synchronization is enforced.
3053 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3055 32. Implemented hosts_connection_nolog.
3057 33. Added an ACL for QUIT.
3059 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3062 35. Added mailbox_size and mailbox_filecount to appendfile.
3064 36. Added control = no_multiline_responses to ACLs.
3066 37. There was a bug in the logic of the code that waits for the clock to tick
3067 in the case where the clock went backwards by a substantial amount such
3068 that the microsecond fraction of "now" was more than the microsecond
3069 fraction of "then" (but the whole seconds number was less).
3071 38. Added support for the libradius Radius client library this is found on
3072 FreeBSD (previously only the radiusclient library was supported).
3078 1. When certain lookups returned multiple values in the form name=value, the
3079 quoting of the values was not always being done properly. Specifically:
3080 (a) If the value started with a double quote, but contained no whitespace,
3082 (b) If the value contained whitespace other than a space character (i.e.
3083 tabs or newlines or carriage returns) it was not quoted.
3084 This fix has been applied to the mysql and pgsql lookups by writing a
3085 separate quoting function and calling it from the lookup code. The fix
3086 should probably also be applied to nisplus, ibase and oracle lookups, but
3087 since I cannot test any of those, I have not disturbed their existing code.
3089 2. A hit in the callout cache for a specific address caused a log line with no
3090 reason for rejecting RCPT. Now it says "Previous (cached) callout
3091 verification failure".
3093 3. There was an off-by-one bug in the queryprogram router. An over-long
3094 return line was truncated at 256 instead of 255 characters, thereby
3095 overflowing its buffer with the terminating zero. As well as fixing this, I
3096 have increased the buffer size to 1024 (and made a note to document this).
3098 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3099 when Exim is waiting for an SMTP response from a remote server, Exim
3100 restarts its select() call on the socket, thereby resetting its timeout.
3101 This is not a problem when such interrupts are rare. Somebody set up a cron
3102 job to run exiwhat every 2 minutes, which is less than the normal select()
3103 timeout (5 or 10 minutes). This meant that the select() timeout never
3104 kicked in because it was always reset. I have fixed this by comparing the
3105 time when an interrupt arrives with the time at the start of the first call
3106 to select(). If more time than the timeout has elapsed, the interrupt is
3107 treated as a timeout.
3109 5. Some internal re-factoring in preparation for the addition of Sieve
3110 extensions (by MH). In particular, the "personal" test is moved to a
3111 separate function, and given an option for scanning Cc: and Bcc: (which is
3112 not set for Exim filters).
3114 6. When Exim created an email address using the login of the caller as the
3115 local part (e.g. when creating a From: or Sender: header line), it was not
3116 quoting the local part when it contained special characters such as @.
3118 7. Installed new OpenBSD configuration files.
3120 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3121 try to make them clearer.
3123 9. Callout options, other than the timeout value, were being ignored when
3124 verifying sender addresses in header lines. For example, when using
3126 verify = header_sender/callout=no_cache
3128 the cache was (incorrectly) being used.
3130 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3131 only the Cygwin environment.
3133 11. When return_path_on_delivery was set as a log selector, if different remote
3134 addresses in the same message used different return paths and parallel
3135 remote delivery occurred, the wrong values would sometimes be logged.
3136 (Whenever a remote delivery process finished, the return path value from
3137 the most recently started remote delivery process was logged.)
3139 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3140 lines when AUTH and/or TLS are in use. This is the "received protocol"
3141 field. Exim used to use "asmtp" for authenticated SMTP, without any
3142 indication (in the protocol name) for TLS use. Now it follows the RFC and
3143 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3144 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3145 names appear in log lines as well as in Received: header lines.
3147 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3148 extensions, and comparison tests, and to fix some bugs.
3150 14. Changes to the "personal" filter test:
3152 (1) The test was buggy in that it was just doing the equivalent of
3153 "contains" tests on header lines. For example, if a user's address was
3154 anne@some.where, the "personal" test would incorrectly be true for
3156 To: susanne@some.where
3158 This test is now done by extracting each address from the header in turn,
3159 and checking the entire address. Other tests that are part of "personal"
3160 are now done using regular expressions (for example, to check local parts
3161 of addresses in From: header lines).
3163 (2) The list of non-personal local parts in From: addresses has been
3164 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3165 taken from the Sieve specification recommendations.
3167 (3) If the message contains any header line starting with "List-" it is
3168 treated as non-personal.
3170 (4) The test for "circular" in the Subject: header line has been removed
3171 because it now seems ill-conceived.
3173 15. Minor typos in src/EDITME comments corrected.
3175 16. Installed latest exipick from John Jetmore.
3177 17. If headers_add on a router specified a text string that was too long for
3178 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3179 of string_sprintf() is now avoided.
3181 18. $message_body_size was not set (it was always zero) when running the DATA
3182 ACL and the local_scan() function.
3184 19. For the "mail" command in an Exim filter, no default was being set for
3185 the once_repeat time, causing a random time value to be used if "once" was
3186 specified. (If the value happened to be <= 0, no repeat happened.) The
3187 default is now 0s, meaning "never repeat". The "vacation" command was OK
3188 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3189 (I found it when inspecting the code).
3191 20. There is now an overall timeout for performing a callout verification. It
3192 defaults to 4 times the callout timeout, which applies to individual SMTP
3193 commands during the callout. The overall timeout applies when there is more
3194 than one host that can be tried. The timeout is checked before trying the
3195 next host. This prevents very long delays if there are a large number of
3196 hosts and all are timing out (e.g. when the network connections are timing
3197 out). The value of the overall timeout can be changed by specifying an
3198 additional sub-option for "callout", called "maxwait". For example:
3200 verify = sender/callout=5s,maxwait=20s
3202 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3203 to the end before writing, but this should make it even safer).
3205 22. Exim was forgetting that it had advertised PIPELINING for the second and
3206 subsequent messages on an SMTP connection. It was also not resetting its
3207 memory on STARTTLS and an internal HELO.
3209 23. When Exim logs an SMTP synchronization error within a session, it now
3210 records whether PIPELINING has been advertised or not.
3212 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3213 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3214 rather than long int.
3216 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3218 26. Added the never_mail option to autoreply.
3224 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3225 crash if the getsockname() call failed; this can happen if a connection is
3226 closed very soon after it is established. The problem was simply in the
3227 order in which certain operations were done, causing Exim to try to write
3228 to the SMTP stream before it had set up the file descriptor. The bug has
3229 been fixed by making things happen in the correct order.
3235 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3236 before the connection was closed, thus losing the rejection response.
3238 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3239 some early Solaris releases, but causes trouble in current releases where
3240 socklen_t is defined.
3242 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3245 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3248 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3249 that releasing the top of it at the end releases what was used for sub-
3250 expansions (unless the block got too big). However, discard this block if
3251 the first thing is a variable or header, so that we can use its block when
3252 it is dynamic (useful for very large $message_headers, for example).
3254 6. Lookups now cache *every* query, not just the most recent. A new, separate
3255 store pool is used for this. It can be recovered when all lookup caches are
3256 flushed. Lookups now release memory at the end of their result strings.
3257 This has involved some general refactoring of the lookup sources.
3259 7. Some code has been added to the store_xxx() functions to reduce the amount
3260 of flapping under certain conditions.
3262 8. log_incoming_interface used to affect only the <= reception log lines. Now
3263 it causes the local interface and port to be added to several more SMTP log
3264 lines, for example "SMTP connection from", and rejection lines.
3266 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3268 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3270 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3271 could be overwritten at the end of the current message (or the start of a
3272 new message if it was set in a HELO ACL). The value is now preserved for
3273 the duration of the SMTP connection.
3275 12. If a transport had a headers_rewrite setting, and a matching header line
3276 contained an unqualified address, that address was qualified, even if it
3277 did not match any rewriting rules. The underlying bug was that the values
3278 of the flags that permit the existence of unqualified sender and recipient
3279 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3280 non-local messages, and by -bnq for local messages) were not being
3281 preserved with the message after it was received.
3283 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3284 "next input=" as part of the text comprising the host identity instead of
3285 the correct text. The code was using the same buffer for two different
3286 strings. However, depending on which order the printing function evaluated
3287 its arguments, the bug did not always show up. Under Linux, for example, my
3288 test suite worked just fine.
3290 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3291 doesn't work with some older versions of Perl. It has been changed to "my",
3292 which in any case is probably the better facility to use.
