1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.562 2009/01/02 17:22:12 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
101 TK/01 Add preliminary DKIM support. Currently requires a forked version of
102 ALT-N's libdkim that I have put here:
103 http://duncanthrax.net/exim-experimental/
105 Note to Michael Haardt: I had to rename some vars in sieve.c. They
106 were called 'true' and it seems that C99 defines that as a reserved
107 keyword to be used with 'bool' variable types. That means you could
108 not include C99-style headers which use bools without triggering
109 build errors in sieve.c.
111 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
112 as mailq or other aliases. Changed the --help handling significantly
113 to do whats expected. exim_usage() emits usage/help information.
115 SC/01 Added the -bylocaldomain option to eximstats.
117 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
119 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
121 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
127 PH/01 Another patch from the Sieve maintainer.
129 PH/02 When an IPv6 address is converted to a string for single-key lookup
130 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
131 dots are used instead of colons so that keys in lsearch files need not
132 contain colons. This was done some time before quoting was made available
133 in lsearch files. However, iplsearch files do require colons in IPv6 keys
134 (notated using the quote facility) so as to distinguish them from IPv4
135 keys. This meant that lookups for IP addresses in host lists did not work
136 for iplsearch lookups.
138 This has been fixed by arranging for IPv6 addresses to be expressed with
139 colons if the lookup type is iplsearch. This is not incompatible, because
140 previously such lookups could never work.
142 The situation is now rather anomolous, since one *can* have colons in
143 ordinary lsearch keys. However, making the change in all cases is
144 incompatible and would probably break a number of configurations.
146 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
149 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
150 conversion specification without a maximum field width, thereby enabling
151 a rogue spamd server to cause a buffer overflow. While nobody in their
152 right mind would setup Exim to query an untrusted spamd server, an
153 attacker that gains access to a server running spamd could potentially
154 exploit this vulnerability to run arbitrary code as the Exim user.
156 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
157 $primary_hostname instead of what libspf2 thinks the hosts name is.
159 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
160 a directory entry by the name of the lookup key. Previously, if a
161 symlink pointed to a non-existing file or a file in a directory that
162 Exim lacked permissions to read, a lookup for a key matching that
163 symlink would fail. Now it is enough that a matching directory entry
164 exists, symlink or not. (Bugzilla 503.)
166 PH/03 The body_linecount and body_zerocount variables are now exported in the
169 PH/04 Added the $dnslist_matched variable.
171 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
172 This means they are set thereafter only if the connection becomes
175 PH/06 Added the client_condition to authenticators so that some can be skipped
176 by clients under certain conditions.
178 PH/07 The error message for a badly-placed control=no_multiline_responses left
179 "_responses" off the end of the name.
181 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
183 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
184 (without spaces) instead of just copying the configuration text.
186 PH/10 Added the /noupdate option to the ratelimit ACL condition.
188 PH/11 Added $max_received_linelength.
190 PH/12 Added +ignore_defer and +include_defer to host lists.
192 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
193 way in which PCRE > 7.0 is built.
195 PH/14 Implemented queue_only_load_latch.
197 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
198 MAIL command. The effect was to mangle the value on 64-bit systems.
200 PH/16 Another patch from the Sieve maintainer.
202 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
204 PH/18 If a system quota error occurred while trying to create the file for
205 a maildir delivery, the message "Mailbox is full" was not appended to the
206 bounce if the delivery eventually timed out. Change 4.67/27 below applied
207 only to a quota excession during the actual writing of the file.
209 PH/19 It seems that peer DN values may contain newlines (and other non-printing
210 characters?) which causes problems in log lines. The DN values are now
211 passed through string_printing() before being added to log lines.
213 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
214 and InterBase are left for another time.)
216 PH/21 Added message_body_newlines option.
218 PH/22 Guard against possible overflow in moan_check_errorcopy().
220 PH/23 POSIX allows open() to be a macro; guard against that.
222 PH/24 If the recipient of an error message contained an @ in the local part
223 (suitably quoted, of course), incorrect values were put in $domain and
224 $local_part during the evaluation of errors_copy.
230 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
231 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
234 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
235 issue a MAIL command.
237 PH/02 In an ACL statement such as
239 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
241 if a client was not listed at all, or was listed with a value other than
242 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
243 the condition was not true (as it should be), so access was not denied.
244 The bug was that the ! inversion was incorrectly passed on to the second
245 item. This has been fixed.
247 PH/03 Added additional dnslists conditions == and =& which are different from
248 = and & when the dns lookup returns more than one IP address.
250 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
251 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
253 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
254 FSYNC, which compiles an option called disable_fsync that allows for
255 bypassing fsync(). The documentation is heavily laced with warnings.
257 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
259 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
260 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
261 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
262 including adding "make clean"; (3) Added -fPIC when compiling the test
263 dynamically loaded module, to get rid of a warning.
265 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
266 message fails, move_frozen_messages = true and ignore_bounce_errors_after
267 = 0s. The bug is otherwise harmless.
269 PH/07 There was a bug in the dovecot authenticator such that the value of
270 $auth1 could be overwritten, and so not correctly preserved, after a
271 successful authentication. This usually meant that the value preserved by
272 the server_setid option was incorrect.
274 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
276 PH/09 Installed PCRE release 7.0.
278 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
279 run for batched SMTP input. It is now run at the start of every message
280 in the batch. While fixing this I discovered that the process information
281 (output by running exiwhat) was not always getting set for -bs and -bS
282 input. This is fixed, and it now also says "batched" for BSMTP.
284 PH/11 Added control=no_pipelining.
286 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
287 patch, slightly modified), and move the expansion of helo_data till after
288 the connection is made in the smtp transport (so it can use these
291 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
293 PH/14 Added log_selector = +pid.
295 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
297 PH/16 Add ${if forany and ${if forall.
299 PH/17 Added dsn_from option to vary the From: line in DSNs.
301 PH/18 Flush SMTP output before performing a callout, unless control =
302 no_callout_flush is set.
304 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
305 was true (the default) a successful delivery failed to delete the retry
306 item, thus causing premature timeout of the address. The bug is now
309 PH/20 Added hosts_avoid_pipelining to the smtp transport.
311 PH/21 Long custom messages for fakedefer and fakereject are now split up
312 into multiline reponses in the same way that messages for "deny" and
313 other ACL rejections are.
315 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
316 with slight modification.
318 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
319 draft, changing the syntax and factoring some duplicate code".
321 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
322 for deliveries of the second and subsequent messages over the same SMTP
325 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
326 ${reduce, with only minor "tidies".
328 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
330 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
331 expansion side effects.
333 PH/27 When a message times out after an over-quota error from an Exim-imposed
334 quota, the bounce message says "mailbox is full". This message was not
335 being given when it was a system quota that was exceeded. It now should
338 MH/03 Made $recipients available in local_scan(). local_scan() already has
339 better access to the recipient list through recipients_list[], but
340 $recipients can be useful in postmaster-provided expansion strings.
342 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
343 in the case of a MAIL command with additional options following the
344 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
345 were accidentally chopped off.
347 PH/29 SMTP synchronization checks are implemented when a command is read -
348 there is a check that no more input is waiting when there shouldn't be
349 any. However, for some commands, a delay in an ACL can mean that it is
350 some time before the response is written. In this time, more input might
351 arrive, invalidly. So now there are extra checks after an ACL has run for
352 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
353 pipelining has not been advertised.
355 PH/30 MH's patch to allow iscntrl() characters to be list separators.
357 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
358 returned in the SMTP response when smtp_return_error_details was false.
361 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
362 instead of the C I/O that was originally supplied, because problems were
365 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
366 Exim which did not show up earlier: it was assuming that a call to
367 SSL_CTX_set_info_callback() might give an error value. In fact, there is
368 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
369 was a macro that became an assignment, so it seemed to work. This has
370 changed to a proper function call with a void return, hence the compile
371 error. Exim's code has been fixed.
373 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
376 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
378 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
380 PH/37 If a message is not accepted after it has had an id assigned (e.g.
381 because it turns out to be too big or there is a timeout) there is no
382 "Completed" line in the log. When some messages of this type were
383 selected by exigrep, they were listed as "not completed". Others were
384 picked up by some special patterns. I have improved the selection
385 criteria to be more general.
387 PH/38 The host_find_failed option in the manualroute router can now be set
388 to "ignore", to completely ignore a host whose IP address cannot be
389 found. If all hosts are ignored, the behaviour is controlled by the new
390 host_all_ignored option.
392 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
393 homing or because of multiple MX records with /mx) generated more than
394 one IP address, and the following item turned out to be the local host,
395 all the secondary addresses of the first item were incorrectly removed
396 from the list, along with the local host and any following hosts (which
397 is what is supposed to happen).
399 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
400 whoever called Exim into the -H file. In the case of the daemon it was
401 behaving confusingly. When first started, it used values for whoever
402 started the daemon, but after a SIGHUP it used the Exim user (because it
403 calls itself on a restart). I have changed the code so that it now always
406 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
407 message are rejected with the same error (e.g. no authentication or bad
408 sender address), and a DATA command is nevertheless sent (as can happen
409 with PIPELINING or a stupid MUA), the error message that was given to the
410 RCPT commands is included in the rejection of the DATA command. This is
411 intended to be helpful for MUAs that show only the final error to their
414 PH/42 Another patch from the Sieve maintainer.
416 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
417 Eximstats - Fixed some broken HTML links and added missing column headers
419 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
420 columns for Rejects, Temp Rejects, Ham, and Spam rows.
422 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
424 PH/43 Yet another patch from the Sieve maintainer.
426 PH/44 I found a way to check for a TCP/IP connection going away before sending
427 the response to the final '.' that terminates a message, but only in the
428 case where the client has not sent further data following the '.'
429 (unfortunately, this is allowed). However, in many cases there won't be
430 any further data because there won't be any more messages to send. A call
431 to select() can be used: if it shows that the input is "ready", there is
432 either input waiting, or the socket has been closed. An attempt to read
433 the next input character can distinguish the two cases. Previously, Exim
434 would have sent an OK response which the client would never have see.
435 This could lead to message repetition. This fix should cure that, at
436 least in a lot of common cases.
438 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
439 advertised in response to EHLO.
445 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
446 fixed by 4.65/MH/01 (is this a record?) are fixed:
448 (i) An empty string was always treated as zero by the numeric comparison
449 operators. This behaviour has been restored.
451 (ii) It is documented that the numeric comparison operators always treat
452 their arguments as decimal numbers. This was broken in that numbers
453 starting with 0 were being interpreted as octal.
455 While fixing these problems I realized that there was another issue that
456 hadn't been noticed. Values of message_size_limit (both the global option
457 and the transport option) were treated as octal if they started with 0.
458 The documentation was vague. These values are now always treated as
459 decimal, and I will make that clear in the documentation.
465 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
466 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
469 MH/01 Don't check that the operands of numeric comparison operators are
470 integers when their expansion is in "skipping" mode (fixes bug
471 introduced by 4.64-PH/07).
473 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
474 child addresses, Exim now panics and dies. Previously, because the count
475 is held in a short int, deliveries were likely to be lost. As such a
476 large number of recipients for a single message is ridiculous
477 (performance will be very, very poor), I have chosen to impose a limit
478 rather than extend the field.
484 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
485 leftover -K file (the existence of which was triggered by #402).
486 While we were at it, introduced process PID as part of the -K
487 filename. This should rule out race conditions when creating
490 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
491 processing considerably. Previous code took too long for large mails,
492 triggering a timeout which in turn triggers #401.
494 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
495 in the DK code in transports.c. sendfile() is not really portable,
496 hence the _LINUX specificness.
498 TF/01 In the add_headers option to the mail command in an Exim filter,
499 there was a bug that Exim would claim a syntax error in any
500 header after the first one which had an odd number of characters
503 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
504 callout verification, Exim cached a "reject" for the entire domain. This
505 is correct for most verifications, but it is not correct for a recipient
506 verification with use_sender or use_postmaster set, because in that case
507 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
508 case of MAIL FROM:<> rejection from other early rejections (e.g.
509 rejection of HELO). When verifying a recipient using a non-null MAIL
510 address, the cache is ignored if it shows MAIL FROM:<> rejection.
511 Whatever the result of the callout, the value of the domain cache is
512 left unchanged (for any other kind of callout, getting as far as trying
513 RCPT means that the domain itself is ok).
515 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
518 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
519 manifest itself as EPIPE rather than ECONNECT. When tidying away a
520 session, the daemon ignores ECONNECT errors and logs others; it now
521 ignores EPIPE as well.
523 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
524 (quoted-printable decoding).
526 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
527 later the small subsequent patch to fix an introduced bug.
529 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
531 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
533 PH/08 An error is now given if message_size_limit is specified negative.
535 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
536 to be given (somewhat) arbitrary names.
538 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
541 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
542 miscellaneous code fixes
544 PH/10 Added the log_reject_target ACL modifier to specify where to log
547 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
548 hostname. This is wrong, because it relates to the incoming message (and
549 probably the interface on which it is arriving) and not to the outgoing
550 callout (which could be using a different interface). This has been
551 changed to use the value of the helo_data option from the smtp transport
552 instead - this is what is used when a message is actually being sent. If
553 there is no remote transport (possible with a router that sets up host
554 addresses), $smtp_active_hostname is used.
556 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
557 tweaks were necessary in order to get it to work (see also 21 below):
558 (a) The code assumed that strncpy() returns a negative number on buffer
559 overflow, which isn't the case. Replaced with Exim's string_format()
561 (b) There were several signed/unsigned issues. I just did the minimum
562 hacking in of casts. There is scope for a larger refactoring.
563 (c) The code used strcasecmp() which is not a standard C function.
564 Replaced with Exim's strcmpic() function.
565 (d) The code set only $1; it now sets $auth1 as well.
566 (e) A simple test gave the error "authentication client didn't specify
567 service in request". It would seem that Dovecot has changed its
568 interface. Fortunately there's a specification; I followed it and
569 changed what the client sends and it appears to be working now.
571 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
574 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
575 address A is aliased to B and C, where B exists and C does not. Without
576 -v the output is "A verified" because verification stops after a
577 successful redirection if more than one address is generated. However,
578 with -v the child addresses are also verified. Exim was outputting "A
579 failed to verify" and then showing the successful verification for C,
580 with its parentage. It now outputs "B failed to verify", showing B's
581 parentage before showing the successful verification of C.
583 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
584 look up a TXT record in a specific list after matching in a combined
587 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
588 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
589 they consult the DNS. I had assumed they would set it the way they
590 wanted; and indeed my experiments on Linux seem to show that in some
591 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
592 To be on the safe side, however, I have now made the interface to
593 host_find_byname() similar to host_find_bydns(), with an argument
594 containing the DNS resolver options. The host_find_byname() function now
595 sets these options at its start, just as host_find_bydns() does. The smtp
596 transport options dns_qualify_single and dns_search_parents are passed to
597 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
598 of host_find_byname() use the default settings of RES_DEFNAMES
599 (qualify_single) but not RES_DNSRCH (search_parents).
601 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
602 spool_read_header() do less string testing, by means of a preliminary
603 switch on the second character of optional "-foo" lines. (This is
604 overdue, caused by the large number of possibilities that now exist.
605 Originally there were few.) While I was there, I also converted the
606 str(n)cmp tests so they don't re-test the leading "-" and the first
607 character, in the hope this might squeeze out yet more improvement.
609 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
610 flag allowing group syntax was set by the header_syntax check but not
611 turned off, possible causing trouble later; (2) The flag was not being
612 set at all for the header_verify test, causing "group"-style headers to
613 be rejected. I have now set it in this case, and also caused header_
614 verify to ignore an empty address taken from a group. While doing this, I
615 came across some other cases where the code for allowing group syntax
616 while scanning a header line wasn't quite right (mostly, not resetting
617 the flag correctly in the right place). These bugs could have caused
618 trouble for malformed header lines. I hope it is now all correct.
620 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
621 with the "reply" argument non-NULL. The code, however (which originally
622 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
623 but it didn't always do it. This confused somebody who was copying the
624 code for some other use. I have removed all the tests.
626 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
627 feature that was used to support insecure browsers during the U.S. crypto
628 embargo. It requires special client support, and Exim is probably the
629 only MTA that supported it -- and would never use it because real RSA is
630 always available. This code has been removed, because it had the bad
631 effect of slowing Exim down by computing (never used) parameters for the
632 RSA_EXPORT functionality.
634 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
635 authenticator to fail if there's a tab character in the incoming data
636 (there should never be unless someone is messing about, as it's supposed
637 to be base64-encoded). Also added, on Timo's advice, the "secured" option
638 if the connection is using TLS or if the remote IP is the same as the
639 local IP, and the "valid-client-cert option" if a client certificate has
642 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
643 authenticators. This can be used for authorization after authentication
644 succeeds. (In the case of plaintext, it servers for both authentication
647 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
648 if any retry times were supplied.
650 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
651 connection, where there is no client IP address to check. In this
652 situation, the verify now always succeeds.
654 PH/25 Applied John Jetmore's -Mset patch.
656 PH/26 Added -bem to be like -Mset, but loading a message from a file.
658 PH/27 In a string expansion for a processed (not raw) header when multiple
659 headers of the same name were present, leading whitespace was being
660 removed from all of them, but trailing whitespace was being removed only
661 from the last one. Now trailing whitespace is removed from each header
662 before concatenation. Completely empty headers in a concatenation (as
665 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
666 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
668 PH/29 [Removed. This was a change that I later backed out, and forgot to
669 correct the ChangeLog entry (that I had efficiently created) before
670 committing the later change.]
672 PH/30 Exim was sometimes attempting to deliver messages that had suffered
673 address errors (4xx response to RCPT) over the same connection as other
674 messages routed to the same hosts. Such deliveries are always "forced",
675 so retry times are not inspected. This resulted in far too many retries
676 for the affected addresses. The effect occurred only when there were more
677 hosts than the hosts_max_try setting in the smtp transport when it had
678 the 4xx errors. Those hosts that it had tried were not added to the list
679 of hosts for which the message was waiting, so if all were tried, there
680 was no problem. Two fixes have been applied:
682 (i) If there are any address or message errors in an SMTP delivery, none
683 of the hosts (tried or untried) are now added to the list of hosts
684 for which the message is waiting, so the message should not be a
685 candidate for sending over the same connection that was used for a
686 successful delivery of some other message. This seems entirely
687 reasonable: after all the message is NOT "waiting for some host".
688 This is so "obvious" that I'm not sure why it wasn't done
689 previously. Hope I haven't missed anything, but it can't do any
690 harm, as the worst effect is to miss an optimization.
692 (ii) If, despite (i), such a delivery is accidentally attempted, the
693 routing retry time is respected, so at least it doesn't keep
694 hammering the server.
696 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
697 in ${readsocket because some servers need this prod.
699 PH/32 Added some extra debug output when updating a wait-xxx database.
701 PH/33 The hint "could be header name not terminated by colon", which has been
702 given for certain expansion errors for a long time, was not being given
703 for the ${if def:h_colon_omitted{... case.
705 PH/34 The spec says: "With one important exception, whenever a domain list is
706 being scanned, $domain contains the subject domain." There was at least
707 one case where this was not true.
709 PH/35 The error "getsockname() failed: connection reset by peer" was being
710 written to the panic log as well as the main log, but it isn't really
711 panic-worthy as it just means the connection died rather early on. I have
712 removed the panic log writing for the ECONNRESET error when getsockname()
715 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
716 runs only) independently of the message's sender address. This meant
717 that, if the 4xx error was in fact related to the sender, a different
718 message to the same recipient with a different sender could confuse
719 things. In particualar, this can happen when sending to a greylisting
720 server, but other circumstances could also provoke similar problems.
721 I have changed the default so that the retry time for these errors is now
722 based a combination of the sender and recipient addresses. This change
723 can be overridden by setting address_retry_include_sender=false in the
726 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
727 remote server are returned as part of bounce messages. This was not
728 happening for LMTP over a pipe (the lmtp transport), but now it is the
729 same for both kinds of LMTP.
731 PH/38 Despite being documented as not happening, Exim was rewriting addresses
732 in header lines that were in fact CNAMEs. This is no longer the case.
734 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
735 and queue runs started by the daemon processed all messages. This has
736 been fixed so that -R and -S can now usefully be given with -q<time>.
738 PH/40 Import PCRE release 6.7 (fixes some bugs).
740 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
742 PH/42 Give an error if -q is specified more than once.
744 PH/43 Renamed the variables $interface_address and $interface_port as
745 $received_ip_address and $received_port, to make it clear that these
746 values apply to message reception, and not to the outgoing interface when
747 a message is delivered. (The old names remain recognized, of course.)
749 PH/44 There was no timeout on the connect() call when using a Unix domain
750 socket in the ${readsocket expansion. There now is.
752 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
753 be meaningful with "accept".
755 SC/01 Eximstats V1.43
756 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
758 SC/02 Eximstats V1.44
759 Use a glob alias rather than an array ref in the generated
760 parser. This improves both readability and performance.
762 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
763 Collect SpamAssassin and rejection statistics.
764 Don't display local sender or destination tables unless
765 there is data to show.
766 Added average volumes into the top table text output.
768 SC/04 Eximstats V1.46
769 Collect data on the number of addresses (recipients)
770 as well as the number of messages.
772 SC/05 Eximstats V1.47
773 Added 'Message too big' to the list of mail rejection
774 reasons (thanks to Marco Gaiarin).
776 SC/06 Eximstats V1.48
777 Mainlog lines which have GMT offsets and are too short to
778 have a flag are now skipped.
780 SC/07 Eximstats V1.49 (Alain Williams)
781 Added the -emptyok flag.
783 SC/08 Eximstats V1.50
784 Fixes for obtaining the IP address from reject messages.
786 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
787 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
788 whitesspace changes from 4.64-PH/27
790 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
793 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
794 are found, allow negative numbers in numeric criteria)
796 JJ/06 exipick.20061117.2, added new $message_body_missing variable
798 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
799 to match changes made in 4.64-PH/43
801 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
803 PH/47 Put in an explicit test for a DNS lookup of an address record where the
804 "domain" is actually an IP address, and force a failure. This locks out
805 those revolvers/nameservers that support "A-for-A" lookups, in
806 contravention of the specifications.
808 PH/48 When a host name was looked up from an IP address, and the subsequent
809 forward lookup of the name timed out, the host name was left in
810 $sender_host_name, contrary to the specification.
812 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
813 restricted to single-key lookups, Exim was not diagnosing an error if
814 * or *@ was used with a query-style lookup.
816 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
818 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
819 long ago, but noone interested enough thought of it. Let's just say that
820 the "1.1" means that there are some new functions that weren't there at
821 some point in the past.
823 PH/51 Error processing for expansion failure of helo_data from an smtp
824 transport during callout processing was broken.
826 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
827 tested/used via the -bh/-bhc/-bs options.
829 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
830 bug, fixed in subsequent PCRE releases).
832 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
833 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
835 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
841 SC/01 Use a glob alias rather than an array ref in eximstats generated
842 parser. This improves both readability and performance.
844 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
845 Don't display local sender or destination tables in eximstats unless
846 there is data to show.
847 Added average volumes into the eximstats top table text output.
849 SC/03 Collect data on the number of addresses (recipients) as well
850 as the number of messages in eximstats.
852 TF/01 Correct an error in the documentation for the redirect router. Exim
853 does (usually) call initgroups() when daemonizing.
855 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
856 with consistent privilege compared to when running as a daemon.
858 TF/03 Note in the spec that $authenticated_id is not set for local
859 submissions from trusted users.
861 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
862 Thanks to Dean Brooks <dean@iglou.com> for the patch.
864 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
865 by adding some example configuration directives to the default
866 configuration file. A little bit of work is required to uncomment the
867 directives and define how usernames and passwords are checked, but
868 there is now a framework to start from.
870 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
871 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
872 without this. I don't know how relevant this is to other LDAP libraries.
874 PH/02 Add the verb name to the "unknown ACL verb" error.
876 PH/03 Magnus Holmgren's patch for filter_prepend_home.
878 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
880 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
881 directory not expanded when it should be if an expanded home directory
882 was set for the address (which is overridden by the transport).
884 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
887 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
888 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
889 because it is too late at that time, and has no effect.
891 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
892 security issue with \' (bugzilla #107). I could not use the
893 PQescapeStringConn() function, because it needs a PGconn value as one of
896 PH/08 When testing addresses using -bt, indicate those final addresses that
897 are duplicates that would not cause an additional delivery. At least one
898 person was confused, thinking that -bt output corresponded to deliveries.
899 (Suppressing duplicates isn't a good idea as you lose the information
900 about possibly different redirections that led to the duplicates.)
902 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
903 systems where poll() doesn't work, in particular OS X.
905 PH/10 Added more information to debugging output for retry time not reached.
907 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
908 operations in malware.c.
910 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
913 PH/13 If write_rejectlog was set false when logging was sent to syslog with
914 syslog_duplication set false, log lines that would normally be written
915 both the the main log and to the reject log were not written to syslog at
918 PH/14 In the default configuration, change the use of "message" in ACL warn
919 statements to "add_header".
921 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
922 not followed by a command (e.g. "seen endif").
924 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
925 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
928 PH/17 Added extra conditions to the default value of delay_warning_condition
932 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
933 { match{$h_precedence:}{(?i)bulk|list|junk} } \
934 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
937 The Auto-Submitted: and various List- headers are standardised, whereas I
938 don't think Precedence: ever was.
940 PH/18 Refactored debugging code in route_finduser() to show more information,
941 in particular, the error code if getpwnam() issues one.
943 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
944 This is apparently needed in addition to the PH/07 change above to avoid
945 any possible encoding problems.
947 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
948 but not after initializing Perl.
950 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
951 output them only if debugging. By default they are written stderr,
952 apparently, which is not desirable.
954 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
957 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
960 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
962 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
963 authenticated or an ident call has been made. Suppress the default
964 values for $authenticated_id and $authenticated_sender (but permit -oMai
965 and -oMas) when testing with -bh.
967 PH/24 Re-jigged the order of the tests in the default configuration so that the
968 tests for valid domains and recipients precede the DNS black list and CSA
969 tests, on the grounds that those ones are more expensive.
971 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
972 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
973 This bug exists in every version of Exim that I still have, right back to
976 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
977 However, an attempt to turn on case-sensitivity in a regex key by
978 including (?-i) didn't work because the subject string was already
979 lowercased, and the effects were non-intuitive. It turns out that a
980 one-line patch can be used to allow (?-i) to work as expected.
986 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
987 other effects) broke the use of negated acl sub-conditions.
989 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
992 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
993 "Deny" causes Exim to reject the incoming connection with a 554 error.
994 Unfortunately, if there is a major crisis, such as a disk failure,
995 tcp-wrappers gives "deny", whereas what one would like would be some
996 kind of temporary error. A kludge has been added to help with this.
997 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
998 554 error is used if errno is still zero or contains ENOENT (which occurs
999 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1002 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1004 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1005 errors. Otherwise a message that provokes a temporary error (when other
1006 messages do not) can cause a whole host to time out.
1008 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1009 addresses were routed directly to files or pipes from a redirect router.
1010 File deliveries just didn't batch; pipe deliveries might have suffered
1013 PH/06 A failure to get a lock for a hints database would erroneously always say
1014 "Failed to get write lock", even when it was really a read lock.
1016 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1017 of 0600. This has been changed to use the value of the lockfile_mode
1018 option (which defaults to 0600).
1020 PH/08 Applied small patch from the Sieve maintainer.
1022 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1023 folder from quota calculations, a direct delivery into this folder messed
1024 up the contents of the maildirsize file. This was because the regex was
1025 used only to exclude .Trash (or whatever) when the size of the mailbox
1026 was calculated. There was no check that a delivery was happening into an
1027 excluded directory. This bug has been fixed by ignoring all quota
1028 processing for deliveries into excluded directories.
1030 PH/10 Added the maildirfolder_create_regex option to appendfile.
1036 PH/01 The code for finding all the local interface addresses on a FreeBSD
1037 system running IPv6 was broken. This may well have applied to all BSD
1038 systems, as well as to others that have similar system calls. The broken
1039 code found IPv4 interfaces correctly, but gave incorrect values for the
1040 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1041 that it would not match correctly against @[] and not recognize the IPv6
1044 PH/02 The ipliteral router was not recognizing addresses of the form user@
1045 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1047 PH/03 Added disable_ipv6.
1049 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1050 decoded form, because it is most often used to construct To: headers
1051 lines in autoreplies, and the decoded form may well be syntactically
1052 invalid. However, $reply_address has leading white space removed, and all
1053 newlines turned into spaces so that the autoreply transport does not
1056 PH/05 If group was specified without a user on a router, and no group or user
1057 was specified on a transport, the group from the router was ignored.
1059 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1060 for visible compile-time settings that can be used to change these
1061 numbers, for those that want even more. Backwards compatibility with old
1062 spool files has been maintained. However, going back to a previous Exim
1063 release will lost any variables that are in spool files.
1065 PH/07 Two small changes when running in the test harness: increase delay when
1066 passing a TCP/IP connection to a new process, in case the original
1067 process has to generate a bounce, and remove special handling of
1068 127.0.0.2 (sic), which is no longer necessary.
1070 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1071 be the same on different OS.
1073 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1076 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1077 whether --show-vars was specified or not
1079 JJ/02 exipick: Added support for new ACL variable spool format introduced
1082 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1083 syntactically invalid From: or Reply-to: line, and a filter used this to
1084 generate an autoreply, and therefore failed to obtain an address for the
1085 autoreply, Exim could try to deliver to a non-existent relative file
1086 name, causing unrelated and misleading errors. What now happens is that
1087 it logs this as a hard delivery error, but does not attempt to create a
1090 PH/11 The exinext utility has a -C option for testing purposes, but although
1091 the given file was scanned by exinext itself; it wasn't being passed on
1092 when Exim was called.
1094 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1095 an end-of-file indication when reading a command response.
1097 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1098 compiled. In many other places in Exim, IPv6 addresses are always
1099 recognized, so I have changed this. It also means that IPv4 domain
1100 literals of the form [IPV4:n.n.n.n] are now always recognized.
1102 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1103 used if the router is not running as root, for example, when verifying at
1104 ACL time, or when using -bh. The debugging output from this situation was
1105 non-existent - all you got was a failure to exec. I have made two
1108 (a) Failures to set uid/gid, the current directory, or a process leader
1109 in a subprocess such as that created by queryprogram now generate
1110 suitable debugging ouput when -d is set.