3294 15. A really picky compiler found some instances of statements for creating
3295 error messages that either had too many or two few arguments for the format
3298 16. The size of the buffer for calls to the DNS resolver has been increased
3299 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3300 for addresses that have a lot of PTR records. This alleviates a problem; it
3301 does not fully solve it.
3303 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3304 buffer now truncates the list and logs the incident, which is the same
3305 action as happens when Exim is looking up a host name and its aliases.
3306 Previously in this situation something unpredictable would happen;
3307 sometimes it was "internal error: store_reset failed".
3309 18. If a server dropped the connection unexpectedly when an Exim client was
3310 using GnuTLS and trying to read a response, the client delivery process
3311 crashed while trying to generate an error log message.
3313 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3314 string, for example:
3316 warn message = H1: something\nH2: something
3318 the text was added as a single header line from Exim's point of view
3319 though it ended up OK in the delivered message. However, searching for the
3320 second and subsequent header lines using $h_h2: did not work. This has been
3321 fixed. Similarly, if a system filter added multiple headers in this way,
3322 the routers could not see them.
3324 20. Expanded the error message when iplsearch is called with an invalid key to
3325 suggest using net-iplsearch in a host list.
3327 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3328 ACLs are no longer actually imposed (and a message to that effect is
3331 22. If a "gecos" field in a passwd entry contained escaped characters, in
3332 particular, if it contained a \" sequence, Exim got it wrong when building
3333 a From: or a Sender: header from that name. A second bug also caused
3334 incorrect handling when an unquoted " was present following a character
3335 that needed quoting.
3337 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3338 was not being matched caselessly.
3340 24. Arranged for all hyphens in the exim.8 source to be escaped with
3343 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3344 properly. Recipient callout cache records were still being keyed to include
3345 the sender, even when use_sender was set false. This led to far more
3346 callouts that were necessary. The sender is no longer included in the key
3347 when use_sender is false.
3349 26. Added "control = submission" modifier to ACLs.
3351 27. Added the ${base62d: operator to decode base 62 numbers.
3353 28. dnsdb lookups can now access SRV records.
3355 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3356 the configuration file.
3358 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3359 (-v) mode. This makes the output for a verbose queue run more intelligible.
3361 31. Added a use_postmaster feature to recipient callouts.
3363 32. Added the $body_zerocount variable, containing the number of binary zero
3364 bytes in the message body.
3366 33. The time of last modification of the "new" subdirectory is now used as the
3367 "mailbox time last read" when there is a quota error for a maildir
3370 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3372 35. Added +ignore_unknown as a special item in host lists.
3374 36. Code for decoding IPv6 addresses in host lists is now included, even if
3375 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3376 address was recognized as an IP address, but was then not correctly decoded
3377 into binary, causing unexpected and incorrect effects when compared with
3384 1. Very minor rewording of debugging text in manualroute to say "list of
3385 hosts" instead of "hostlist".
3387 2. If verify=header_syntax was set, and a header line with an unqualified
3388 address (no domain) and a large number of spaces between the end of the
3389 name and the colon was received, the reception process suffered a buffer
3390 overflow, and (when I tested it) crashed. This was caused by some obsolete
3391 code that should have been removed. The fix is to remove it!
3393 3. When running in the test harness, delay a bit after writing a bounce
3394 message to get a bit more predictability in the log output.
3396 4. Added a call to search_tidyup() just before forking a reception process. In
3397 theory, someone could use a lookup in the expansion of smtp_accept_max_
3398 per_host which, without the tidyup, could leave open a database connection.
3400 5. Added the variables $recipient_data and $sender_data which get set from a
3401 lookup success in an ACL "recipients" or "senders" condition, or a router
3402 "senders" option, similar to $domain_data and $local_part_data.
3404 6. Moved the writing of debug_print from before to after the "senders" test
3407 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3408 problems for message scanning, either using a data ACL, or using
3409 local_scan() because the Received: header was not generated till after they
3410 were called (in order to set the time as the time of reception completion).
3411 I have revised the way this works. The header is now generated after the
3412 body is received, but before the ACL or local_scan() are called. After they
3413 are run, the timestamp in the header is updated.
3419 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3420 before starting a queue runner without re-exec. This happened only when
3421 deliver_drop_privilege was set or when the Exim user was set to root. The
3422 effect of the bug was that timeouts during subsequent deliveries caused
3423 crashes instead of being properly handled. The handler is now left at its
3424 default (and expected) setting.
3426 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3427 message, again when deliver_drop_privilege is set or Exim is run as root.
3428 The bug described in (1) was not present in this case, but the tidying up
3429 of the other signals was missing. I have made the two cases consistent.
3431 3. The ignore_target_hosts setting on a manualroute router was being ignored
3432 for hosts that were looked up using the /MX notation.
3434 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3437 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3438 operated on the sender address. After changing the $sender_address to <>
3439 for the sender address verify, Exim was re-instated it as the original
3440 (before rewriting) address, but remembering that it had rewritten it, so it
3441 wasn't rewriting it again. This bug also had the effect of breaking the
3442 sender address verification caching when the sender address was rewritten.
3444 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3445 This has been changed so that if the ip literal address matches
3446 ignore_target_hosts, the router declines.
3448 7. Added expansion conditions match_domain, match_address, and match_local_
3449 part (NOT match_host).
3451 8. The placeholder for the Received: header didn't have a length field set.
3453 9. Added code to Exim itself and to exim_lock to test for a specific race
3454 condition that could lead to file corruption when using MBX delivery. The
3455 issue is with the lockfile that is created in /tmp. If this file is removed
3456 after a process has opened it but before that process has acquired a lock,
3457 there is the potential for a second process to recreate the file and also
3458 acquire a lock. This could lead to two Exim processes writing to the file
3459 at the same time. The added code performs the same test as UW imapd; it
3460 checks after acquiring the lock that its file descriptor still refers to
3461 the same named file.
3463 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3464 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3467 11. Added the smtp_active_hostname option. If used, this will typically be made
3468 to depend on the incoming interface address. Because $interface_address is
3469 not set up until the daemon has forked a reception process, error responses
3470 that can happen earlier (such as "too many connections") no longer contain
3473 12. If an expansion in a condition on a "warn" statement fails because a lookup
3474 defers, the "warn" statement is abandoned, and the next ACL statement is
3475 processed. Previously this caused the whole ACL to be aborted.
3477 13. Added the iplsearch lookup type.
3479 14. Added ident_timeout as a log selector.
3481 15. Added tls_certificate_verified as a log selector.
3483 16. Added a global option tls_require_ciphers (compare the smtp transport
3484 option of the same name). This controls incoming TLS connections.
3486 17. I finally figured out how to make tls_require_ciphers do a similar thing
3487 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3488 before starting the TLS session.
3490 18. Tabs are now shown as \t in -bP output.
3492 19. If the log selector return_path_on_delivery was set, Exim crashed when
3493 bouncing a message because it had too many Received: header lines.
3495 20. If two routers both had headers_remove settings, and the first one included
3496 a superfluous trailing colon, the final name in the first list and the
3497 first name in the second list were incorrectly joined into one item (with a
3498 colon in the middle).
3504 1. Added -C and -D options to the exinext utility, mainly to make it easier
3505 to include in the automated testing, but these could be helpful when
3506 multiple configurations are in use.
3508 2. The exinext utility was not formatting the output nicely when there was
3509 an alternate port involved in the retry record key, nor when there was a
3510 message id as well (for retries that were specific to a specific message
3511 and a specific host). It was also confused by IPv6 addresses, because of
3512 the additional colons they contain. I have fixed the IPv4 problem, and
3513 patched it up to do a reasonable job for IPv6.
3515 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3516 delivery, the log line now contains "pipelined" if PIPELINING was used.
3518 4. An SMTP transport process used to panic and die if the bind() call to set
3519 an explicit outgoing interface failed. This has been changed; it is now
3520 treated in the same way as a connect() failure.
3522 5. A reference to $sender_host_name in the part of a conditional expansion
3523 that was being skipped was still causing a DNS lookup. This no longer
3526 6. The def: expansion condition was not recognizing references to header lines
3527 that used bh_ and bheader_.
3529 7. Added the _cache feature to named lists.
3531 8. The code for checking quota_filecount in the appendfile transport was
3532 allowing one more file than it should have been.
3534 9. For compatibility with Sendmail, the command line option
3542 and sets the incoming protocol and host name (for trusted callers). The
3543 host name and its colon can be omitted when only the protocol is to be set.
3544 Note the Exim already has two private options, -pd and -ps, that refer to
3545 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3546 "s", but I don't think that's a major issue.
3548 10. A number of refactoring changes to the code, none of which should affect
3551 (a) The number of logging options was getting close to filling up the
3552 32-bit word that was used as a bit map. I have split them into two classes:
3553 those that are passed in the argument to log_write(), and those that are
3554 only ever tested independently outside of that function. These are now in
3555 separate 32-bit words, so there is plenty of room for expansion again.
3556 There is no change in the user interface or the logging behaviour.
3558 (b) When building, for example, log lines, the code previously used a
3559 macro that called string_cat() twice, in order to add two strings. This is
3560 not really sufficiently general. Furthermore, there was one instance where
3561 it was actually wrong because one of the argument was used twice, and in
3562 one call a function was used. (As it happened, calling the function twice
3563 did not affect the overall behaviour.) The macro has been replaced by a
3564 function that can join an arbitrary number of extra strings onto a growing
3567 (c) The code for expansion conditions now uses a table and a binary chop
3568 instead of a serial search (which was left over from when there were very
3569 few conditions). Also, it now recognizes conditions like "pam" even when
3570 the relevant support is not compiled in: a suitably worded error message is
3571 given if an attempt is made to use such a condition.
3573 11. Added ${time_interval:xxxxx}.
3575 12. A bug was causing one of the ddress fields not to be passed back correctly
3576 from remote delivery subprocesses. The field in question was not being
3577 subsequently used, so this caused to problems in practice.
3579 13. Added new log selectors queue_time and deliver_time.
3581 14. Might have fixed a bug in maildirsizefile handling that threw up
3582 "unexpected character" debug warnings, and recalculated the data
3583 unnecessarily. In any case, I expanded the warning message to give more
3586 15. Added the message "Restricted characters in address" to the statements in
3587 the default ACL that block characters like @ and % in local parts.
3589 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3590 Three changes have been made:
3592 (a) There was a serious bug; a negative response to MAIL caused the whole
3593 recipient domain to be cached as invalid, thereby blocking all messages
3594 to all local parts at the same domain, from all senders. This bug has
3595 been fixed. The domain is no longer cached after a negative response to
3596 MAIL if the sender used is not empty.
3598 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3601 (c) A new callout option, "use_sender" has been added for people who want
3602 the modified behaviour.
3608 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3611 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3612 indeed breaks things for older releases.
3614 3. Added additional logging to the case where there is a problem reading data
3615 from a filter that is running in a subprocess using a pipe, in order to
3616 try to track down a specific problem.
3618 4. Testing facility fudge: when running in the test harness and attempting
3619 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3620 getting "No route to host". Convert this to a timeout.
3622 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3625 6. Some OS don't have socklen_t but use size_t instead. This affects the
3626 fifth argument of getsockopt() amongst other things. This is now
3627 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3628 can be set for individual OS. I have set it for SunOS5, OSF1, and
3629 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3630 some earlier ones do not.
3632 7. Change 4.30/15 was not doing the test caselessly.
3634 8. The standard form for an IPv6 address literal was being rejected by address
3635 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3636 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3637 this, as well as the form without the "IPv6" on the front (but only when
3638 address literals are enabled, of course).
3640 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3642 10. Exim crashed if a message with an empty sender address specified by -f
3643 encountered a router with an errors_to setting. This could be provoked only
3644 by a command such as
3648 where an empty string was supplied; "<>" did not hit this bug.
3650 11. Installed PCRE release 4.5.
3652 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3653 remained set. It is now erased.
3655 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3656 times from message ids (which are base 36 rather than the normal 62).
3658 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3659 were being counted as actual protocol errors, and logged if the log
3660 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3661 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3662 and DATA following a set of rejected RCPTs do not count as protocol errors.
3663 In other words, Exim assumes they were pipelined, though this may not
3664 actually be the case. Of course, in all cases the client gets an
3665 appropriate error code.
3667 15. If a lookup fails in an ACL condition, a message about the failure may
3668 be available; it is used if testing the ACL cannot continue, because most
3669 such messages specify what the cause of the deferral is. However, some
3670 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3671 that caused an old message to be retained and used if a later statement
3672 caused a defer, replacing the real cause of the deferral.
3674 16. If an IP address had so many PTR records that the DNS lookup buffer
3675 was not large enough to hold them, Exim could crash while trying to process
3676 the truncated data. It now detects and logs this case.
3678 17. Further to 4.21/58, another change has been made: if (and only if) the
3679 first line of a message (the first header line) ends with CRLF, a bare LF
3680 in a subsequent header line has a space inserted after it, so as not to
3681 terminate the header.
3683 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3684 unnecessarily, used atoi() instead of strtol(), and didn't check the
3685 termination when getting file sizes from file names by regex.
3687 19. Completely re-implemented the support for maildirsize files, in the light
3688 of a number of problems with the previous contributed implementation
3689 (4.30/29). In particular:
3691 . If the quota is zero, the maildirsize file is maintained, but no quota is
3694 . If the maildir directory does not exist, it is created before any attempt
3695 to write a maildirsize file.
3697 . The quota value in the file is just a cache; if the quota is changed in
3698 the transport, the new value overrides.
3700 . A regular expression is available for excluding directories from the
3703 20. The autoreply transport checks the characters in options that define the
3704 message's headers; it allows continued headers, but it was checking with
3705 isspace() after an embedded newline instead of explicitly looking for a
3708 21. If all the "regular" hosts to which an address was routed had passed their
3709 expiry times, and had not reached their retry times, the address was
3710 bounced, even if fallback hosts were defined. Now Exim should go on to try
3713 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3714 equivalent code in the SMTP transport. Some hosts send humungous responses
3715 to HELO/EHLO, more than 1024 it seems.
3717 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3718 gives compiler warnings in some environments. I've now done it "properly",
3721 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3722 (because of problems with the built-in one) was declared to return uschar *
3723 instead of char *, causing compiler failure.
3725 25. Fixed a file descriptor leak when processing alias/forward files.
3727 26. Fixed a minor format string issue in dbfn.c.
3729 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3731 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3732 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3735 29. When a sender address is verified, it is cached, to save repeating the test
3736 when there is more than one recipient in a message. However, when the
3737 verification involves a callout, it is possible for different callout
3738 options to be set for different recipients. It is too complicated to keep
3739 track of this in the cache, so now Exim always runs a verification when a
3740 callout is required, relying on the callout cache for the optimization.
3741 The overhead is duplication of the address routing, but this should not be
3744 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3745 to be verified with callout=postmaster, and the main callout worked but the
3746 postmaster check failed, the verification correctly failed. However, if a
3747 subsequent RCPT command asked for sender verification *without* the
3748 postmaster check, incorrect caching caused this verification also to fail,
3751 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3752 it was not caching the DNS options (qualify_single, search_parents) that
3753 were used when the lookup failed. A subsequent lookup with different
3754 options therefore always gave the same answer, though there were cases
3755 where it should not have. (Example: a "domains = !$mx_any" option on a
3756 dnslookup router: the "domains" option is always processed without any
3757 widening, but the router might have qualify_single set.) Now Exim uses the
3758 cached value only when the same options are set.
3760 32. Added John Jetmore's "exipick" utility to the distribution.
3762 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3763 than a timeout (e.g. a certificate is required, and is not provided), an
3764 Exim server now closes the connection immediately. Previously it waited for
3765 the client to close - but if the client is SSL, it seems that they each
3766 wait for each other, leading to a delay before one of them times out.
3768 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3769 maintained 0.8.x compatibility because I don't think many are using it, and
3770 it is clearly obsolete.
3772 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3775 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3776 client certificate was expired. A simple patch fixes this, though I don't
3777 understand the full logic of why the verify callback is called multiple
3780 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3781 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3782 which causes problems with some clients (such as the Certicom SSL Plus
3783 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3784 disables the coutermeasure allowing Eudora to connect."
3786 38. Exim was not checking that a write() to a log file succeeded. This could
3787 lead to Bad Things if a log got too big, in particular if it hit a file
3788 size limit. Exim now panics and dies if it cannot write to a log file, just
3789 as it does if it cannot open a log file.