1112 (b) The queryprogram router detects when it is not running as root,
1113 outputs suitable debugging information if -d is set, and then runs
1114 the subprocess without attempting to change uid/gid.
1116 PH/15 Minor change to Makefile for building test_host (undocumented testing
1119 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1120 additional section of a DNS packet that returns MX or SRV records.
1121 Instead, it always explicitly searches for A/AAAA records. This avoids
1122 major problems that occur when a DNS server includes only records of one
1123 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1124 fixed another bug: if SRV records were looked up and the corresponding
1125 address records were *not* found in the additional section, the port
1126 values from the SRV records were lost.
1128 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1129 using the correct key (the original address) when searching the retry
1130 rules in order to find which one to use for generating the retry hint.
1132 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1133 adding the default one. Similarly, if it contains a Reply-To: header, the
1134 errors_reply_to option, if set, is not used.
1136 PH/19 When calculating a retry time, Exim used to measure the "time since
1137 failure" by looking at the "first failed" field in the retry record. Now
1138 it does not use this if it is later than than the arrival time of the
1139 message. Instead it uses the arrival time. This makes for better
1140 behaviour in cases where some deliveries succeed, thus re-setting the
1141 "first failed" field. An example is a quota failure for a huge message
1142 when small messages continue to be delivered. Without this change, the
1143 "time since failure" will always be short, possible causing more frequent
1144 delivery attempts for the huge message than are intended.
1145 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1147 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1148 $1, $2, $3) because the numerical variables can be reset during some
1149 expansion items (e.g. "match"), thereby losing the authentication data.
1151 PH/21 Make -bV show the size of off_t variables so that the test suite can
1152 decide whether to run tests for quotas > 2G.
1154 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1155 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1156 filecount value is greater than 2G or if a quota value is greater than 2G
1157 on a system where the size of off_t is not greater than 4, a panic error
1160 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1161 never match. The debug and -bh output now contains an explicit error
1162 message indicating a malformed IPv4 address or mask.
1164 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1165 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1166 PH/23 above applies.
1168 PH/25 Do not write to syslog when running in the test harness. The only
1169 occasion when this arises is a failure to open the main or panic logs
1170 (for which there is an explicit test).
1172 PH/26 Added the /no_tell option to "control=freeze".
1174 PH/27 If a host name lookup failed very early in a connection, for example, if
1175 the IP address matched host_lookup and the reverse lookup yielded a name
1176 that did not have a forward lookup, an error message of the form "no IP
1177 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1178 could be logged. Now it outputs the IP address instead of "NULL".
1180 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1181 allows the sender and the authenticated sender to be set when
1182 submitting a message from within Exim. Since child_open_exim() is
1183 documented for local_scan(), the new function should be too.
1185 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1186 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1187 results in an empty string is now treated as unset.
1189 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1191 PH/31 Added sender_verify_fail logging option.
1193 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1194 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1195 tidied the source and removed it altogether.
1197 PH/33 When a queue run was abandoned because the load average was too high, a
1198 log line was always written; now it is written only if the queue_run log
1199 selector is set. In addition, the log line for abandonment now contains
1200 information about the queue run such as the pid. This is always present
1201 in "start" and "stop" lines but was omitted from the "abandon" line.
1203 PH/34 Omit spaces between a header name and the colon in the error message that
1204 is given when verify = headers_syntax fails (if there are lots of them,
1205 the message gets confusing).
1207 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1208 names, as there are now some PTR records that contain slashes. This check
1209 is only to protect against broken name servers that fall over on strange
1210 characters, so the fact that it applies to all lookups doesn't matter.
1212 PH/36 Now that the new test suite is complete, we can remove some of the
1213 special code in Exim that was needed for the old test suite. For example,
1214 sorting DNS records because real resolvers return them in an arbitrary
1215 order. The new test suite's fake resolver always returns records in the
1218 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1219 bounces) except when queue_only is set, to avoid logging races between
1220 the different processes.
1222 PH/38 Panic-die if .include specifies a non-absolute path.
1224 PH/39 A tweak to the "H" retry rule from its user.
1226 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1227 a label. They prevented compilation on older perls.
1229 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1230 a warning to be raised on newish perls.
1232 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1233 on queue. Changes to match documented behaviour of showing count of
1234 messages matching specified criteria.
1236 PH/40 Changed the default ident timeout from 30s to 5s.
1238 PH/41 Added support for the use of login_cap features, on those BSD systems
1239 that have them, for controlling the resources used by pipe deliveries.
1241 PH/42 The content-scanning code uses fopen() to create files in which to put
1242 message data. Previously it was not paying any attention to the mode of
1243 the files. Exim runs with umask(0) because the rest of the code creates
1244 files with open(), and sets the required mode explicitly. Thus, these
1245 files were ending up world-writeable. This was not a big issue, because,
1246 being within the spool directory, they were not world-accessible. I have
1247 created a function called modefopen, which takes an additional mode
1248 argument. It sets umask(777), creates the file, chmods it to the required
1249 mode, then resets the umask. All the relevant calls to fopen() in the
1250 content scanning code have been changed to use this function.
1252 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1253 to 24 hours. This avoids potential overflow problems when processing G
1254 and H retry rules. I suspect nobody ever tinkers with this value.
1256 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1258 PH/45 When the plaintext authenticator is running as a client, the server's
1259 challenges are checked to ensure they are valid base64 strings. By
1260 default, the authentication attempt is cancelled if an invalid string is
1261 received. Setting client_ignore_invalid_base64 true ignores these errors.
1262 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1263 they are received. Thus, the responses can be made to depend on the
1264 challenges. If an invalid string is ignored, an empty string is placed in
1267 PH/46 Messages that are created by the autoreply transport now contains a
1268 References: header, in accordance with RFCs 2822 and 3834.
1270 PH/47 Added authenticated_sender_force to the smtp transport.
1272 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1274 PH/49 Installed latest patch from the Sieve maintainer.
1276 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1277 was also set, the appendfile transport was unnecessarily scanning a
1278 directory of message files (e.g. for maildir delivery) to find the count
1279 of files (along with the size), even though it did not need this
1280 information. It now does the scan only if it needs to find either the
1281 size of the count of files.
1283 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1285 PH/52 Two bugs concerned with error handling when the smtp transport is
1288 (i) Exim was not creating retry information for temporary errors given
1289 for individual recipients after the DATA command when the smtp transport
1290 was used in LMTP mode. This meant that they could be retried too
1291 frequently, and not timed out correctly.
1293 (ii) Exim was setting the flag that allows error details to be returned
1294 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1295 recipients that were returned after the DATA command.
1297 PH/53 This is related to PH/52, but is more general: for any failing address,
1298 when detailed error information was permitted to be returned to the
1299 sender, but the error was temporary, then after the final timeout, only
1300 "retry timeout exceeded" was returned. Now it returns the full error as
1301 well as "retry timeout exceeded".
1303 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1304 do this, and (what is worse) MTAs that accept it.
1306 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1307 will now be deprecated.
1309 PH/56 New os.c-cygwin from the Cygwin maintainer.
1311 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1312 formats (previously only available in exim formats via -bpr, -bpru,
1313 and -bpra. Now also available in native and exiqgrep formats)
1315 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1316 with very large, slow to parse queues
1318 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1320 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1322 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1323 responses to authentication challenges, though it was showing the
1324 challenges; (ii) I've removed the CR characters from the debug output for
1327 PH/58 Allow for the insertion of a newline as well as a space when a string
1328 is turned into more than one encoded-word during RFC 2047 encoding. The
1329 Sieve code now uses this.
1331 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1332 data_4xx, lost_connection, tls_required.
1334 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1335 message was being sent as an SMTP response.
1337 PH/61 Add -l and -k options to exicyclog.
1339 PH/62 When verifying, if an address was redirected to one new address, so that
1340 verification continued, and the new address failed or deferred after
1341 having set something in $address_data, the value of $address_data was not
1342 passed back to the ACL. This was different to the case when no
1343 redirection occurred. The value is now passed back in both cases.
1345 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1346 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1347 use login_cap.h, so on its own it isn't the distinguishing feature. The
1348 new name refers directly to the setclassresources() function.
1350 PH/65 Added configuration files for NetBSD3.
1352 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1354 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1355 is preferred over IPv4.
1357 PH/68 The bounce_return_message and bounce_return_body options were not being
1358 honoured for bounces generated during the reception of non-SMTP messages.
1359 In particular, this applied to messages rejected by the ACL. This bug has
1360 been fixed. However, if bounce_return_message is true and bounce_return_
1361 body is false, the headers that are returned for a non-SMTP message
1362 include only those that have been read before the error was detected.
1363 (In the case of an ACL rejection, they have all been read.)
1365 PH/69 The HTML version of the specification is now built in a directory called
1366 spec_html instead of spec.html, because the latter looks like a path with
1367 a MIME-type, and this confuses some software.
1369 PH/70 Catch two compiler warnings in sieve.c.
1371 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1372 function verify_get_ident() calls ip_connect() to connect a socket, but
1373 if the "connect()" function timed out, ip_connect() used to close the
1374 socket. However, verify_get_ident() also closes the socket later, and in
1375 between Exim writes to the log, which may get opened at this point. When
1376 the socket was closed in ip_connect(), the log could get the same file
1377 descriptor number as the socket. This naturally causes chaos. The fix is
1378 not to close the socket in ip_connect(); the socket should be closed by
1379 the function that creates it. There was only one place in the code where
1380 this was missing, in the iplookup router, which I don't think anybody now
1381 uses, but I've fixed it anyway.
1383 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1384 well as to direct DNS lookups. Otherwise the handling of names in host
1385 lists is inconsistent and therefore confusing.
1391 PH/01 Two changes to the default runtime configuration:
1393 (1) Move the checks for relay_from_hosts and authenticated clients from
1394 after to before the (commented out) DNS black list checks.
1396 (2) Add control=submission to the relay_from_hosts and authenticated
1397 clients checks, on the grounds that messages accepted by these
1398 statements are most likely to be submissions.
1400 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1402 (1) Generate an error if the third argument for the ${prvs expansion is
1405 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1408 (3) Reset the variables that are obtained from the first argument of
1409 ${prvscheck and used in the second argument before leaving the code,
1410 because their memory is reclaimed, so using them afterwards may do
1413 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1414 one (it's much easier than Tom thought :-).
1416 (5) Because of (4), we can now allow for the use of $prvscheck_result
1417 inside the third argument.
1419 PH/03 For some reason, the default setting of PATH when running a command from
1420 a pipe transport was just "/usr/bin". I have changed it to
1423 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1424 anything to be listed in the output from -bV.
1426 PH/05 When a filter generated an autoreply, the entire To: header line was
1427 quoted in the delivery log line, like this:
1429 => >A.N.Other <ano@some.domain> <original@ddress> ...
1431 This has been changed so that it extracts the operative address. There
1432 may be more than one such address. If so, they are comma-separated, like
1435 => >ano@some.domain,ona@other.domain <original@ddress> ...
1437 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1438 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1439 not being looked up in the rDNS to get a host name, Exim was showing the
1440 IP address twice in Received: lines, even though the IP addresses were
1441 identical. For example:
1443 Received: from [1.2.3.4] (helo=[1.2.3.4])
1445 However, if the real host name was known, it was omitting the HELO data
1446 if it matched the actual IP address. This has been tidied up so that it
1447 doesn't show the same IP address twice.
1449 PH/07 When both +timestamp and +memory debugging was on, the value given by
1450 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1451 called by the debug printing, thereby overwriting the timestamp buffer.
1452 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1454 PH/08 When the original message was included in an autoreply transport, it
1455 always said "this is a copy of the message, including all the headers",
1456 even if body_only or headers_only was set. It now gives an appropriate
1459 PH/09 Applied a patch from the Sieve maintainer which:
1461 o fixes some comments
1462 o adds the (disabled) notify extension core
1463 o adds some debug output for the result of if/elsif tests
1464 o points to the current vacation draft in the documentation
1465 and documents the missing references header update
1469 o fixes a bug in processing the envelope test (when testing
1470 multiple envelope elements, the last element determinted the
1473 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1474 Electronic Mail") by including:
1476 Auto-submitted: auto-generated
1478 in the messages that it generates (bounce messages and others, such as
1479 warnings). In the case of bounce messages for non-SMTP mesages, there was
1480 also a typo: it was using "Auto_submitted" (underscore instead of
1481 hyphen). Since every message generated by Exim is necessarily in response
1482 to another message, thes have all been changed to:
1484 Auto-Submitted: auto-replied
1486 in accordance with these statements in the RFC:
1488 The auto-replied keyword:
1490 - SHOULD be used on messages sent in direct response to another
1491 message by an automatic process,
1493 - MUST NOT be used on manually-generated messages,
1495 - MAY be used on Delivery Status Notifications (DSNs) and Message
1496 Disposition Notifications (MDNs),
1498 - MUST NOT be used on messages generated by automatic or periodic
1499 processes, except for messages which are automatic responses to
1502 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1503 to the default Received: header definition.
1505 PH/12 Added log selector acl_warn_skipped (default on).
1507 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1508 variables because (a) they are in the wrong storage pool and (b) even if
1509 they were copied, it wouldn't work properly because of the caching.
1511 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1512 checking when decoding. Apparently there are clients that generate
1513 overlong encoded strings. Why am I not surprised?
1515 PH/15 If the first argument of "${if match_address" was not empty, but did not
1516 contain an "@" character, Exim crashed. Now it writes a panic log message
1517 and treats the condition as false.
1519 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1521 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1522 extension "envelope-auth". The code is finished and in agreement with
1523 other implementations, but there is no documentation so far and in fact,
1524 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1525 not changing the active code.
1527 Print executed "if" and "elsif" statements when debugging is used. This
1528 helps a great deal to understand what a filter does.
1530 Document more things not specified clearly in RFC3028. I had all this
1531 sorted out, when out of a sudden new issues came to my mind. Oops."
1533 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1536 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1537 canonical form (as documented). However, after a host name lookup from
1538 the IP address, check_host() was doing a simple string comparison with
1539 addresses acquired from the DNS when checking that the found name did
1540 have the original IP as one of its addresses. Since any found IPv6
1541 addresses are likely to be in abbreviated form, the comparison could
1542 fail. Luckily, there already exists a function for doing the comparison
1543 by converting both addresses to binary, so now that is used instead of
1544 the text comparison.
1546 PH/20 There was another similar case to PH/19, when a complete host name was
1547 given in a host list; looking up its IP address could give an abbreviated
1548 form, whereas the current host's name might or might not be abbreviated.
1549 The same fix has been applied.
1555 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1556 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1559 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1560 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1562 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1564 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1565 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1566 Also, for "Auto-Submitted", treat anything other than "no" as
1567 non-personal, in accordance with RFC 3834. (Previously it treated
1568 anything starting "auto-" as non-personal.)
1570 TF/01 The control=submission/name=... option had a problem with syntax
1571 errors if the name included a slash character. The /name= option
1572 now slurps the rest of the string, so it can include any characters
1573 but it must come last in the list of options (after /sender_retain
1576 PH/05 Some modifications to the interface to the fake nameserver for the new
1584 TK/01 Added the "success_on_redirect" address verification option. See
1585 NewStuff for rationale and an example.
1587 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1589 PH/02 Patch to exigrep to allow it to work on syslog lines.