3791 39. Modified OS/Makefile-Linux so that it now contains
3793 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3795 The two -D definitions ensure that Exim is compiled with large file
3796 support, which makes it possible to handle log files that are bigger than
3799 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3800 instance) a domain was checked against a named list that involved a lookup,
3801 causing $domain_data to be set, then another domain was checked against the
3802 same list, then the first domain was re-checked, the value of $domain_data
3803 after the final check could be wrong. In particular, if the second check
3804 failed, it could be set empty. This bug probably also applied to
3807 41. The strip_trailing_dot option was not being applied to the address given
3808 with the -f command-line option.
3810 42. The code for reading a message's header from the spool was incrementing
3811 $received_count, but never initializing it. This meant that the value was
3812 incorrect (doubled) while delivering a message in the same process in which
3813 it was received. In the most common configuration of Exim, this never
3814 happens - a fresh exec is done - but it can happen when
3815 deliver_drop_privilege is set.
3817 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3818 it now includes up to 150 characters of the unexpected data in the log
3821 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3822 and building data strings. The size of both of these buffers was 10 000
3823 bytes - far larger than anybody would *ever* want, thought I. Needless to
3824 say, somebody hit the limit. I have increased the maximum line length to
3825 20 000 and the maximum data length of concatenated lines to 100 000. I have
3826 also fixed two bugs, because there was no checking on these buffers. Tsk,
3827 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3828 buffer is too small.
3830 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3831 lsearch lookups. Now it does.
3833 46. When parsing a route_list item in a manualroute router, a fixed-length
3834 buffer was used for the list of hosts. I made this 1024 bytes long,
3835 thinking that nobody would ever have a list of hosts that long. Wrong.
3836 Somebody had a whole pile of complicated expansion conditions, and the
3837 string was silently truncated, leading to an expansion error. It turns out
3838 that it is easier to change to an unlimited length (owing to other changes
3839 that have happened since this code was originally written) than to build
3840 structure for giving a limitation error. The length of the item that
3841 expands into the list of hosts is now unlimited.
3843 47. The lsearch lookup could not handle data where the length of text line was
3844 more than 4095 characters. Such lines were truncated, leading to shortened
3845 data being returned. It should now handle lines of any length.
3847 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3848 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3851 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3852 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3853 no longer happens in comment lines. A list of replacements is now placed
3854 at the head of all of the source files, except those whose only change is
3855 to replace PERL_COMMAND in the very first #! line.
3857 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3858 messages on the queue, with a bottom-up merge sort, using code contributed
3859 by Michael Haardt. This should make operations like -bp somewhat faster on
3860 large queues. It won't affect queue runners, except when queue_run_in_order
3863 51. Installed eximstats 1.31 in the distribution.
3865 52. Added support for SRV lookups to the dnslookup router.
3867 53. If an ACL referred to $message_body or $message_body_end, the value was not
3868 reset for any messages that followed in the same SMTP session.
3870 54. The store-handling optimization for building very long strings was not
3871 differentiating between the different store pools. I don't think this
3872 actually made any difference in practice, but I've tidied it.
3874 55. While running the routers to verify a sender address, $sender_address
3875 was still set to the sender address. This is wrong, because when routing to
3876 send a bounce to the sender, it would be empty. Therefore, I have changed
3877 it so that, while verifying a sender address, $sender_address is set to <>.
3878 (There is no change to what happens when verifying a recipient address.)
3880 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3881 target A or AAAA records (if not already returned) without resetting the
3882 qualify_single or search_parents options of the DNS resolver. These are
3883 inappropriate in this case because the targets of MX and SRV records must
3884 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3885 target that, when qualified, matched something in the local domain. These
3886 two options are now turned off when doing these lookups.
3888 57. It seems that at least some releases of Reiserfs (which does not have the
3889 concept of a fixed number of inodes) returns zero and not -1 for the
3890 number of available inodes. This interacted badly with check_spool_inodes,
3891 which assumed that -1 was the "no such thing" setting. What I have done is
3892 to check that the total number of inodes is greater than zero before doing
3893 the test of how many are available.
3895 58. When a "warn" ACL statement has a log_message modifier, the message is
3896 remembered, and not repeated. This is to avoid a lot of repetition when a
3897 message has many recipients that cause the same warning to be written.
3898 Howewer, Exim was preserving the list of already written lines for an
3899 entire SMTP session, which doesn't seem right. The memory is now reset if a
3900 new message is started.
3902 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3903 debugging output unless log_rewrite was also set.
3905 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3906 of (int)(handle) when we know that handle contains (void *)(-1).
3908 61. The Exim daemon panic-logs an error return when it closes the incoming
3909 connection. However "connection reset by peer" seems to be common, and
3910 isn't really an error worthy of noting specially, so that particular error
3913 62. When Exim is trying to find all the local interfaces, it used to panic and
3914 die if the ioctl to get the interface flags failed. However, it seems that
3915 on at least one OS (Solaris 9) it is possible to have an interface that is
3916 included in the list of interfaces, but for which you get a failure error
3917 for this call. This happens when the interface is not "plumbed" into a
3918 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3919 failure of the "get flags" call assumes that the interface is down.
3921 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3922 makes life easier for people who are doing arithmetic on fields extracted
3923 from dates, where you often get leading zeros that should not be
3924 interpreted as octal.
3926 64. Added qualify_domain to the redirect router, to override the global
3929 65. If a pathologically long header line contained very many addresses (the
3930 report of this problem mentioned 10 000) and each of them was rewritten,
3931 Exim could use up a very large amount of memory. (It kept on making new
3932 copies of the header line as it rewrote, and never released the old ones.)
3933 At the expense of a bit more processing, the header rewriting function has
3934 been changed so that it no longer eats memory in this way.
3936 66. The generation of the Received: header has been moved from the time that a
3937 message starts to be received, to the time that it finishes. The timestamp
3938 in the Received: header should now be very close to that of the <= log
3939 line. There are two side-effects of this change:
3941 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3942 logged header lines no longer include the local Received: line, because
3943 it has not yet been created. The same applies to a copy of the message
3944 that is returned to a non-SMTP sender when a message is rejected.
3946 (b) When a filter file is tested using -bf, no additional Received: header
3947 is added to the test message. After some thought, I decided that this
3950 This change does not affect the value of $received_for. It is still set
3951 after address rewriting, but before local_scan() is called.
3953 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3955 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3956 gave an unhelpful panic error message, and a defer error. I have managed to
3957 change this behaviour so that it now rejects any supplied certificate,
3958 which seems right, as the list of acceptable certificates is empty.
3960 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3961 gave an unhelpful defer error. I have not managed to make this reject any
3962 supplied certificates, but the error message it gives is "no certificate
3963 supplied", which is not helpful.
3965 70. exigrep's output now also includes lines that are not associated with any
3966 message, but which match the given pattern. Implemented by a patch from
3967 Martin Sluka, which also tidied up the Perl a bit.
3969 71. Recipient callout verification, like sender verification, was using <> in
3970 the MAIL FROM command. This isn't really the right thing, since the actual
3971 sender may affect whether the remote host accepts the recipient or not. I
3972 have changed it to use the actual sender in the callout; this means that
3973 the cache record is now keyed on a recipient/sender pair, not just the
3974 recipient address. There doesn't seem to be a real danger of callout loops,
3975 since a callout by the remote host to check the sender would use <>.
3976 [SEE ABOVE: changed after hitting problems.]
3978 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3979 temporary errors. However, in the case of such a code being given after
3980 the end of a data transmission (i.e. after ".") Exim was failing to write
3981 a retry record for the message. (Yes, there was some broken host that was
3982 actually sending 8xx at this point.)
3984 73. An unknown lookup type in a host list could cause Exim to panic-die when
3985 the list was checked. (An example that provoked this was putting <; in the
3986 middle of a list instead of at the start.) If this happened during a DATA
3987 ACL check, a -D file could be left lying around. This kind of configuration
3988 error no longer causes Exim to die; instead it causes a defer errror. The
3989 incident is still logged to the main and panic logs.
3991 74. Buglet left over from Exim 3 conversion. The message "too many messages
3992 in one connection" was written to the rejectlog but not the mainlog, except
3993 when address rewriting (yes!) was being logged.
3995 75. Added write_rejectlog option.
3997 76. When a system filter was run not as root (that is, when system_filter_user
3998 was set), the values of the $n variables were not being returned to the
3999 main process; thus, they were not subsequently available in the $sn
4002 77. Added +return_path_on_delivery log selector.
4004 78. A connection timeout was being treated differently from recipients deferred
4005 when testing hosts_max_try with a message that was older than the host's
4006 retry timeout. (The host should not be counted, thus allowing all hosts to
4007 be tried at least once before bouncing.) This may have been the cause of an
4008 occasionally reported bug whereby a message would remain on the queue
4009 longer than the retry timeout, but would be bounced if a delivery was
4010 forced. I say "may" because I never totally pinned down the problem;
4011 setting up timeout/retry tests is difficult. See also the next item.
4013 79. The ultimate address timeout was not being applied to errors that involved
4014 a combination of host plus message (for example, a timeout on a MAIL
4015 command). When an address resolved to a number of possible hosts, and they
4016 were not all tried for each delivery (e.g. because of hosts_max_try), a
4017 message could remain on the queue longer than the retry timeout.
4019 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4022 81. Fixed an obscure SMTP outgoing bug which required at least the following
4023 conditions: (a) there was another message waiting for the same server;
4024 (b) the server returned 5xx to all RCPT commands in the first message so
4025 that the message was not completed; (c) the server dropped the connection
4026 or gave a negative response to the RSET that Exim sends to abort the
4027 transaction. The observed case was a dropped connection after DATA that had
4028 been sent in pipelining mode. That is, the server had advertised PIPELINING
4029 but was not implementing it correctly. The effect of the bug was incorrect
4030 behaviour, such as trying another host, and this could lead to a crash.
4036 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4037 and daemon.c were passed as pointers to ints; they should have been
4038 pointers to socklen_t variables (which are typically unsigned ints).
4040 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4043 3. Fixed a really odd bug that affected only the testing scheme; patching a
4044 certain fixed string in the binary changed the value of another string that
4045 happened to be identical to the end of the original first string.
4047 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4048 name", it returns that address as the IP address. On some operating
4049 systems (e.g. Solaris), it also passes back the IP address string as the
4050 "host name". However, on others (e.g. Linux), it passes back an empty
4051 string. Exim wasn't checking for this, and was changing the host name to an
4052 empty string, assuming it had been canonicized.
4054 5. Although rare, it is permitted to have more than one PTR record for a given
4055 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4056 all the names associated with an address, because they do in Solaris.
4057 However, it seems that they do not in Linux for data that comes from the
4058 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4059 I found this out when I moved to a new Linux workstation and tried to run
4060 the Exim test suite.
4062 To get round this problem I have changed the code so that it now does its
4063 own call to the DNS to look up PTR records when searching for a host name.
4064 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4065 addresses that are only in /etc/hosts are still found.
4067 This behaviour is, however, controlled by an option called host_lookup_
4068 order, which defaults to "bydns:byaddr". If people want to use the other
4069 order, or indeed, just use one or the other means of lookup, they can
4070 specify it in this variable.
4072 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4073 some operating systems, this comes back from gethostbyaddr() as an empty
4074 string, and this is what Exim used to test for. However, it seems that in
4075 other systems, "." is yielded. Exim now tests for this case too.
4077 7. The values of check_spool_space and check_log_space are now held internally
4078 as a number of kilobytes instead of an absolute number of bytes. If a
4079 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4080 kilobyte. This means that much larger values can be stored.
4082 8. Exim monitor: an attempt to get the action menu when not actually pointing
4083 at a message produces an empty menu entitled "No message selected". This
4084 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4085 no entries in it ("Shell widget menu has zero width and/or height"). So I
4086 have added a single, blank menu entry in this case.
4088 9. Added ${quote_local_part.
4090 10. MIME decoding is now applied to the contents of Subject: header lines when
4093 11. Now that a reference to $sender_host_address automatically causes a reverse
4094 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4095 host lookup before query-style lookups in lists that might use this
4096 variable. This has therefore been abolished, and the "net-" prefix is no
4097 longer necessary for query-style lookups.
4099 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4100 have been a typo for LFLAGS, so it has been changed.
4102 13. The install script calls Exim with "-C /dev/null" in order to find the
4103 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4104 to be output. Howeve, since Exim outputs its version number before the
4105 error, it didn't break the script. It just looked ugly. I fixed this by
4106 always allowing "-C /dev/null" if the caller is root.
4108 14. Ignore overlarge ACL variable number when reading spool file - insurance
4109 against a later release with more variables having written the file.
4111 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4112 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4113 this, as well as the form without the "IPv6" on the front.
4115 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4116 OS/Makefile-Darwin file.
4118 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4119 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4121 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4122 variable. However, some OS use ints for inodes. I've added cast to long int
4123 to get rid of the compiler warning.
4125 19. I had forgotten to lock out "/../" in configuration file names when
4126 ALT_CONFIG_PREFIX was set.
4128 20. Routers used for verification do not need to specify transports. However,
4129 if such a router generated a host list, and callout was configured, Exim
4130 crashed, because it could not find a port number from the (non-existent)
4131 transport. It now assumes port 25 in this circumstance.
4133 21. Added the -t option to exigrep.
4135 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4136 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4137 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4138 them is set, LOOKUP_LSEARCH is forced.
4140 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4142 24. Added sender and host information to the "rejected by local_scan()" log
4143 line; previously there was no indication of these.
4145 25. Added .include_if_exists.
4147 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4148 Linux. It turns out that not all file systems support this. Apparently some
4149 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4150 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4151 not supported on the file descriptor, is now ignored when Exim is trying to
4152 sync a directory. This applies only to Linux.
4154 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4156 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4157 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4160 29. Added support for maildirsize files from supplied patch (modified a bit).
4162 30. The use of :fail: followed by an empty string could lead Exim to respond to
4163 sender verification failures with (e.g.):
4165 550 Verification failed for <xxx>
4166 550 Sender verify failed
4168 where the first response line was missing the '-' that indicates it is not
4169 the final line of the response.
4171 31. The loop for finding the name of the user that called Exim had a hardwired
4172 limit of 10; it now uses the value of finduser_retries, which is used for
4173 all other user lookups.
4175 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4178 33. Exim was neglecting to zero errno before one call of strtol() when
4179 expanding a string and expecting an integer value. On some systems this
4180 resulted in spurious "integer overflow" errors. Also, it was casting the
4181 result into an int without checking.
4183 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4184 did not work. The code looks as if it has *never* worked, though it appears
4185 to have been documented since at least releast 1.62. I have made it work.
4187 35. The "timeout_DNS" error in retry rules, also documented since at least
4188 1.62, also never worked. As it isn't clear exactly what this means, and
4189 clearly it isn't a major issue, I have abolished the feature by treating it
4190 as "timeout", and writing a warning to the main and panic logs.
4192 36. The display of retry rules for -brt wasn't always showing the error code
4195 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4196 timeout_connect_A, timeout_connect_MX.
4198 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4199 to the empty sender.
4201 39. The daemon was not analysing the content of -oX till after it had closed
4202 stderr and disconnected from the controlling terminal. This meant that any
4203 syntax errors were only noted on the panic log, and the return code from
4204 the command was 0. By re-arranging the code a little, I've made the
4205 decoding happen first, so such errors now appear on stderr, and the return
4206 code is 1. However, the actual setting up of the sockets still happens in
4207 the disconnected process, so errors there are still only recorded on the
4210 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4211 connections (as happens on some IP stacks) was logged at start up time as
4212 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4213 it from "IPv6 and IPv4", which means that two separate sockets are being
4216 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4217 says whether AF_INET or AF_INET6 was passed as an argument.
4219 42. Exiwhat output was messed up when time zones were included in log
4222 43. Exiwhat now gives more information about the daemon's listening ports,
4223 and whether -tls-on-connect was used.
4225 44. The "port" option of the smtp transport is now expanded.
4227 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4228 silently ignored. Now an error message is written to the main and panic
4231 46. There's a new ACL modifier called "logwrite" which writes to a log file
4232 as soon as it is encountered.
4234 47. Added $local_user_uid and $local_user_gid at routing time.
4236 48. Exim crashed when trying to verify a sender address that was being
4239 49. Exim was recognizing only a space character after ".include". It now also
4240 recognizes a tab character.
4242 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4243 extracting the relevant information from the specification. The man page no
4244 longer contains scrambled data for the -d option, and I've added a section
4245 at the front about calling Exim under different names.
4247 51. Added "extra_headers" argument to the "mail" command in filter files.
4249 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4252 53. Installed eximstats 1.29.