1591 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1592 fread() to skip over the body file's header line, because in Cygwin the
1593 header line is locked and is inaccessible.
1595 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1596 co-exist for some time) to make it clear that it is the Exim ID that is
1597 referenced, not the Message-ID: header line.
1599 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1600 string_format() function, because snprintf() does not exist on all
1603 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1604 ${stat: expansion item.
1606 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1607 protocol synchronization error", to keep the pedants happy.
1609 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1610 well as for IRIX systems, when gcc is being used. See the host.c source
1613 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1615 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1618 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1619 transport and to the smtp transport in LMTP mode.
1621 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1623 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1624 run a filter in a subprocess. This could lead to confusion in subsequent
1625 lookups in the parent process. There should also be a search_tidyup() at
1626 the end of the subprocess.
1628 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1629 only if the host matched helo_try_verify_hosts, which caused the
1630 verification to occur when the EHLO/HELO command was issued. The ACL just
1631 tested the remembered result. Now, if a previous verification attempt has
1632 not happened, "verify = helo" does it there and then.
1634 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1636 TK/03 Fix log output including CR from clamd.
1638 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1639 exist provoked a memory error which could cause a segfault.
1641 PH/15 Installed PCRE 6.2
1643 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1645 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1646 of the problem. Specifically, suggested +O2 rather than +O1 for the
1649 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1651 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1652 record, the port was not correctly logged when the outgoing_port log
1653 selector was set (it logged the transort's default port).
1655 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1656 fallback_hosts, and "hosts" in the smtp transport.
1658 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1659 host errors such as "Connection refused".
1661 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1662 authentication with radiusclient 0.4.9:
1664 - Error returned from rc_read_config was caught wrongly
1665 - Username/password not passed on to radius server due to wrong length.
1667 The presumption is that some radiusclient API changes for 4.51/PH/17
1668 were not taken care of correctly. The code is still untested by me (my
1669 Linux distribution still has 0.3.2 of radiusclient), but it was
1670 contributed by a Radius user.
1672 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1673 expanding the "port" option of the smtp transport.
1675 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1676 while reading a MIME header. Thanks to Tom Hughes for a patch.
1678 PH/24 Include config.h inside local_scan.h so that configuration settings are
1681 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1682 that in an ACL for RCPT (for example), you can examine exactly what was
1685 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1686 commands, but it was not correctly comparing the address with the actual
1687 client host address. Thus, it would show the EHLO address in Received:
1688 header lines when this was not necessary.
1690 PH/27 Added the % operator to ${eval:}.
1692 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1693 it should be ignoring failures (because with -C, for example, it has lost
1694 privilege). It wasn't ignoring creation failures other than "already
1697 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1700 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1701 because an input line was too long, either on its own, or by virtue of
1702 too many continuations, the temporary file was not being removed, and the
1703 return code was incorrect.
1705 PH/31 Missing "BOOL" in function definition in filtertest.c.
1707 PH/32 Applied Sieve patches from the maintainer.
1709 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1711 PH/33 Added "verify = not_blind".
1713 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1714 Local/Makefile (with some defaults set). These are used in built scripts
1715 such as exicyclog, but they have never been used in the exim_install
1716 script (though there are many overriding facilities there). I have
1717 arranged that the exim_install script now takes note of these two
1720 PH/35 Installed configuration files for Dragonfly.
1722 PH/36 When a locally submitted message by a trusted user did not contain a
1723 From: header, and the sender address was obtained from -f or from an SMTP
1724 MAIL command, and the trusted user did not use -F to supply a sender
1725 name, $originator_name was incorrectly used when constructing a From:
1726 header. Furthermore, $originator_name was used for submission mode
1727 messages from external hosts without From: headers in a similar way,
1728 which is clearly wrong.
1730 PH/37 Added control=suppress_local_fixups.
1732 PH/38 When log_selector = +received_sender was set, and the addition of the
1733 sender made the log line's construction buffer exactly full, or one byte
1734 less than full, an overflow happened when the terminating "\n" was
1737 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1738 when the result of a list match is failure because a DNS lookup failed.
1740 PH/40 RM_COMMAND is now used in the building process.
1742 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1743 the "build-* directories that it finds.
1745 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1746 address in a domain literal was a prefix of an interface address.
1748 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1749 when verifying a sender address, unless rewrite_headers is false.
1751 PH/44 Wrote a long comment about why errors_to addresses are verified as
1752 recipients, not senders.
1754 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1755 the ratelimit ACL was added.
1757 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1759 PH/46 Added extra information about PostgreSQL errors to the error string.
1761 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1762 avoiding the need to install special zones in a real server. This is
1763 backwards compatible; if it can't find the fake resolver, it drops back.
1764 Thus, both old and new test suites can be run.
1766 TF/02 Added util/ratelimit.pl
1768 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1769 clock is set back in time.
1771 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1772 Candler <B.Candler@pobox.com>.
1774 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1775 OK for addresses that are the result of redirections.
1777 PH/48 A number of further additions for the benefit of the new test suite,
1778 including a fake gethostbyname() that interfaces to the fake DNS resolver
1781 TF/06 The fix for widen_domains has also been applied to qualify_single and
1782 search_parents which are the other dnslookup options that can cause
1785 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1788 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1790 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1791 service" if it encountered an MX record with an empty target hostname.
1792 The message is now "an MX or SRV record indicated no SMTP service".
1794 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1795 if the DNS of the sending site is misconfigured. This is quite a
1796 common situation. This change restores the behaviour of treating a
1797 helo verification defer as a failure.
1799 PH/51 If self=fail was set on a router, the bounce message did not include the
1800 actual error message.
1806 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1808 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1809 command itself ended in error, the underlying message about the transport
1810 filter timeout was being overwritten with the pipe command error. Now the
1811 underlying error message should be appended to the second error message.
1813 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1815 PH/02 Reduce the amount of output that "make" produces by default. Full output
1816 can still be requested.
1818 PH/03 The warning log line about a condition test deferring for a "warn" verb
1819 was being output only once per connection, rather than after each
1820 occurrence (because it was using the same function as for successful
1821 "warn" verbs). This seems wrong, so I have changed it.
1823 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1824 it should not have, which might have caused a crash in the right
1825 circumstances, but probably never did.
1827 PH/04 Installed a modified version of Tony Finch's patch to make submission
1828 mode fix the return path as well as the Sender: header line, and to
1829 add a /name= option so that you can make the user's friendly name appear
1832 TF/03 Added the control = fakedefer ACL modifier.
1834 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1835 Mark Lowes for thorough testing.
1837 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1839 TK/03 Merged latest SRS patch from Miles Wilton.
1841 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1842 with the definition in sysexits.h (which is #included earlier).
1843 Fortunately, Exim does not actually use EX_OK. The code used to try to
1844 preserve the sysexits.h value, by assumimg that macro definitions were
1845 scanned for macro replacements. I have been disabused of this notion,
1846 so now the code just undefines EX_OK before #including unistd.h.
1848 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1849 in the smtp transport. When a block could not be written in a single
1850 write() function, the timeout was being re-applied to each part-write.
1851 This seems wrong - if the receiver was accepting one byte at a time it
1852 would take for ever. The timeout is now adjusted when this happens. It
1853 doesn't have to be particularly precise.
1855 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1856 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1858 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1859 without a domain if the check to <postmaster@domain> fails.
1861 SC/01 Eximstats: added -xls and the ability to specify output files
1862 (patch written by Frank Heydlauf).
1864 SC/02 Eximstats: use FileHandles for outputing results.
1866 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1868 SC/04 Eximstats: fixed display of large numbers with -nvr option
1870 SC/05 Eximstats: fixed merging of reports with empty tables.
1872 SC/06 Eximstats: added the -include_original_destination flag
1874 SC/07 Eximstats: removed tabs and trailing whitespace.
1876 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1878 TK/06 MBOX spool code: Add real "From " MBOX separator line
1879 so the .eml file is really in mbox format (even though
1880 most programs do not really care). Patch from Alex Miller.
1882 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1883 The latter is generated from $received_to and is only set if the
1884 message has one envelope recipient. SA can use these headers,
1885 obviously out-of-the-box. Patch from Alex Miller.
1887 PH/08 The ${def test on a variable was returning false if the variable's
1888 value was "0", contrary to what the specification has always said!
1889 The result should be true unless the variable is empty.
1891 PH/09 The syntax error of a character other than { following "${if
1892 def:variable_name" (after optional whitespace) was not being diagnosed.
1893 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1894 accidental colon was present, for example, could give incorrect results.
1896 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1897 result is used (not including appendfile, where other changes are about
1900 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1901 This involved changing a lot of size variables from int to off_t. It
1902 should work with maildirs and everything.
1904 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1905 spamd dying while we are connected to it.
1907 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1910 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1911 function for BDB 4.3.
1913 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1915 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1916 the sources for PCRE have been re-arranged and more files are now
1919 PH/15 The code I had for printing potentially long long variables in PH/11
1920 above was not the best (it lost precision). The length of off_t variables
1921 is now inspected at build time, and an appropriate printing format (%ld
1922 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1923 to be "long long int" or "long int". This is needed for the internal
1924 formatting function string_vformat().
1926 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1927 the configuration file to be ":syslog", then the script "guesses" where
1928 the logs files are, rather than using the compiled in default. In our
1929 case the guess is not the same as the compiled default, so the script
1930 suddenly stopped working when I started to use syslog. The patch checks
1931 to see if log_file_path is "". If so, it attempts to read it from exim
1932 with no configuration file to get the compiled in version, before it
1933 falls back to the previous guessing code."
1935 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1936 implementing BATV in an Exim configuration. See NewStuff for the gory
1939 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1940 Makefile that are specific to HP-UX.
1942 PH/18 If the "use_postmaster" option was set for a recipient callout together
1943 with the "random" option, the postmaster address was used as the MAIL
1944 FROM address for the random test, but not for the subsequent recipient
1945 test. It is now used for both.
1947 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1948 patch removes a few documentation additions to RFC 3028, because the
1949 latest draft now contains them. It adds the new en;ascii-case comparator
1950 and a new error check for 8bit text in MIME parts. Comparator and
1951 require names are now matched exactly. I enabled the subaddress
1952 extension, but it is not well tested yet (read: it works for me)."
1954 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1955 rework some of the code of TK/09 above to avoid the hardwired use of
1956 "%lld" and "long long". Replaced the call to snprintf() with a call to
1959 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1960 records point to non-existent hosts", "retry timeout exceeded", and
1961 "retry time not reached for any host after a long failure period".
1963 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1964 experimental DomainKeys support:
1966 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1967 (2) On an error such as an illegally used "control", the wrong name for
1968 the control was given.
1970 These problems did NOT occur unless DomainKeys support was compiled.
1972 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1974 PH/24 Added ${if match_ip condition.
1976 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1977 databases so that it will be absolutely obvious if a crash occurs in the
1978 DB library. This is a regular occurrence (often caused by mis-matched
1981 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1982 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1983 picked up on a user's system that detects such things. There doesn't seem
1984 to be a gcc warning option for this - only an attribute that has to be
1985 put on the function's prototype. It seems that in Fedora Core 4 they have
1986 set this on a number of new functions. No doubt there will be more in due
1989 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1990 specify a transport. However, if an address that was verified by such a
1991 router was the subject of a callout, Exim crashed because it tried to
1992 read the rcpt_include_affixes from the non-existent transport. Now it
1993 just assumes that the setting of that option is false. This bug was
1994 introduced by 4.51/PH/31.
1996 PH/28 Changed -d+all to exclude +memory, because that information is very
1997 rarely of interest, but it makes the output a lot bigger. People tend to
1998 do -d+all out of habit.
2000 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2001 code in os-type was giving problems when libc.so lives in lib64, like on
2004 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2005 aren't the modern standard, and it seems that some systems' include files
2006 don't always have them. Exim was already checking for some of the newer
2007 ones like T_AAAA, and defining it itself. I've added checks for all the
2008 record types that Exim uses.
2010 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2011 not automatically generating a new one, as it is supposed to. This
2012 prevented TLS from working. If the file did exist, but contained invalid
2013 data, a new version was generated, as expected. It was only the case of a
2014 non-existent file that was broken.
2016 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2017 with a change in libdomainkeys > 0.64.
2019 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2020 from DNS. If the selector record carries the flag, it now has
2021 precedence over the domain-wide flag.
2023 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2025 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2026 the use of a port name that isn't defined in /etc/services) occurred, the
2027 message was deferred as in a normal delivery, and thus remained on the
2028 spool, instead of being failed because of the mua_wrapper setting. This
2029 is now fixed, and I tidied up some of the mua_wrapper messages at the
2032 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2033 the messages in a hash of arrays rather than using individual hashes.
2034 This is a bit cleaner and results in dramatic memory savings, albeit
2035 at a slight CPU cost.
2037 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2038 as requested by Marc Sherman.
2040 SC/10 Eximstats: added histograms for user specified patterns as requested
2043 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2045 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2046 fopen() in the content-scanning modules that did not already have it.
2052 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2053 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2055 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2057 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2059 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2060 to test Sieve filters that use "vacation".
2062 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2063 that changes the way the GnuTLS parameters are stored in the cache file.
2064 The new format can be generated externally. For backward compatibility,
2065 if the data in the cache doesn't make sense, Exim assumes it has read an
2066 old-format file, and it generates new data and writes a new file. This
2067 means that you can't go back to an older release without removing the
2070 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2071 work if there are any delivery delays because "one_time" forces the
2072 parent to be marked "delivered", so its unseen clone is never tried
2073 again. For this reason, Exim now forbids the simultaneous setting of
2076 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2077 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2078 entry if you want to know the details. The fix, however, neglected to
2079 consider the case when local delivery batching is involved. The test for
2080 "previously delivered" was not happening when checking to see if an
2081 address could be batched with a previous (undelivered) one; under
2082 certain circumstances this could lead to multiple deliveries to the same
2085 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2086 in its include files, and this causes problems building Exim.
2088 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2089 header_syntax) but Exim was just ignoring anything given after a slash.
2090 In particular, this caused confusion with an attempt to use "verify =
2091 reverse_host_lookup/defer_ok". An error is now given when options are
2092 supplied for verify items that do not have them. (Maybe reverse_host_
2093 lookup should have a defer_ok option, but that's a different point.)
2095 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2096 defined by RFC 821) to 2048, because there were problems with some AUTH
2097 commands, and RFC 1869 says the size should be increased for extended
2098 SMTP commands that take arguments.
2100 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2103 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2104 "unknown" error; now it says that the functionality isn't in the binary.
2106 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2107 an address' error message when a string expansion fails (syntax or
2108 whatever). Otherwise the password may appear in the log. Following change
2109 PH/42 below, there is no longer a chance of it appearing in a bounce
2112 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2114 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2115 of its addresses was ever tried. (Bugzilla bug #2.)
2117 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2118 the result incorrectly in the debug output. (It correctly added a newline
2119 to what was transported.)
2121 TF/01 Added $received_time.
2123 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2124 commented out examples of how to interface to a virus scanner and to
2125 SpamAssassin. Also added commented examples of av_scanner and
2126 spamd_address settings.