4254 54. Added transport_filter_timeout as a generic transport option.
4256 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4257 Cc: header lines. This was required by RFC 822, but it not required by RFC
4260 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4261 incoming messages that did not have them. Now it does so only if the
4262 message originates locally, that is, if there is no associated remote host
4263 address. When Resent- header lines are present, this applies to the Resent-
4264 lines rather than the non-Resent- lines.
4266 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4267 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4269 58. Messages for configuration errors now include the name of the main
4270 configuration file - useful now that there may be more than one file in a
4271 list (.included file names were always shown).
4273 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4274 for those rare installations that do not start the daemon as root or run it
4275 setuid root. I've cut out the call to initgroups() if the daemon is not
4278 60. The Exim user and group can now be bound into the binary as text strings
4279 that are looked up at the start of Exim's processing.
4281 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4283 62. Added $mailstore_basename variable.
4285 63. Installed patch to sieve.c from Michael Haardt.
4287 64. When Exim failed to open the panic log after failing to open the main log,
4288 the original message it was trying to log was written to stderr and debug
4289 output, but if they were not available (the usual case in production), it
4290 was lost. Now it is written to syslog before the two lines that record the
4291 failures to open the logs.
4293 65. Users' Exim filters run in subprocesses under the user's uid. It is
4294 possible for a "deliver" command or an alias in a "personal" command to
4295 provoke an address rewrite. If logging of address rewriting is configured,
4296 this fails because the process is not running as root or exim. There may be
4297 a better way of dealing with this, but for the moment (because 4.30 needs
4298 to be released), I have disabled address rewrite logging when running a
4299 filter in a non-root, non-exim process.
4305 1. The buildconfig auxiliary program wasn't quoting the value set for
4306 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4307 not defined. This bug was masked in 4.22 by the effect that was fixed in
4310 2. Some messages that were rejected after a message id was allocated were
4311 shown as "incomplete" by exigrep. It no longer does this for messages that
4312 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4314 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4315 have allow_domain_literals set, the ID did not get logged in the <= line.
4316 Domain literals are now always recognized in Message-ID: header lines.
4318 4. The first argument for a ${extract expansion item is the key name or field
4319 number. Leading and trailing spaces in this item were not being ignored,
4320 causing some misleading effects.
4322 5. When deliver_drop_privilege was set, single queue runner processes started
4323 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4324 same command in the process it spins off) were not dropping privilege.
4326 6. When the daemon running as "exim" started a queue runner, it always
4327 re-executed Exim in the spun-off process. This is a waste of effort when
4328 deliver_drop_privilege is set. The new process now just calls the
4329 queue-runner function directly.
4335 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4338 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4339 ignored. Though the use of -oP was forcing the writing of a pid file, it
4340 was always written to the default place.
4342 3. If the message "no IP address found for host xxxx" is generated during
4343 incoming verification, it is now followed by identification of the incoming
4344 connection (so you can more easily find what provoked it).
4346 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4348 5. Added some features to "harden" Exim a bit more against certain attacks:
4350 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4351 be put in Local/Makefile. This is like the never_users runtime option,
4352 but it cannot be overridden. The default setting is "root".
4354 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4355 prefix string with which any file named in a -C command line option
4358 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4359 is retained for -C and -D only if the caller of Exim is root. Without
4360 it, the exim user may also use -C and -D and retain privilege.
4362 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4363 command line option is disabled.
4365 6. Macro names set by the -D option must start with an upper case letter, just
4366 like macro names defined in the configuration file.
4368 7. Added "dereference=" facility to LDAP.
4370 8. Two instances of the typo "uknown" in the source files are fixed.
4372 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4373 the Configure-Makefile script screwed up while processing it.
4375 10. Incorporated PCRE 4.4.
4377 11. The SMTP synchronization check was not operating right at the start of an
4378 SMTP session. For example, it could not catch a HELO sent before the client
4379 waited for the greeting. There is now a check for outstanding input at the
4380 point when the greeting is written. Because of the duplex, asynchronous
4381 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4382 way, but not yet received, when the check is performed.
4384 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4385 on TCP/IP sockets, because this apparently causes some broken clients to
4388 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4389 unchanged) from the Cygwin maintainer.
4391 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4392 was supported instead of testing for mbx. Effectively a typo.
4394 15. The spa authenticator server code was not checking that the input it
4395 received was valid base64.
4397 16. The debug output line for the "set" modifier in ACLs was not showing the
4398 name of the variable that was being set.
4400 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4402 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4403 lookup on its own. Something else was needed to trigger the lookup. For
4404 example, a match in host_lookup or the need for a host name in a host list.
4405 Now, if $sender_host_name is referenced and the host name has not yet been
4406 looked up, a lookup is performed. If the lookup fails, the variable remains
4407 empty, and $host_lookup_failed is set to "1".
4409 19. Added "eqi" as a case-independent comparison operator.
4411 20. The saslauthd authentication condition could segfault if neither service
4412 nor realm was specified.
4414 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4415 error message that was logged was misleading, and incoming SMTP
4416 connections were dropped. The message is now more accurate, and temporary
4417 errors are given to SMTP connections.
4419 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4420 (except RSET) with a 421 error, until QUIT is received. However, it was
4421 failing to send a response to QUIT.
4423 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4424 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4425 ACL is now run after the helo_try_verify_hosts code.
4427 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4428 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4429 case-independent, so other case variants are also recognized). Apparently
4430 some systems use these upper case variants.
4432 25. If more than two messages were waiting for the same host, and a transport
4433 filter was specified for the transport, Exim sent two messages over the
4434 same TCP/IP connection, and then failed with "socket operation on non-
4435 socket" when it tried to send the third.
4437 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4439 27. The extern definition of crypt16() in expand.c was not being excluded when
4440 the OS had its own crypt16() function.
4442 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4443 as a preferred synonym for return_size_limit, both as an option and as an
4446 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4448 30. Changed the default configuration ACL to relax the local part checking rule
4449 for addresses that are not in any local domains. For these addresses,
4450 slashes and pipe symbols are allowed within local parts, but the sequence
4451 /../ is explicitly forbidden.
4453 31. SPA server authentication was not clearing the challenge buffer before
4456 32. log_message in a "warn" ACL statement was writing to the reject log as
4457 well as to the main log, which contradicts the documentation and doesn't
4458 seem right (because no rejection is happening). So I have stopped it.
4460 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4461 However, I am unable to do any testing of this.
4463 34. Fixed an infelicity in the appendfile transport. When checking directories
4464 for a mailbox, to see if any needed to be created, it was accidentally
4465 using path names with one or more superfluous leading slashes; tracing
4466 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4468 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4469 contents are added to the log line that is written for every discarded
4470 recipient. (Previously a log_message setting was ignored.)
4472 36. The ${quote: operator now quotes the string if it is empty.
4474 37. The install script runs exim in order to find its version number. If for
4475 some reason other than non-existence or emptiness, which it checks, it
4476 could not run './exim', it was installing it with an empty version number,
4477 i.e. as "exim-". This error state is now caught, and the installation is
4480 38. An argument was missing from the function that creates an error message
4481 when Exim fails to connect to the socket for saslauthd authentication.
4482 This could cause Exim to crash, or give a corrupted message.
4484 39. Added isip, isip4, and isip6 to ${if conditions.
4486 40. The ACL variables $acl_xx are now saved with the message, and can be
4487 accessed later in routers, transports, and filters.
4489 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4490 strings in the file are not string-expanded.
4492 42. If a MAIL command specified a SIZE value that was too large to fit into an
4493 int variable, the check against message_size_limit failed. Such values are
4494 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4495 day this will have to be increased, but I don't think I want to be around
4496 when emails are that large.
4503 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4504 iconv() is not standard in FreeBSD.
4506 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4507 IPv6 enabled. The observed symptom was a segmentation fault on return from
4508 the function os_common_find_running_interfaces() in src/os.c.
4510 3. In the check_special_case() function in daemon.c I had used "errno" as an
4511 argument name, which causes warnings on some systems. This was basically a
4512 typo, since it was named "eno" in the comments!
4514 4. The code that waits for the clock to tick (at a resolution of some fraction
4515 of a second) so as to ensure message-id uniqueness was always waiting for
4516 at least one whole tick, when it could have waited for less. [This is
4517 almost certainly not relevant at current processor speeds, where it is
4518 unlikely to ever wait at all. But we try to future-proof.]
4520 5. The function that sleeps for a time interval that includes fractions of a
4521 second contained a race. It did not block SIGALRM between setting the
4522 timer, and suspending (a couple of lines later). If the interval was short
4523 and the sigsuspend() was delayed until after it had expired, the suspension
4524 never ended. On busy systems this could lead to processes getting stuck for
4527 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4528 process, before it forks any delivery processes. The open lookup caching
4529 mechanism meant that the open file or database connection was passed into
4530 the delivery process. The problem was that delivery processes always tidy
4531 up cached lookup data. This could cause a problem for the next delivery
4532 process started by the queue runner, because the external queue runner
4533 process does not know about the closure. So the next delivery process
4534 still has data in the lookup cache. In the case of a file lookup, there was
4535 no problem because closing a file descriptor in a subprocess doesn't affect
4536 the parent. However, if the lookup was caching a connection to a database,
4537 the connection was closed, and the second delivery process was likely to
4538 see errors such as "PGSQL: query failed: server closed the connection
4539 unexpectedly". The problem has been fixed by closing all cached lookups
4540 in a queue runner before running a delivery process.
4542 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4543 seem to have the "const" qualifier which it has on other OS. I've
4546 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4547 *of the same type* (client or server) with the same public name that an
4548 error should be diagnosed.
4550 9. When Exim looked up a host name for an IP address, but failed to find the
4551 original IP address when looking up the host name (a safety check), it
4552 output the message "<ip address> does not match any IP for NULL", which was
4553 confusing, to say the least. The bug was that the host name should have
4554 appeared instead of "NULL".
4556 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4557 user that is built into the binary, and the -C or -D options is used, root
4558 privilege is dropped before the configuration file is read. In addition,
4559 logging is switched to stderr instead of the normal log files. If the
4560 configuration then re-defines the Exim user, the unprivileged environment
4561 is probably not what is expected, so Exim logs a panic warning message (but
4564 However, if deliver_drop_privilege is set, the unprivileged state may well
4565 be exactly what is intended, so the warning has been cut out in that case,
4566 and Exim is allowed to try to write to its normal log files.
4572 1. smtp_return_error_details was not giving details for temporary sender
4573 or receiver verification errors.
4575 2. Diagnose a configuration error if two authenticators have the same public
4578 3. Exim used not to create the message log file for a message until the first
4579 delivery attempt. This could be confusing when incoming messages were held
4580 for policy or load reasons. The message log file is now created at the time
4581 the message is received, and an initial "Received" line is written to it.
4583 4. The automatically generated man page for command line options had a minor
4584 bug that caused no ill effects; however, a more serious problem was that
4585 the procedure for building the man page automatically didn't always
4586 operate. Consequently, release 4.20 contains an out-of-date version. This
4587 shouldn't happen again.
4589 5. When building Exim with embedded Perl support, the script that builds the
4590 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4591 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4593 6. The freeze_tell option was not being used for messages that were frozen on
4594 arrival, either by an ACL or by local_scan().
4596 7. Added the smtp_incomplete_transaction log selector.
4598 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4599 was accepting AUTH without a new EHLO.
4601 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4602 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4605 10. Logging of TCP/IP connections (when configured) now happens in the main
4606 daemon process instead of the child process, so that the TCP/IP connection
4607 count is more accurate (but it can never be perfect).
4609 11. The use of "drop" in a nested ACL was not being handled correctly in the
4610 outer ACL. Now, if condition failure induced by the nested "drop" causes
4611 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4612 or "require"), the connection is dropped.
4614 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4615 that yield "discard" can now be used with an "accept" or a "discard" verb,
4616 but an error is generated for any others (because I can't see a useful way
4617 to define what should happen).
4619 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4620 are now processed in the same way as lines in the Exim configuration file.
4621 In particular, continuation lines are supported.
4623 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4625 15. Added -ti meaning -t -i.
4627 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4628 domains, and warn by logging if others are found.
4630 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4631 returned from GIFCONF when Exim is trying to find the list of interfaces on
4632 a host. The code in os.c has been modified to copy each ifreq to an aligned
4633 structure in all cases.
4635 Also, in some cases, the returned ifreq's were being copied to a 'struct
4636 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4637 means the last couple of bytes of an IPv6 address could be chopped if the
4638 ifreq contained only a normal sockaddr (14 bytes storage).
4640 18. Named domain lists were not supported in the hosts_treat_as_local option.
4641 An entry such as +xxxx was not recognized, and was treated as a literal
4644 19. Ensure that header lines added by a DATA ACL are included in the reject log
4645 if the ACL subsequently rejects the message.
4647 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4648 MD5 (which is deprecated).
4650 21. When testing a filter file using -bf, Exim was writing a message when it
4651 took the sender from a "From " line in the message, but it was not doing so
4652 when it took $return_path from a Return-Path: header line. It now does.
4654 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4655 with a valid header line field name (a series of printing characters
4656 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4658 23. Changed "disc" in the source to "disk" to conform to the documentation and
4659 the book and for uniformity.
4661 24. Ignore Sendmail's -Ooption=value command line item.
4663 25. When execve() failed while trying to run a command in a pipe transport,
4664 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4665 could be confused with a return value of 69 from the command itself. This
4666 has been changed to 127, the value the shell returns if it is asked to run
4667 a non-existent command. The wording for the related log line suggests a
4668 non-existent command as the problem.
4670 26. If received_header_text expands to an empty string, do not add a Received:
4671 header line to the message. (Well, it adds a token one on the spool, but
4672 marks it "old" so that it doesn't get used or transmitted.)
4674 27. Installed eximstats 1.28 (addition of -nt option).
4676 28. There was no check for failure on the call to getsockname() in the daemon
4677 code. This can fail if there is a shortage of resources on the system, with
4678 ENOMEM, for example. A temporary error is now given on failure.
4680 29. Contrary to the C standard, it seems that in some environments, the
4681 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4682 program. Exim now does this explicitly; it affects the formatting of
4683 timestamps using strftime().
4685 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4686 complaints. I've now initialized all the variables, to avoid this.
4688 32. Header lines added by a system filter were not being "seen" during
4689 transport-time rewrites.
4691 33. The info_callback() function passed to OpenSSL is set up with type void
4692 (*)(SSL *, int, int), as described somewhere. However, when calling the
4693 function (actually a macro) that sets it up, the type void(*)() is
4694 expected. I've put in a cast to prevent warnings from picky compilers.
4696 34. If a DNS black list lookup found a CNAME record, but there were no A
4697 records associated with the domain it pointed at, Exim crashed.
4699 35. If a DNS black list lookup returned more than one A record, Exim ignored
4700 all but the first. It now scans all returned addresses if a particular IP
4701 value is being sought. In this situation, the contents of the
4702 $dnslist_value variable are a list of all the addresses, separated by a
4705 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4706 to accept a host name and all its aliases if the forward lookup for any of
4707 them yielded the IP address of the incoming connection. Now it accepts only
4708 those names whose forward lookup yields the correct IP address. Any other
4709 names are discarded. This closes a loophole whereby a rogue DNS
4710 administrator could create reverse DNS records to break through a
4711 wildcarded host restriction in an ACL.
4713 37. If a user filter or a system filter that ran in a subprocess used any of
4714 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4715 the wrong values were passed to the pipe command ($thisaddress had the
4716 value of $0, $0 had the value of $1, etc). This bug was introduced by
4717 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4719 38. Improved the line breaking for long SMTP error messages from ACLs.
4720 Previously, if there was no break point between 40 and 75 characters, Exim
4721 left the rest of the message alone. Two changes have been made: (a) I've
4722 reduced the minimum length to 35 characters; (b) if it can't find a break
4723 point between 35 and 75 characters, it looks ahead and uses the first one
4724 that it finds. This may give the occasional overlong line, but at least the
4725 remaining text gets split now.
4727 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4728 file descriptors might be low, and that setting 1000 would always raise it.
4729 It turns out that in some environments, the limit is already over 1000 and
4730 that lowering it causes trouble. So now Exim takes care not to decrease it.
4732 40. When delivering a message, the value of $return_path is set to $sender_
4733 address at the start of routing (routers may change the value). By an
4734 oversight, this default was not being set up when an address was tested by
4735 -bt or -bv, which affected the outcome if any router or filter referred to
4738 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4739 log only when "message" or "log_message" are set. However, if one of the
4740 conditions was an address verification, or a call to a nested ACL, the
4741 messages generated by the underlying test were being passed through. This
4742 no longer happens. The underlying message is available in $acl_verify_
4743 message for both "message" and "log_message" expansions, so it can be
4744 passed through if needed.