2128 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2129 and controls are allowed in which ACLs. There were a couple of minor
2130 errors. Some of the entries in the conditions table (which is a table of
2131 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2132 negation of where the condition IS allowed.
2134 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2136 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2137 header file does not have a version number, so I've had to invent a new
2138 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2139 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2140 radiusclient), but it was contributed by a Radius user.
2142 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2143 files or directories, for OpenSSL.
2145 PH/19 When an Exim process that is running as root has to create an Exim log
2146 file, it does so in a subprocess that runs as exim:exim so as to get the
2147 ownership right at creation (otherwise, other Exim processes might see
2148 the file with the wrong ownership). There was no test for failure of this
2149 fork() call, which would lead to the process getting stuck as it waited
2150 for a non-existent subprocess. Forks do occasionally fail when resources
2151 run out. I reviewed all the other calls to fork(); they all seem to check
2154 PH/20 When checking for unexpected SMTP input at connect time (before writing
2155 the banner), Exim was not dealing correctly with a non-positive return
2156 from the read() function. If the client had disconnected by this time,
2157 the result was a log entry for a synchronization error with an empty
2158 string after "input=" when read() returned zero. If read() returned -1
2159 (an event I could not check), uninitialized data bytes were printed.
2160 There were reports of junk text (parts of files, etc) appearing after
2163 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2165 PH/22 Added support for macro redefinition, and (re)definition in between
2166 driver and ACL definitions.
2168 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2169 forgetting to use the resulting value; it was using the unexpanded value.
2171 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2172 hadn't been configured. The fix is from Juergen Kreileder, who
2173 understands it better than I do:
2175 "Here's what I see happening with three configured cyrus_sasl
2176 authenticators configured (plain, login, cram-md5):
2178 On startup auth_cyrus_sasl_init() gets called for each of these.
2179 This means three calls to sasl_listmech() without a specified mech_list.
2180 => SASL tests which mechs of all available mechs actually work
2181 => three warnings about OTP not working
2182 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2184 With the patch, sasl_listmech() also gets called three times. But now
2185 SASL's mech_list option is set to the server_mech specified in the the
2186 authenticator. Or in other words, the answer from sasl_listmech()
2187 gets limited to just the mech you're testing for (which is different
2189 => the return list contains just 'plain' or 'login', 'cram-md5' or
2190 nothing depending on the value of ob->server_mech.
2192 I've just tested the patch: Authentication still works fine,
2193 unavailable mechs specified in the exim configuration are still
2194 caught, and the auth.log warnings about OTP are gone."
2196 PH/25 When debugging is enabled, the contents of the command line are added
2197 to the debugging output, even when log_selector=+arguments is not
2200 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2201 answer is "GNU", and only if the return is "GNU/something" is the answer
2204 PH/27 $acl_verify_message is now set immediately after the failure of a
2205 verification in an ACL, and so is available in subsequent modifiers. In
2206 particular, the message can be preserved by coding like this:
2208 warn !verify = sender
2209 set acl_m0 = $acl_verify_message
2211 Previously, $acl_verify_message was set only while expanding "message"
2212 and "log_message" when a very denied access.
2214 PH/28 Modified OS/os.c-Linux with
2216 -#ifndef OS_LOAD_AVERAGE
2217 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2219 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2220 nomenclature these days.)
2222 PH/29 Installed patch from the Sieve maintainer that adds the options
2223 sieve_useraddress and sieve_subaddress to the redirect router.
2225 PH/30 In these circumstances:
2226 . Two addresses routed to the same list of hosts;
2227 . First host does not offer TLS;
2228 . First host accepts first address;
2229 . First host gives temporary error to second address;
2230 . Second host offers TLS and a TLS session is established;
2231 . Second host accepts second address.
2232 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2233 and peerdn, if requested) that were in fact used only for the second
2236 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2237 attention to any local part prefix or suffix that was matched by the
2238 router that accepted the address. It now behaves in the same way as it
2239 does for delivery: the affixes are removed from the local part unless
2240 rcpt_include_affixes is set on the transport.
2242 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2243 timeout during the DATA phase of an incoming message.
2245 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2246 applied a patch sanctioned by the Sieve maintainer.
2248 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2249 the uid or gid is negative. A case of a negative gid caused this to be
2250 noticed. The fix allows for either to be negative.
2252 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2253 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2256 PH/36 The change PH/12 above was broken. Fixed it.
2258 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2259 the grounds that routing the same address twice would always produce the
2260 same answer. This might have been true once, but it is certainly no
2261 longer true now. Routing a child address may depend on the previous
2262 routing that produced that child. Some complicated redirection strategies
2263 went wrong when messages had multiple recipients, and made Exim's
2264 behaviour dependent on the order in which the addresses were given.
2266 I have moved the duplicate checking until after the routing is complete.
2267 Exim scans the addresses that are assigned to local and remote
2268 transports, and removes any duplicates. This means that more work will be
2269 done, as duplicates will always all be routed, but duplicates are
2270 presumably rare, so I don't expect this is of any significance.
2272 For deliveries to pipes, files, and autoreplies, the duplicate checking
2273 still happens during the routing process, since they are not going to be
2276 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2277 It corrects a timeout issue with spamd. This is Ian's comment: "The
2278 background is that sometimes spamd either never reads data from a
2279 connection it has accepted, or it never writes response data. The exiscan
2280 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2281 blindly assumes that writes won't block so it may never time out."
2283 PH/39 Allow G after quota size as well as K and M.
2285 PH/40 The value set for $authenticated_id in an authenticator may not contain
2286 binary zeroes or newlines because the value is written to log lines and
2287 to spool files. There was no check on this. Now the value is run through
2288 the string_printing() function so that such characters are converted to
2289 printable escape sequences.
2291 PH/41 $message_linecount is a new variable that contains the total number of
2292 lines in the message. Compare $body_linecount, which is the count for the
2295 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2296 bounce and delay warning messages, except in certain special cases, which
2299 (a) An SMTP error message from a remote host;
2300 (b) A message specified in a :fail: redirection;
2301 (c) A message specified in a "fail" command in a system filter;
2302 (d) A message specified in a FAIL return from the queryprogram router;
2303 (e) A message specified by the cannot_route_message router option.
2305 In these cases only, Exim does include the error details in bounce and
2306 warning messages. There are also a few cases where bland messages such
2307 as "unrouteable address" or "local delivery error" are given.
2309 PH/43 $value is now also set for the "else" part of a ${run expansion.
2311 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2312 being worked on, but at least Exim now implements the latest version to
2315 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2316 process to complete was treated as a delivery failure, a timeout while
2317 writing the message to the pipe was logged, but erroneously treated as a
2318 successful delivery. Such timeouts include transport filter timeouts. For
2319 consistency with the overall process timeout, these timeouts are now
2320 treated as errors, giving rise to delivery failures by default. However,
2321 there is now a new Boolean option for the pipe transport called
2322 timeout_defer, which, if set TRUE, converts the failures into defers for
2323 both kinds of timeout. A transport filter timeout is now identified in
2326 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2327 systems where "make" and "gmake" are different, calling "gmake" at top
2328 level broke things. I've arranged for the value of $(MAKE) to be passed
2329 from the Makefile to this script so that it can call the same version of
2333 A note about Exim versions 4.44 and 4.50
2334 ----------------------------------------
2336 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2337 changes of various kinds. As a consequence, a big documentation update was
2338 needed. This delayed the release for rather longer than seemed good, especially
2339 in the light of a couple of (minor) security issues. Therefore, the changes
2340 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2341 release. So 4.44 and 4.50 are in effect two different branches that both start
2344 I have left the 4.50 change log unchanged; it contains all the changes since
2345 4.43. The change log for 4.44 is below; many of its items are identical to
2346 those for 4.50. This seems to be the most sensible way to preserve the
2347 historical information.
2353 1. Minor wording change to the doc/README.SIEVE file.
2355 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2356 computation of the current number of files was incorrect.
2358 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2359 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2362 4. Give more explanation in the error message when the command for a transport
2363 filter fails to execute.
2365 5. There are several places where Exim runs a non-Exim command in a
2366 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2367 was being done only for the command run by the queryprogram router. It is
2368 now done for all such subprocesses. The other cases are: ${run, transport
2369 filters, and the commands run by the lmtp and pipe transports.
2371 6. Added CONFIGURE_GROUP build-time option.
2373 7. Some older OS have a limit of 256 on the maximum number of file
2374 descriptors. Exim was using setrlimit() to set 1000 as a large value
2375 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2376 systems. I've change it so that if it can't get 1000, it tries for 256.
2378 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2379 was an oversight, and furthermore, ever since the addition of extra
2380 controls (e.g. 4.43/32), the checks on when to allow different forms of
2381 "control" were broken. There should now be diagnostics for all cases when a
2382 control that does not make sense is encountered.
2384 9. Added the /retain_sender option to "control=submission".
2386 10. $recipients is now available in the predata ACL (oversight).
2388 11. Tidy the search cache before the fork to do a delivery from a message
2389 received from the command line. Otherwise the child will trigger a lookup
2390 failure and thereby defer the delivery if it tries to use (for example) a
2391 cached ldap connection that the parent has called unbind on.
2393 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2394 of $address_data from the recipient verification was clobbered by the
2395 sender verification.
2397 13. The value of address_data from a sender verification is now available in
2398 $sender_address_data in subsequent conditions in the ACL statement.
2400 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2402 15. Added a new option "connect=<time>" to callout options, to set a different
2405 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2406 was its contents. (It was OK if the option was not defined at all.)
2408 17. A "Completed" log line is now written for messages that are removed from
2409 the spool by the -Mrm option.
2411 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2412 information about exactly what failed.
2414 19. Added -dd to debug only the daemon process.
2416 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2417 handles timeouts, both on the server side and network timeouts. Renamed the
2418 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2420 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2421 It is now set to "smtps".
2423 22. $host_address is now set to the target address during the checking of
2424 ignore_target_hosts.
2426 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2427 being passed; this would have caused $sender_host_name to have been used if
2428 matching the list had actually called for a host name (not very likely,
2429 since this list is usually IP addresses). A host name is now passed as
2432 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2433 code that specifies a non-restarting handler (typically sigaction() in
2434 modern systems) in an attempt to fix a rare and obscure crash bug.
2436 25. Narrowed the window for a race in the daemon that could cause it to ignore
2437 SIGCHLD signals. This is not a major problem, because they are used only to
2438 wake it up if nothing else does.
2440 26. A malformed maildirsize file could cause Exim to calculate negative values
2441 for the mailbox size or file count. Odd effects could occur as a result.
2442 The maildirsize information is now recalculated if the size or filecount
2445 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2446 support for a long time. Removed HAVE_SYS_VFS_H.
2448 28. Installed the latest version of exipick from John Jetmore.
2450 29. In an address list, if the pattern was not a regular expression, an empty
2451 subject address (from a bounce message) matched only if the pattern was an
2452 empty string. Non-empty patterns were not even tested. This was the wrong
2453 because it is perfectly reasonable to use an empty address as part of a
2454 database query. An empty address is now tested by patterns that are
2455 lookups. However, all the other forms of pattern expect the subject to
2456 contain a local part and a domain, and therefore, for them, an empty
2457 address still always fails if the pattern is not itself empty.
2459 30. Exim went into a mad DNS loop when attempting to do a callout where the
2460 host was specified on an smtp transport, and looking it up yielded more
2461 than one IP address.
2463 31. Re-factored the code for checking spool and log partition space into a
2464 function that finds that data and another that does the check. The former
2465 is then used to implement four new variables: $spool_space, $log_space,
2466 $spool_inodes, and $log_inodes.
2468 32. The RFC2047 encoding function was originally intended for short strings
2469 such as real names; it was not keeping to the 75-character limit for
2470 encoded words that the RFC imposes. It now respects the limit, and
2471 generates multiple encoded words if necessary. To be on the safe side, I
2472 have increased the buffer size for the ${rfc2047: expansion operator from
2475 33. It is now permitted to omit both strings after an "if" condition; if the
2476 condition is true, the result is "true". As before, when the second string
2477 is omitted, a false condition yields an empty string. This makes it less
2478 cumbersome to write custom ACL and router conditions.
2480 34. Failure to deliver a bounce message always caused it to be frozen, even if
2481 there was an errors_to setting on the router. The errors_to setting is now
2484 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2485 canonical form (fully expanded) before being placed in
2486 $sender_host_address.
2488 36. The table in the code that translates DNS record types into text (T_A to
2489 "A" for instance) was missing entries for NS and CNAME. It is just possible
2490 that this could have caused confusion if both these types were looked up
2491 for the same domain, because the text type is used as part of Exim's
2492 per-process caching. But the chance of anyone hitting this buglet seems
2495 37. The dnsdb lookup has been extended in a number of ways.
2497 (1) There is a new type, "zns", which walks up the domain tree until it
2498 finds some nameserver records. It should be used with care.
2500 (2) There is a new type, "mxh", which is like "mx" except that it returns
2501 just the host names, not the priorities.
2503 (3) It is now possible to give a list of domains (or IP addresses) to be
2504 looked up. The behaviour when one of the lookups defers can be
2505 controlled by a keyword.
2507 (4) It is now possible to specify the separator character for use when
2508 multiple records are returned.
2510 38. The dnslists ACL condition has been extended: it is now possible to supply
2511 a list of IP addresses and/or domains to be looked up in a particular DNS
2514 39. Added log_selector=+queue_time_overall.
2516 40. When running the queue in the test harness, wait just a tad after forking a
2517 delivery process, to get repeatability of debugging output.
2519 41. Include certificate and key file names in error message when GnuTLS fails
2520 to set them up, because the GnuTLS error message doesn't include the name
2521 of the failing file when there is a problem reading it.
2523 42. Allow both -bf and -bF in the same test run.
2525 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2527 44. The "Exiscan patch" is now merged into the mainline Exim source.
2529 45. Sometimes the final signoff response after QUIT could fail to get
2530 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2531 < 0 before doing a fflush(). This bug looks as though it goes back to the
2532 introduction of TLS in release 3.20, but "sometimes" must have been rare
2533 because the tests only now provoked it.
2535 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2536 (this can affect the format of dates).
2538 47. exim_tidydb, when checking for the continued existence of a message for
2539 which it has found a message-specific retry record, was not finding
2540 messages that were in split spool directories. Consequently, it was
2541 deleting retry records that should have stayed in existence.
2543 48. Steve fixed some bugs in eximstats.
2545 49. The SPA authentication driver was not abandoning authentication and moving
2546 on to the next authenticator when an expansion was forced to fail,
2547 contradicting the general specification for all authenticators. Instead it
2548 was generating a temporary error. It now behaves as specified.
2550 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2551 (the order specifies the preference for clients). The order is now AES256,
2552 AES128, 3DES, ARCFOUR128.
2554 51. Small patch to Sieve code - explicitly set From: when generating an
2557 52. Exim crashed if a remote delivery caused a very long error message to be
2558 recorded - for instance if somebody sent an entire SpamAssassin report back
2559 as a large number of 550 error lines. This bug was coincidentally fixed by
2560 increasing the size of one of Exim's internal buffers (big_buffer) that
2561 happened as part of the Exiscan merge. However, to be on the safe side, I
2562 have made the code more robust (and fixed the comments that describe what
2565 53. Now that there can be additional text after "Completed" in log lines (if
2566 the queue_time_overall log selector is set), a one-byte patch to exigrep
2567 was needed to allow it to recognize "Completed" as not the last thing in
2570 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2571 patch that reportedly fixes this has been added. I am not expert enough to
2572 create a test for it. This is what the patch creator wrote:
2574 "I found a little strange behaviour of ldap code when working with
2575 Windows 2003 AD Domain, where users was placed in more than one
2576 Organization Units. When I tried to give exim partial DN, the exit code
2577 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2578 But simultaneously result of request was absolutely normal ldap result,
2579 so I produce this patch..."
2581 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2582 REFERENCE, so I have modified the code to exclude the patch when that macro
2585 55. Some experimental protocols are using DNS PTR records for new purposes. The
2586 keys for these records are domain names, not reversed IP addresses. The
2587 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2588 leaves it alone. Component reversal etc. now happens only for IP addresses.
2591 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2593 57. Double the size of the debug message buffer (to 2048) so that more of very
2594 long debug lines gets shown.
2596 58. The exicyclog utility now does better if the number of log files to keep
2597 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2599 59. Two changes related to the smtp_active_hostname option:
2601 (1) $smtp_active_hostname is now available as a variable.
2602 (2) The default for smtp_banner uses $smtp_active_hostname instead
2603 of $primary_hostname.
2605 60. The host_aton() function is supposed to be passed a string that is known
2606 to be a valid IP address. However, in the case of IPv6 addresses, it was
2607 not checking this. This is a hostage to fortune. Exim now panics and dies
2608 if the condition is not met. A case was found where this could be provoked
2609 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2610 components; fortuitously, this particular loophole had already been fixed
2611 by change 4.50/55 above.
2613 If there are any other similar loopholes, the new check in host_aton()
2614 itself should stop them being exploited. The report I received stated that
2615 data on the command line could provoke the exploit when Exim was running as
2616 exim, but did not say which command line option was involved. All I could
2617 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2618 running as the user.
2621 61. There was a buffer overflow vulnerability in the SPA authentication code
2622 (which came originally from the Samba project). I have added a test to the
2623 spa_base64_to_bits() function which I hope fixes it.
2626 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2627 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2629 63. The daemon start-up calls getloadavg() while still root for those OS that
2630 need the first call to be done as root, but it missed one case: when
2631 deliver_queue_load_max is set with deliver_drop_privilege. This is
2632 necessary for the benefit of the queue runner, because there is no re-exec
2633 when deliver_drop_privilege is set.
2635 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2636 This has been fixed.
2638 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2639 was in use, was not putting the data itself into the right store pool;
2640 consequently, it could be overwritten for a subsequent message in the same
2641 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2644 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2646 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2647 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2648 the function were treating the return as a boolean value, which happened to
2649 work because 0=false and not-0=true, but is not correct code.
2651 68. The host_aton() function was not handling scoped IPv6 addresses (those
2652 with, for example, "%eth0" on the end) correctly.
2654 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2655 negated items (that is, ~something) in unsigned ints. Some compilers
2656 apparently mutter when there is no cast.
2658 70. If an address verification called from an ACL failed, and did not produce a
2659 user-specific message (i.e. there was only a "system" message), nothing was
2660 put in $acl_verify_message. In this situation, it now puts the system
2663 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2664 session; change 4.31/43 added the unwanted input to the log line - except
2665 that it did not do this in the start of session case. It now does.
2667 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2668 This is wrong and can cause the other end to generate a synchronization
2669 error if it is another Exim or anything else that does the synchronization
2670 check. A QUIT command is no longer sent after a timeout.
2672 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2673 during host lookups.
2675 74. The defer_ok option of callout verification was not working if it was used
2676 when verifying addresses in header lines, that is, for this case:
2678 verify = header_sender/callout=defer_ok
2680 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2681 those file descriptors could be used for SMTP connections. If anything
2682 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2683 could be sent to the SMTP client, causing chaos. The daemon now opens
2684 stdin, stdout, and stderr to /dev/null when it puts itself into the
2687 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2688 log by default. The user can override this with suitable Perl magic.
2690 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2691 the log message when discard triggers, was not working for the DATA ACL or
2692 for the non-SMTP ACL.
2694 78. Error message wording change in sieve.c.
2696 79. If smtp_accept_max_per_host was set, the number of connections could be
2697 restricted to fewer than expected, because the daemon was trying to set up
2698 a new connection before checking whether the processes handling previous
2699 connections had finished. The check for completed processes is now done
2700 earlier. On busy systems, this bug wouldn't be noticed because something
2701 else would have woken the daemon, and it would have reaped the completed
2704 80. If a message was submitted locally by a user whose login name contained one
2705 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2706 It caused a spool format error. I have fixed the spool reading code. A
2707 related problem was that the "from" clause in the Received: line became
2708 illegal because of the space(s). It is now covered by ${quote_local_part.
2710 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2713 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2715 83. Patch from Sieve maintainer to fix unterminated string problem in
2716 "vacation" handling.
2718 84. Some minor changes to the Linux configuration files to help with other
2719 OS variants using glibc.
2721 85. One more patch for Sieve to update vacation handling to latest spec.
2724 ----------------------------------------------------
2725 See the note above about the 4.44 and 4.50 releases.
2726 ----------------------------------------------------
2732 1. Change 4.43/35 introduced a bug that caused file counts to be
2733 incorrectly computed when quota_filecount was set in an appendfile
2736 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2737 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2740 3. Give more explanation in the error message when the command for a transport
2741 filter fails to execute.
2743 4. There are several places where Exim runs a non-Exim command in a
2744 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2745 was being done only for the command run by the queryprogram router. It is
2746 now done for all such subprocesses. The other cases are: ${run, transport
2747 filters, and the commands run by the lmtp and pipe transports.
2749 5. Some older OS have a limit of 256 on the maximum number of file
2750 descriptors. Exim was using setrlimit() to set 1000 as a large value
2751 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2752 systems. I've change it so that if it can't get 1000, it tries for 256.
2754 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2755 was an oversight, and furthermore, ever since the addition of extra
2756 controls (e.g. 4.43/32), the checks on when to allow different forms of
2757 "control" were broken. There should now be diagnostics for all cases when a
2758 control that does not make sense is encountered.
2760 7. $recipients is now available in the predata ACL (oversight).
2762 8. Tidy the search cache before the fork to do a delivery from a message
2763 received from the command line. Otherwise the child will trigger a lookup
2764 failure and thereby defer the delivery if it tries to use (for example) a
2765 cached ldap connection that the parent has called unbind on.
2767 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2768 of $address_data from the recipient verification was clobbered by the
2769 sender verification.
2771 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2772 was its contents. (It was OK if the option was not defined at all.)
2774 11. A "Completed" log line is now written for messages that are removed from
2775 the spool by the -Mrm option.
2777 12. $host_address is now set to the target address during the checking of
2778 ignore_target_hosts.
2780 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2781 being passed; this would have caused $sender_host_name to have been used if
2782 matching the list had actually called for a host name (not very likely,
2783 since this list is usually IP addresses). A host name is now passed as
2786 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2787 code that specifies a non-restarting handler (typically sigaction() in
2788 modern systems) in an attempt to fix a rare and obscure crash bug.
2790 15. Narrowed the window for a race in the daemon that could cause it to ignore
2791 SIGCHLD signals. This is not a major problem, because they are used only to
2792 wake it up if nothing else does.
2794 16. A malformed maildirsize file could cause Exim to calculate negative values
2795 for the mailbox size or file count. Odd effects could occur as a result.
2796 The maildirsize information is now recalculated if the size or filecount
2799 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2800 support for a long time. Removed HAVE_SYS_VFS_H.
2802 18. Updated exipick to current release from John Jetmore.
2804 19. Allow an empty sender to be matched against a lookup in an address list.
2805 Previously the only cases considered were a regular expression, or an
2808 20. Exim went into a mad DNS lookup loop when doing a callout where the
2809 host was specified on the transport, if the DNS lookup yielded more than
2812 21. The RFC2047 encoding function was originally intended for short strings
2813 such as real names; it was not keeping to the 75-character limit for
2814 encoded words that the RFC imposes. It now respects the limit, and
2815 generates multiple encoded words if necessary. To be on the safe side, I
2816 have increased the buffer size for the ${rfc2047: expansion operator from
2819 22. Failure to deliver a bounce message always caused it to be frozen, even if
2820 there was an errors_to setting on the router. The errors_to setting is now
2823 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2824 canonical form (fully expanded) before being placed in
2825 $sender_host_address.
2827 24. Updated eximstats to version 1.33
2829 25. Include certificate and key file names in error message when GnuTLS fails
2830 to set them up, because the GnuTLS error message doesn't include the name
2831 of the failing file when there is a problem reading it.
2833 26. Expand error message when OpenSSL has problems setting up cert/key files.
2836 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2837 (this can affect the format of dates).
2839 28. exim_tidydb, when checking for the continued existence of a message for
2840 which it has found a message-specific retry record, was not finding
2841 messages that were in split spool directories. Consequently, it was
2842 deleting retry records that should have stayed in existence.
2844 29. eximstats updated to version 1.35
2845 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2846 1.35 - bugfix such that pie charts by volume are generated correctly
2848 30. The SPA authentication driver was not abandoning authentication and moving
2849 on to the next authenticator when an expansion was forced to fail,
2850 contradicting the general specification for all authenticators. Instead it
2851 was generating a temporary error. It now behaves as specified.
2853 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2854 (the order specifies the preference for clients). The order is now AES256,
2855 AES128, 3DES, ARCFOUR128.
2857 31. Small patch to Sieve code - explicitly set From: when generating an
2860 32. Exim crashed if a remote delivery caused a very long error message to be
2861 recorded - for instance if somebody sent an entire SpamAssassin report back
2862 as a large number of 550 error lines. This bug was coincidentally fixed by
2863 increasing the size of one of Exim's internal buffers (big_buffer) that
2864 happened as part of the Exiscan merge. However, to be on the safe side, I
2865 have made the code more robust (and fixed the comments that describe what
2868 33. Some experimental protocols are using DNS PTR records for new purposes. The
2869 keys for these records are domain names, not reversed IP addresses. The
2870 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2871 leaves it alone. Component reversal etc. now happens only for IP addresses.
2874 34. The host_aton() function is supposed to be passed a string that is known
2875 to be a valid IP address. However, in the case of IPv6 addresses, it was
2876 not checking this. This is a hostage to fortune. Exim now panics and dies
2877 if the condition is not met. A case was found where this could be provoked
2878 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2879 components; fortuitously, this particular loophole had already been fixed
2880 by change 4.50/55 or 4.44/33 above.
2882 If there are any other similar loopholes, the new check in host_aton()
2883 itself should stop them being exploited. The report I received stated that
2884 data on the command line could provoke the exploit when Exim was running as
2885 exim, but did not say which command line option was involved. All I could
2886 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2887 running as the user.
2890 35. There was a buffer overflow vulnerability in the SPA authentication code
2891 (which came originally from the Samba project). I have added a test to the
2892 spa_base64_to_bits() function which I hope fixes it.
2895 36. The daemon start-up calls getloadavg() while still root for those OS that
2896 need the first call to be done as root, but it missed one case: when
2897 deliver_queue_load_max is set with deliver_drop_privilege. This is
2898 necessary for the benefit of the queue runner, because there is no re-exec
2899 when deliver_drop_privilege is set.
2901 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2902 was in use, was not putting the data itself into the right store pool;
2903 consequently, it could be overwritten for a subsequent message in the same
2904 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2907 38. Sometimes the final signoff response after QUIT could fail to get
2908 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2909 < 0 before doing a fflush(). This bug looks as though it goes back to the
2910 introduction of TLS in release 3.20, but "sometimes" must have been rare
2911 because the tests only now provoked it.
2917 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2918 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2919 Now it returns an integer. A number of calls were still expecting a T/F
2920 return. Fortuitously, in all cases, the tests worked in OK situations,
2921 which is the norm. However, things would have gone wrong on any write
2922 failures on the smtp file descriptor. This function is used when sending
2923 messages over SMTP and also when doing verify callouts.
2925 2. When Exim is called to do synchronous delivery of a locally submitted
2926 message (the -odf or -odi options), it no longer closes stderr before doing
2929 3. Implemented the mua_wrapper option.
2931 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2933 5. Implemented the functions header_remove(), header_testname(),
2934 header_add_at_position(), and receive_remove_recipient(), and exported them
2937 6. If an ACL "warn" statement specified the addition of headers, Exim already
2938 inserted X-ACL-Warn: at the start if there was no header name. However, it
2939 was not making this test for the second and subsequent header lines if
2940 there were newlines in the string. This meant that an invalid header could
2941 be inserted if Exim was badly configured.
2943 7. Allow an ACL "warn" statement to add header lines at the start or after all
2944 the Received: headers, as well as at the end.
2946 8. Added the rcpt_4xx retry error code.
2948 9. Added postmaster_mailfrom=xxx to callout verification option.
2950 10. Added mailfrom=xxxx to the callout verification option, for verify=
2953 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2954 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2956 12. Inserted some casts to stop certain compilers warning when using pointer
2957 differences as field lengths or precisions in printf-type calls (mostly
2958 affecting debugging statements).
2960 13. Added optional readline() support for -be (dynamically loaded).
2962 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2963 same clock tick as a message's arrival, so that its received time was the
2964 same as the "first fail" time on the retry record, and that message
2965 remained on the queue past the ultimate address timeout, every queue runner
2966 would try a delivery (because it was past the ultimate address timeout) but
2967 after another failure, the ultimate address timeout, which should have then
2968 bounced the address, did not kick in. This was a "< instead of <=" error;
2969 in most cases the first failure would have been in the next clock tick
2970 after the received time, and all would be well.
2972 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2973 being recognized when the domain list was tested by the match_domain
2974 condition in an expansion string.
2976 16. Added the ${str2b64: operator.
2978 17. Exim was always calling setrlimit() to set a large limit for the number of
2979 processes, without checking whether the existing limit was already
2980 adequate. (It did check for the limit on file descriptors.) Furthermore,
2981 errors from getrlimit() and setrlimit() were being ignored. Now they are
2982 logged to the main and panic logs, but Exim does carry on, to try to do its
2983 job under whatever limits there are.
2985 18. Imported PCRE 5.0.
2987 19. Trivial typo in log message " temporarily refused connection" (the leading
2990 20. If the log selector return_path_on_delivery was set and an address was
2991 redirected to /dev/null, the delivery process crashed because it assumed
2992 that a return path would always be set for a "successful" delivery. In this
2993 case, the whole delivery is bypassed as an optimization, and therefore no
2996 21. Internal re-arrangement: the function for sending a challenge and reading
2997 a response while authentication was assuming a zero-terminated challenge
2998 string. It's now changed to take a pointer and a length, to allow for
2999 binary data in such strings.
3001 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3003 23. Exim was not respecting finduser_retries when seeking the login of the
3004 uid under which it was called; it was always trying 10 times. (The default
3005 setting of finduser_retries is zero.) Also, it was sleeping after the final
3006 failure, which is pointless.
3008 24. Implemented tls_on_connect_ports.
3010 25. Implemented acl_smtp_predata.
3012 26. If the domain in control=submission is set empty, Exim assumes that the
3013 authenticated id is a complete email address when it generates From: or
3014 Sender: header lines.
3016 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3017 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3018 chown and chgrp in /bin and hostname in /usr/bin.
3020 28. Exim was keeping the "process log" file open after each use, just as it
3021 does for the main log. This opens the possibility of it remaining open for
3022 long periods when the USR1 signal hits a daemon. Occasional processlog
3023 errors were reported, that could have been caused by this. Anyway, it seems
3024 much more sensible not to leave this file open at all, so that is what now
3027 29. The long-running daemon process does not normally write to the log once it
3028 has entered its main loop, and it closes the log before doing so. This is
3029 so that log files can straightforwardly be renamed and moved. However,
3030 there are a couple of unusual error situations where the daemon does write
3031 log entries, and I had neglected to close the log afterwards.
3033 30. The text of an SMTP error response that was received during a remote
3034 delivery was being truncated at 512 bytes. This is too short for some of
3035 the long messages that one sometimes sees. I've increased the limit to
3038 31. It is now possible to make retry rules that apply only when a message has a
3039 specific sender, in particular, an empty sender.
3041 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3042 it possible to be selective about when SMTP synchronization is enforced.
3044 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3046 32. Implemented hosts_connection_nolog.
3048 33. Added an ACL for QUIT.
3050 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3053 35. Added mailbox_size and mailbox_filecount to appendfile.
3055 36. Added control = no_multiline_responses to ACLs.
3057 37. There was a bug in the logic of the code that waits for the clock to tick
3058 in the case where the clock went backwards by a substantial amount such
3059 that the microsecond fraction of "now" was more than the microsecond
3060 fraction of "then" (but the whole seconds number was less).
3062 38. Added support for the libradius Radius client library this is found on
3063 FreeBSD (previously only the radiusclient library was supported).
3069 1. When certain lookups returned multiple values in the form name=value, the
3070 quoting of the values was not always being done properly. Specifically:
3071 (a) If the value started with a double quote, but contained no whitespace,
3073 (b) If the value contained whitespace other than a space character (i.e.
3074 tabs or newlines or carriage returns) it was not quoted.
3075 This fix has been applied to the mysql and pgsql lookups by writing a
3076 separate quoting function and calling it from the lookup code. The fix
3077 should probably also be applied to nisplus, ibase and oracle lookups, but
3078 since I cannot test any of those, I have not disturbed their existing code.
3080 2. A hit in the callout cache for a specific address caused a log line with no
3081 reason for rejecting RCPT. Now it says "Previous (cached) callout
3082 verification failure".
3084 3. There was an off-by-one bug in the queryprogram router. An over-long
3085 return line was truncated at 256 instead of 255 characters, thereby
3086 overflowing its buffer with the terminating zero. As well as fixing this, I
3087 have increased the buffer size to 1024 (and made a note to document this).
3089 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3090 when Exim is waiting for an SMTP response from a remote server, Exim
3091 restarts its select() call on the socket, thereby resetting its timeout.
3092 This is not a problem when such interrupts are rare. Somebody set up a cron
3093 job to run exiwhat every 2 minutes, which is less than the normal select()
3094 timeout (5 or 10 minutes). This meant that the select() timeout never
3095 kicked in because it was always reset. I have fixed this by comparing the
3096 time when an interrupt arrives with the time at the start of the first call
3097 to select(). If more time than the timeout has elapsed, the interrupt is
3098 treated as a timeout.
3100 5. Some internal re-factoring in preparation for the addition of Sieve
3101 extensions (by MH). In particular, the "personal" test is moved to a
3102 separate function, and given an option for scanning Cc: and Bcc: (which is
3103 not set for Exim filters).
3105 6. When Exim created an email address using the login of the caller as the
3106 local part (e.g. when creating a From: or Sender: header line), it was not
3107 quoting the local part when it contained special characters such as @.
3109 7. Installed new OpenBSD configuration files.
3111 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3112 try to make them clearer.
3114 9. Callout options, other than the timeout value, were being ignored when
3115 verifying sender addresses in header lines. For example, when using
3117 verify = header_sender/callout=no_cache
3119 the cache was (incorrectly) being used.
3121 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3122 only the Cygwin environment.
3124 11. When return_path_on_delivery was set as a log selector, if different remote
3125 addresses in the same message used different return paths and parallel
3126 remote delivery occurred, the wrong values would sometimes be logged.
3127 (Whenever a remote delivery process finished, the return path value from
3128 the most recently started remote delivery process was logged.)
3130 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3131 lines when AUTH and/or TLS are in use. This is the "received protocol"
3132 field. Exim used to use "asmtp" for authenticated SMTP, without any
3133 indication (in the protocol name) for TLS use. Now it follows the RFC and
3134 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3135 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3136 names appear in log lines as well as in Received: header lines.
3138 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3139 extensions, and comparison tests, and to fix some bugs.
3141 14. Changes to the "personal" filter test:
3143 (1) The test was buggy in that it was just doing the equivalent of
3144 "contains" tests on header lines. For example, if a user's address was
3145 anne@some.where, the "personal" test would incorrectly be true for
3147 To: susanne@some.where
3149 This test is now done by extracting each address from the header in turn,
3150 and checking the entire address. Other tests that are part of "personal"
3151 are now done using regular expressions (for example, to check local parts
3152 of addresses in From: header lines).
3154 (2) The list of non-personal local parts in From: addresses has been
3155 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3156 taken from the Sieve specification recommendations.
3158 (3) If the message contains any header line starting with "List-" it is
3159 treated as non-personal.
3161 (4) The test for "circular" in the Subject: header line has been removed
3162 because it now seems ill-conceived.
3164 15. Minor typos in src/EDITME comments corrected.
3166 16. Installed latest exipick from John Jetmore.
3168 17. If headers_add on a router specified a text string that was too long for
3169 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3170 of string_sprintf() is now avoided.
3172 18. $message_body_size was not set (it was always zero) when running the DATA
3173 ACL and the local_scan() function.
3175 19. For the "mail" command in an Exim filter, no default was being set for
3176 the once_repeat time, causing a random time value to be used if "once" was
3177 specified. (If the value happened to be <= 0, no repeat happened.) The
3178 default is now 0s, meaning "never repeat". The "vacation" command was OK
3179 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3180 (I found it when inspecting the code).
3182 20. There is now an overall timeout for performing a callout verification. It
3183 defaults to 4 times the callout timeout, which applies to individual SMTP
3184 commands during the callout. The overall timeout applies when there is more
3185 than one host that can be tried. The timeout is checked before trying the
3186 next host. This prevents very long delays if there are a large number of
3187 hosts and all are timing out (e.g. when the network connections are timing
3188 out). The value of the overall timeout can be changed by specifying an
3189 additional sub-option for "callout", called "maxwait". For example:
3191 verify = sender/callout=5s,maxwait=20s
3193 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3194 to the end before writing, but this should make it even safer).
3196 22. Exim was forgetting that it had advertised PIPELINING for the second and
3197 subsequent messages on an SMTP connection. It was also not resetting its
3198 memory on STARTTLS and an internal HELO.
3200 23. When Exim logs an SMTP synchronization error within a session, it now
3201 records whether PIPELINING has been advertised or not.
3203 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3204 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3205 rather than long int.
3207 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3209 26. Added the never_mail option to autoreply.
3215 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3216 crash if the getsockname() call failed; this can happen if a connection is
3217 closed very soon after it is established. The problem was simply in the
3218 order in which certain operations were done, causing Exim to try to write
3219 to the SMTP stream before it had set up the file descriptor. The bug has
3220 been fixed by making things happen in the correct order.
3226 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3227 before the connection was closed, thus losing the rejection response.
3229 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3230 some early Solaris releases, but causes trouble in current releases where
3231 socklen_t is defined.
3233 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3236 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3239 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3240 that releasing the top of it at the end releases what was used for sub-
3241 expansions (unless the block got too big). However, discard this block if
3242 the first thing is a variable or header, so that we can use its block when
3243 it is dynamic (useful for very large $message_headers, for example).
3245 6. Lookups now cache *every* query, not just the most recent. A new, separate
3246 store pool is used for this. It can be recovered when all lookup caches are
3247 flushed. Lookups now release memory at the end of their result strings.
3248 This has involved some general refactoring of the lookup sources.
3250 7. Some code has been added to the store_xxx() functions to reduce the amount
3251 of flapping under certain conditions.
3253 8. log_incoming_interface used to affect only the <= reception log lines. Now
3254 it causes the local interface and port to be added to several more SMTP log
3255 lines, for example "SMTP connection from", and rejection lines.
3257 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3259 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3261 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3262 could be overwritten at the end of the current message (or the start of a
3263 new message if it was set in a HELO ACL). The value is now preserved for
3264 the duration of the SMTP connection.
3266 12. If a transport had a headers_rewrite setting, and a matching header line
3267 contained an unqualified address, that address was qualified, even if it
3268 did not match any rewriting rules. The underlying bug was that the values
3269 of the flags that permit the existence of unqualified sender and recipient
3270 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3271 non-local messages, and by -bnq for local messages) were not being
3272 preserved with the message after it was received.
3274 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3275 "next input=" as part of the text comprising the host identity instead of
3276 the correct text. The code was using the same buffer for two different
3277 strings. However, depending on which order the printing function evaluated
3278 its arguments, the bug did not always show up. Under Linux, for example, my
3279 test suite worked just fine.
3281 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3282 doesn't work with some older versions of Perl. It has been changed to "my",
3283 which in any case is probably the better facility to use.
3285 15. A really picky compiler found some instances of statements for creating
3286 error messages that either had too many or two few arguments for the format
3289 16. The size of the buffer for calls to the DNS resolver has been increased
3290 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3291 for addresses that have a lot of PTR records. This alleviates a problem; it
3292 does not fully solve it.
3294 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3295 buffer now truncates the list and logs the incident, which is the same
3296 action as happens when Exim is looking up a host name and its aliases.
3297 Previously in this situation something unpredictable would happen;
3298 sometimes it was "internal error: store_reset failed".
3300 18. If a server dropped the connection unexpectedly when an Exim client was
3301 using GnuTLS and trying to read a response, the client delivery process
3302 crashed while trying to generate an error log message.
3304 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3305 string, for example:
3307 warn message = H1: something\nH2: something
3309 the text was added as a single header line from Exim's point of view
3310 though it ended up OK in the delivered message. However, searching for the
3311 second and subsequent header lines using $h_h2: did not work. This has been
3312 fixed. Similarly, if a system filter added multiple headers in this way,
3313 the routers could not see them.
3315 20. Expanded the error message when iplsearch is called with an invalid key to
3316 suggest using net-iplsearch in a host list.
3318 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3319 ACLs are no longer actually imposed (and a message to that effect is
3322 22. If a "gecos" field in a passwd entry contained escaped characters, in
3323 particular, if it contained a \" sequence, Exim got it wrong when building
3324 a From: or a Sender: header from that name. A second bug also caused
3325 incorrect handling when an unquoted " was present following a character
3326 that needed quoting.
3328 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3329 was not being matched caselessly.
3331 24. Arranged for all hyphens in the exim.8 source to be escaped with
3334 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3335 properly. Recipient callout cache records were still being keyed to include
3336 the sender, even when use_sender was set false. This led to far more
3337 callouts that were necessary. The sender is no longer included in the key
3338 when use_sender is false.
3340 26. Added "control = submission" modifier to ACLs.
3342 27. Added the ${base62d: operator to decode base 62 numbers.
3344 28. dnsdb lookups can now access SRV records.
3346 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3347 the configuration file.
3349 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3350 (-v) mode. This makes the output for a verbose queue run more intelligible.
3352 31. Added a use_postmaster feature to recipient callouts.
3354 32. Added the $body_zerocount variable, containing the number of binary zero
3355 bytes in the message body.
3357 33. The time of last modification of the "new" subdirectory is now used as the
3358 "mailbox time last read" when there is a quota error for a maildir
3361 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3363 35. Added +ignore_unknown as a special item in host lists.
3365 36. Code for decoding IPv6 addresses in host lists is now included, even if
3366 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3367 address was recognized as an IP address, but was then not correctly decoded
3368 into binary, causing unexpected and incorrect effects when compared with
3375 1. Very minor rewording of debugging text in manualroute to say "list of
3376 hosts" instead of "hostlist".
3378 2. If verify=header_syntax was set, and a header line with an unqualified
3379 address (no domain) and a large number of spaces between the end of the
3380 name and the colon was received, the reception process suffered a buffer
3381 overflow, and (when I tested it) crashed. This was caused by some obsolete
3382 code that should have been removed. The fix is to remove it!
3384 3. When running in the test harness, delay a bit after writing a bounce
3385 message to get a bit more predictability in the log output.
3387 4. Added a call to search_tidyup() just before forking a reception process. In
3388 theory, someone could use a lookup in the expansion of smtp_accept_max_
3389 per_host which, without the tidyup, could leave open a database connection.
3391 5. Added the variables $recipient_data and $sender_data which get set from a
3392 lookup success in an ACL "recipients" or "senders" condition, or a router
3393 "senders" option, similar to $domain_data and $local_part_data.
3395 6. Moved the writing of debug_print from before to after the "senders" test
3398 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3399 problems for message scanning, either using a data ACL, or using
3400 local_scan() because the Received: header was not generated till after they
3401 were called (in order to set the time as the time of reception completion).
3402 I have revised the way this works. The header is now generated after the
3403 body is received, but before the ACL or local_scan() are called. After they
3404 are run, the timestamp in the header is updated.
3410 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3411 before starting a queue runner without re-exec. This happened only when
3412 deliver_drop_privilege was set or when the Exim user was set to root. The
3413 effect of the bug was that timeouts during subsequent deliveries caused
3414 crashes instead of being properly handled. The handler is now left at its
3415 default (and expected) setting.
3417 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3418 message, again when deliver_drop_privilege is set or Exim is run as root.
3419 The bug described in (1) was not present in this case, but the tidying up
3420 of the other signals was missing. I have made the two cases consistent.
3422 3. The ignore_target_hosts setting on a manualroute router was being ignored
3423 for hosts that were looked up using the /MX notation.
3425 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3428 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3429 operated on the sender address. After changing the $sender_address to <>
3430 for the sender address verify, Exim was re-instated it as the original
3431 (before rewriting) address, but remembering that it had rewritten it, so it
3432 wasn't rewriting it again. This bug also had the effect of breaking the
3433 sender address verification caching when the sender address was rewritten.
3435 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3436 This has been changed so that if the ip literal address matches
3437 ignore_target_hosts, the router declines.
3439 7. Added expansion conditions match_domain, match_address, and match_local_
3440 part (NOT match_host).
3442 8. The placeholder for the Received: header didn't have a length field set.
3444 9. Added code to Exim itself and to exim_lock to test for a specific race
3445 condition that could lead to file corruption when using MBX delivery. The
3446 issue is with the lockfile that is created in /tmp. If this file is removed
3447 after a process has opened it but before that process has acquired a lock,
3448 there is the potential for a second process to recreate the file and also
3449 acquire a lock. This could lead to two Exim processes writing to the file
3450 at the same time. The added code performs the same test as UW imapd; it
3451 checks after acquiring the lock that its file descriptor still refers to
3452 the same named file.
3454 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3455 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3458 11. Added the smtp_active_hostname option. If used, this will typically be made
3459 to depend on the incoming interface address. Because $interface_address is
3460 not set up until the daemon has forked a reception process, error responses
3461 that can happen earlier (such as "too many connections") no longer contain
3464 12. If an expansion in a condition on a "warn" statement fails because a lookup
3465 defers, the "warn" statement is abandoned, and the next ACL statement is
3466 processed. Previously this caused the whole ACL to be aborted.
3468 13. Added the iplsearch lookup type.
3470 14. Added ident_timeout as a log selector.
3472 15. Added tls_certificate_verified as a log selector.
3474 16. Added a global option tls_require_ciphers (compare the smtp transport
3475 option of the same name). This controls incoming TLS connections.
3477 17. I finally figured out how to make tls_require_ciphers do a similar thing
3478 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3479 before starting the TLS session.
3481 18. Tabs are now shown as \t in -bP output.
3483 19. If the log selector return_path_on_delivery was set, Exim crashed when
3484 bouncing a message because it had too many Received: header lines.
3486 20. If two routers both had headers_remove settings, and the first one included
3487 a superfluous trailing colon, the final name in the first list and the
3488 first name in the second list were incorrectly joined into one item (with a
3489 colon in the middle).
3495 1. Added -C and -D options to the exinext utility, mainly to make it easier
3496 to include in the automated testing, but these could be helpful when
3497 multiple configurations are in use.
3499 2. The exinext utility was not formatting the output nicely when there was
3500 an alternate port involved in the retry record key, nor when there was a
3501 message id as well (for retries that were specific to a specific message
3502 and a specific host). It was also confused by IPv6 addresses, because of
3503 the additional colons they contain. I have fixed the IPv4 problem, and
3504 patched it up to do a reasonable job for IPv6.
3506 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3507 delivery, the log line now contains "pipelined" if PIPELINING was used.
3509 4. An SMTP transport process used to panic and die if the bind() call to set
3510 an explicit outgoing interface failed. This has been changed; it is now
3511 treated in the same way as a connect() failure.
3513 5. A reference to $sender_host_name in the part of a conditional expansion
3514 that was being skipped was still causing a DNS lookup. This no longer
3517 6. The def: expansion condition was not recognizing references to header lines
3518 that used bh_ and bheader_.
3520 7. Added the _cache feature to named lists.
3522 8. The code for checking quota_filecount in the appendfile transport was
3523 allowing one more file than it should have been.
3525 9. For compatibility with Sendmail, the command line option
3533 and sets the incoming protocol and host name (for trusted callers). The
3534 host name and its colon can be omitted when only the protocol is to be set.
3535 Note the Exim already has two private options, -pd and -ps, that refer to
3536 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3537 "s", but I don't think that's a major issue.
3539 10. A number of refactoring changes to the code, none of which should affect
3542 (a) The number of logging options was getting close to filling up the
3543 32-bit word that was used as a bit map. I have split them into two classes:
3544 those that are passed in the argument to log_write(), and those that are
3545 only ever tested independently outside of that function. These are now in
3546 separate 32-bit words, so there is plenty of room for expansion again.
3547 There is no change in the user interface or the logging behaviour.
3549 (b) When building, for example, log lines, the code previously used a
3550 macro that called string_cat() twice, in order to add two strings. This is
3551 not really sufficiently general. Furthermore, there was one instance where
3552 it was actually wrong because one of the argument was used twice, and in
3553 one call a function was used. (As it happened, calling the function twice
3554 did not affect the overall behaviour.) The macro has been replaced by a
3555 function that can join an arbitrary number of extra strings onto a growing
3558 (c) The code for expansion conditions now uses a table and a binary chop
3559 instead of a serial search (which was left over from when there were very
3560 few conditions). Also, it now recognizes conditions like "pam" even when
3561 the relevant support is not compiled in: a suitably worded error message is
3562 given if an attempt is made to use such a condition.
3564 11. Added ${time_interval:xxxxx}.
3566 12. A bug was causing one of the ddress fields not to be passed back correctly
3567 from remote delivery subprocesses. The field in question was not being
3568 subsequently used, so this caused to problems in practice.
3570 13. Added new log selectors queue_time and deliver_time.
3572 14. Might have fixed a bug in maildirsizefile handling that threw up
3573 "unexpected character" debug warnings, and recalculated the data
3574 unnecessarily. In any case, I expanded the warning message to give more
3577 15. Added the message "Restricted characters in address" to the statements in
3578 the default ACL that block characters like @ and % in local parts.
3580 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3581 Three changes have been made:
3583 (a) There was a serious bug; a negative response to MAIL caused the whole
3584 recipient domain to be cached as invalid, thereby blocking all messages
3585 to all local parts at the same domain, from all senders. This bug has
3586 been fixed. The domain is no longer cached after a negative response to
3587 MAIL if the sender used is not empty.
3589 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3592 (c) A new callout option, "use_sender" has been added for people who want
3593 the modified behaviour.
3599 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3602 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3603 indeed breaks things for older releases.
3605 3. Added additional logging to the case where there is a problem reading data
3606 from a filter that is running in a subprocess using a pipe, in order to
3607 try to track down a specific problem.
3609 4. Testing facility fudge: when running in the test harness and attempting
3610 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3611 getting "No route to host". Convert this to a timeout.
3613 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3616 6. Some OS don't have socklen_t but use size_t instead. This affects the
3617 fifth argument of getsockopt() amongst other things. This is now
3618 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3619 can be set for individual OS. I have set it for SunOS5, OSF1, and
3620 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3621 some earlier ones do not.
3623 7. Change 4.30/15 was not doing the test caselessly.
3625 8. The standard form for an IPv6 address literal was being rejected by address
3626 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3627 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3628 this, as well as the form without the "IPv6" on the front (but only when
3629 address literals are enabled, of course).
3631 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3633 10. Exim crashed if a message with an empty sender address specified by -f
3634 encountered a router with an errors_to setting. This could be provoked only
3635 by a command such as
3639 where an empty string was supplied; "<>" did not hit this bug.
3641 11. Installed PCRE release 4.5.
3643 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3644 remained set. It is now erased.
3646 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3647 times from message ids (which are base 36 rather than the normal 62).
3649 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3650 were being counted as actual protocol errors, and logged if the log
3651 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3652 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3653 and DATA following a set of rejected RCPTs do not count as protocol errors.
3654 In other words, Exim assumes they were pipelined, though this may not
3655 actually be the case. Of course, in all cases the client gets an
3656 appropriate error code.
3658 15. If a lookup fails in an ACL condition, a message about the failure may
3659 be available; it is used if testing the ACL cannot continue, because most
3660 such messages specify what the cause of the deferral is. However, some
3661 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3662 that caused an old message to be retained and used if a later statement
3663 caused a defer, replacing the real cause of the deferral.
3665 16. If an IP address had so many PTR records that the DNS lookup buffer
3666 was not large enough to hold them, Exim could crash while trying to process
3667 the truncated data. It now detects and logs this case.
3669 17. Further to 4.21/58, another change has been made: if (and only if) the
3670 first line of a message (the first header line) ends with CRLF, a bare LF
3671 in a subsequent header line has a space inserted after it, so as not to
3672 terminate the header.
3674 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3675 unnecessarily, used atoi() instead of strtol(), and didn't check the
3676 termination when getting file sizes from file names by regex.
3678 19. Completely re-implemented the support for maildirsize files, in the light
3679 of a number of problems with the previous contributed implementation
3680 (4.30/29). In particular:
3682 . If the quota is zero, the maildirsize file is maintained, but no quota is
3685 . If the maildir directory does not exist, it is created before any attempt
3686 to write a maildirsize file.
3688 . The quota value in the file is just a cache; if the quota is changed in
3689 the transport, the new value overrides.
3691 . A regular expression is available for excluding directories from the
3694 20. The autoreply transport checks the characters in options that define the
3695 message's headers; it allows continued headers, but it was checking with
3696 isspace() after an embedded newline instead of explicitly looking for a
3699 21. If all the "regular" hosts to which an address was routed had passed their
3700 expiry times, and had not reached their retry times, the address was
3701 bounced, even if fallback hosts were defined. Now Exim should go on to try
3704 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3705 equivalent code in the SMTP transport. Some hosts send humungous responses
3706 to HELO/EHLO, more than 1024 it seems.
3708 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3709 gives compiler warnings in some environments. I've now done it "properly",
3712 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3713 (because of problems with the built-in one) was declared to return uschar *
3714 instead of char *, causing compiler failure.
3716 25. Fixed a file descriptor leak when processing alias/forward files.
3718 26. Fixed a minor format string issue in dbfn.c.
3720 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3722 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3723 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3726 29. When a sender address is verified, it is cached, to save repeating the test
3727 when there is more than one recipient in a message. However, when the
3728 verification involves a callout, it is possible for different callout
3729 options to be set for different recipients. It is too complicated to keep
3730 track of this in the cache, so now Exim always runs a verification when a
3731 callout is required, relying on the callout cache for the optimization.
3732 The overhead is duplication of the address routing, but this should not be
3735 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3736 to be verified with callout=postmaster, and the main callout worked but the
3737 postmaster check failed, the verification correctly failed. However, if a
3738 subsequent RCPT command asked for sender verification *without* the
3739 postmaster check, incorrect caching caused this verification also to fail,
3742 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3743 it was not caching the DNS options (qualify_single, search_parents) that
3744 were used when the lookup failed. A subsequent lookup with different
3745 options therefore always gave the same answer, though there were cases
3746 where it should not have. (Example: a "domains = !$mx_any" option on a
3747 dnslookup router: the "domains" option is always processed without any
3748 widening, but the router might have qualify_single set.) Now Exim uses the
3749 cached value only when the same options are set.
3751 32. Added John Jetmore's "exipick" utility to the distribution.
3753 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3754 than a timeout (e.g. a certificate is required, and is not provided), an
3755 Exim server now closes the connection immediately. Previously it waited for
3756 the client to close - but if the client is SSL, it seems that they each
3757 wait for each other, leading to a delay before one of them times out.
3759 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3760 maintained 0.8.x compatibility because I don't think many are using it, and
3761 it is clearly obsolete.
3763 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3766 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3767 client certificate was expired. A simple patch fixes this, though I don't
3768 understand the full logic of why the verify callback is called multiple
3771 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3772 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3773 which causes problems with some clients (such as the Certicom SSL Plus
3774 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3775 disables the coutermeasure allowing Eudora to connect."
3777 38. Exim was not checking that a write() to a log file succeeded. This could
3778 lead to Bad Things if a log got too big, in particular if it hit a file
3779 size limit. Exim now panics and dies if it cannot write to a log file, just
3780 as it does if it cannot open a log file.
3782 39. Modified OS/Makefile-Linux so that it now contains
3784 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3786 The two -D definitions ensure that Exim is compiled with large file
3787 support, which makes it possible to handle log files that are bigger than
3790 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3791 instance) a domain was checked against a named list that involved a lookup,
3792 causing $domain_data to be set, then another domain was checked against the
3793 same list, then the first domain was re-checked, the value of $domain_data
3794 after the final check could be wrong. In particular, if the second check
3795 failed, it could be set empty. This bug probably also applied to
3798 41. The strip_trailing_dot option was not being applied to the address given
3799 with the -f command-line option.
3801 42. The code for reading a message's header from the spool was incrementing
3802 $received_count, but never initializing it. This meant that the value was
3803 incorrect (doubled) while delivering a message in the same process in which
3804 it was received. In the most common configuration of Exim, this never
3805 happens - a fresh exec is done - but it can happen when
3806 deliver_drop_privilege is set.
3808 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3809 it now includes up to 150 characters of the unexpected data in the log
3812 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3813 and building data strings. The size of both of these buffers was 10 000
3814 bytes - far larger than anybody would *ever* want, thought I. Needless to
3815 say, somebody hit the limit. I have increased the maximum line length to
3816 20 000 and the maximum data length of concatenated lines to 100 000. I have
3817 also fixed two bugs, because there was no checking on these buffers. Tsk,
3818 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3819 buffer is too small.
3821 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3822 lsearch lookups. Now it does.
3824 46. When parsing a route_list item in a manualroute router, a fixed-length
3825 buffer was used for the list of hosts. I made this 1024 bytes long,
3826 thinking that nobody would ever have a list of hosts that long. Wrong.
3827 Somebody had a whole pile of complicated expansion conditions, and the
3828 string was silently truncated, leading to an expansion error. It turns out
3829 that it is easier to change to an unlimited length (owing to other changes
3830 that have happened since this code was originally written) than to build
3831 structure for giving a limitation error. The length of the item that
3832 expands into the list of hosts is now unlimited.
3834 47. The lsearch lookup could not handle data where the length of text line was
3835 more than 4095 characters. Such lines were truncated, leading to shortened
3836 data being returned. It should now handle lines of any length.
3838 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3839 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3842 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3843 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3844 no longer happens in comment lines. A list of replacements is now placed
3845 at the head of all of the source files, except those whose only change is
3846 to replace PERL_COMMAND in the very first #! line.
3848 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3849 messages on the queue, with a bottom-up merge sort, using code contributed
3850 by Michael Haardt. This should make operations like -bp somewhat faster on
3851 large queues. It won't affect queue runners, except when queue_run_in_order
3854 51. Installed eximstats 1.31 in the distribution.
3856 52. Added support for SRV lookups to the dnslookup router.
3858 53. If an ACL referred to $message_body or $message_body_end, the value was not
3859 reset for any messages that followed in the same SMTP session.
3861 54. The store-handling optimization for building very long strings was not
3862 differentiating between the different store pools. I don't think this
3863 actually made any difference in practice, but I've tidied it.
3865 55. While running the routers to verify a sender address, $sender_address
3866 was still set to the sender address. This is wrong, because when routing to
3867 send a bounce to the sender, it would be empty. Therefore, I have changed
3868 it so that, while verifying a sender address, $sender_address is set to <>.
3869 (There is no change to what happens when verifying a recipient address.)
3871 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3872 target A or AAAA records (if not already returned) without resetting the
3873 qualify_single or search_parents options of the DNS resolver. These are
3874 inappropriate in this case because the targets of MX and SRV records must
3875 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3876 target that, when qualified, matched something in the local domain. These
3877 two options are now turned off when doing these lookups.
3879 57. It seems that at least some releases of Reiserfs (which does not have the
3880 concept of a fixed number of inodes) returns zero and not -1 for the
3881 number of available inodes. This interacted badly with check_spool_inodes,
3882 which assumed that -1 was the "no such thing" setting. What I have done is
3883 to check that the total number of inodes is greater than zero before doing
3884 the test of how many are available.
3886 58. When a "warn" ACL statement has a log_message modifier, the message is
3887 remembered, and not repeated. This is to avoid a lot of repetition when a
3888 message has many recipients that cause the same warning to be written.
3889 Howewer, Exim was preserving the list of already written lines for an
3890 entire SMTP session, which doesn't seem right. The memory is now reset if a
3891 new message is started.
3893 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3894 debugging output unless log_rewrite was also set.
3896 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3897 of (int)(handle) when we know that handle contains (void *)(-1).
3899 61. The Exim daemon panic-logs an error return when it closes the incoming
3900 connection. However "connection reset by peer" seems to be common, and
3901 isn't really an error worthy of noting specially, so that particular error
3904 62. When Exim is trying to find all the local interfaces, it used to panic and
3905 die if the ioctl to get the interface flags