4746 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4747 new expansion $bh_ to give the encoded byte string without charset
4748 translation. Translation happens only if iconv() is available; HAVE_ICONV
4749 indicates this at build time. HEADERS_CHARSET gives the charset to
4750 translate to; headers_charset can change it in the configuration, and
4751 "headers charset" can change it in an individual filter file.
4753 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4754 that creates RFC 2047 encoded "words" labels them as that charset instead
4755 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4756 expansion operator; (ii) when Exim creates a From: line for a local
4757 message; (iii) when a header line is rewritten to include a "phrase" part.
4759 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4760 buggy, causing it to skip the first lines of messages whose message ID
4761 ended in 'D'. This would not have bitten before Exim release 4.14, because
4762 message IDs were unlikely to end in 'D' before then. The effect was to have
4763 incorrect size information for certain domains.
4765 45. #include "config.h" was missing at the start of the crypt16.c module. This
4766 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4769 46. If there was a timeout during a "random" callout check, Exim treated it as
4770 a failure of the random address, and carried on sending RSET and the real
4771 address. If the delay was just some slowness somewhere, the response to the
4772 original RCPT would be taken as a response to RSET and so on, causing
4773 mayhem of various kinds.
4775 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4776 when I implemented it. It didn't allow for the fact that some option values
4777 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4778 the right test for positive values.
4780 48. Domain names in DNS records are case-independent. Exim always looks them up
4781 in lower case. Some resolvers return domain names in exactly the case they
4782 appear in the zone file, that is, they may contain uppercase letters. Not
4783 all resolvers do this - some return always lower case. Exim was treating a
4784 change of case by a resolver as a change of domain, similar to a widening
4785 of a domain abbreviation. This triggered its re-routing code and so it was
4786 trying to route what was effectively the same domain again. This normally
4787 caused routing to fail (because the router wouldn't handle the domain
4788 twice). Now Exim checks for this case specially, and just changes the
4789 casing of the domain that it ultimately uses when it transmits the message
4792 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4795 50. If a filter generated a file delivery with a non-absolute name (possible if
4796 no home directory exists for the router), the forbid_file option was not
4799 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4800 the existing equality matching.
4802 52. Exim was using ints instead of ino_t variables in some places where it was
4803 dealing with inode numbers.
4805 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4806 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4807 called TMPDIR, and if it finds it is different, it changes its value.
4809 54. The smtp_printf() function is now made available to local_scan() so
4810 additional output lines can be written before returning. There is also an
4811 smtp_fflush() function to enable the detection of a dropped connection.
4812 The variables smtp_input and smtp_batched_input are exported to
4815 55. Changed the default runtime configuration: the message "Unknown user"
4816 has been removed from the ACL, and instead placed on the localuser router,
4817 using the cannot_route_message feature. This means that any verification
4818 failures that generate their own messages won't get overridden. Similarly,
4819 the "Unrouteable address" message that was in the ACL for unverifiable
4820 relay addresses has also been removed.
4822 56. Added hosts_avoid_esmtp to the smtp transport.
4824 57. The exicyclog script was not checking for the esoteric option
4825 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4826 will work only if exicyclog is run under the appropriate euid.
4828 58. Following a discussion on the list, the rules by which Exim recognises line
4829 endings on incoming messages have been changed. The -dropcr and drop_cr
4830 options are now no-ops, retained only for backwards compatibility. The
4831 following line terminators are recognized: LF CRLF CR. However, special
4832 processing applies to CR:
4834 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4835 nor a local message in the state where . is a terminator.
4837 (ii) If a bare CR is encountered in a header line, an extra space is added
4838 after the line terminator so as not to end the header. The reasoning
4839 behind this is that bare CRs in header lines are most likely either
4840 to be mistakes, or people trying to play silly games.
4842 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4843 was being incorrectly given as 18 bytes larger than it should have been.
4844 This is a VOB (very old bug).
4846 60. This may never have affected anything current, but just in case it has:
4847 When the local host is found other than at the start of a list of hosts,
4848 the local host, those with the same MX, and any that follow, are discarded.
4849 When the list in question was part of a longer list of hosts, the following
4850 hosts (not currently being processed) were also being discarded. This no
4851 longer happens. I'm not sure if this situation could ever has previously
4854 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4857 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4858 header. This is something that is recommended in a new Internet Draft, and
4859 is something that is documented as being done by Sendmail. There are two
4860 possible values. For messages generated by the autoreply transport, Exim
4863 Auto-Submitted: auto-replied
4865 whereas for all other generated messages (e.g. bounces) it adds
4867 Auto-Submitted: auto-generated
4869 63. The "personal" condition in filters now includes a test for the
4870 Auto-Submitted: header. If it contains the string "auto-" the message it
4871 not considered personal.
4873 64. Added rcpt_include_affixes as a generic transport option.
4875 65. Added queue_only_override (default true).
4877 66. Added the syslog_duplication option.
4879 67. If what should have been the first header line of a message consisted of
4880 a space followed by a colon, Exim was mis-interpreting it as a header line.
4881 It isn't of course - it is syntactically invalid and should therefore be
4882 treated as the start of the message body. The misbehaviour could have
4883 caused a number of strange effects, including loss of data in subsequent
4884 header lines, and spool format errors.
4886 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4887 client host had authenticated. This control can now be exercised by an ACL
4888 for more flexibility.
4890 69. By default, callouts do not happen when testing with -bh. There is now a
4891 variant, -bhc, which does actually run the callout code, including
4892 consulting and updating the callout cache.
4894 70. Added support for saslauthd authentication, courtesy of Alexander
4897 71. If statvfs() failed on the spool or log directories while checking their
4898 size for availability, Exim confusingly gave the error "space shortage".
4899 Furthermore, in debugging mode it crashed with a floating point exception.
4900 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4901 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4902 serious problem, Exim now writes to the main and panic logs when this
4903 happens, with details of the failure. It then refuses to accept the
4904 incoming message, giving the message "spool directory problem" or "log
4905 directory problem" with a 421 code for SMTP messages.
4907 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4908 0, 1, and 2 exist, because some OS complain for execs without them (see
4909 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4910 descriptors. However, the code omitted to check that the open succeeded,
4911 causing mysterious errors if for some reason the permissions on /dev/null
4912 got screwed. Now Exim writes a message to the main and panic logs, and
4913 bombs out if it can't open /dev/null.
4915 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4916 interact so that it is all more flexible. It is supposed to remain
4917 backwards compatible. Also added extra_local_interfaces.
4919 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4920 to bomb out with an assertion failure - to the client this appears as a
4921 connection drop. This problem occurs in the part of the code that was taken
4922 from the Samba project. Fortunately, the assertion is in a very simple
4923 function, so I have fixed this by reproducing the function inline in the
4924 one place where it is called, and arranging for authentication to fail
4925 instead of killing the process with assert().
4927 75. The SPA client code was not working when the server requested OEM rather
4928 than Unicode encoding.
4930 76. Added code to make require_files with a specific uid setting more usable in
4931 the case where statting the file as root fails - usually a non-root-mounted
4932 NFS file system. When this happens and the failure is EACCES, Exim now
4933 forks a subprocess and does the per-uid checking as the relevant uid.
4935 77. Added process_log_path.
4937 78. If log_file_path was not explicitly set, a setting of check_log_space or
4938 check_log_inodes was ignored.
4940 79. If a space check for the spool or log partitions fails, the incident is now
4941 logged. Of course, in the latter case the data may get lost...
4943 80. Added the %p formatting code to string_format() so that it can be used to
4944 print addresses in debug_print(). Adjusted all the address printing in the
4945 debugging in store.c to use %p rather than %d.
4947 81. There was a concern that a line of code in smtp_in.c could overflow a
4948 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4949 initially expressed, the concern was not well-founded, because trailing
4950 spaces are removed early. However, if the trailing spaces were followed by
4951 a NULL, they did not get removed, so the overflow was possible. Two fixes
4954 (a) I re-wrote the offending code in a cleaner fashion.
4955 (b) If an incoming SMTP command contains a NULL character, it is rejected
4958 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4959 runs initgroups(), so that if the Exim user is in any additional groups,
4960 they will be used during message reception.
4966 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob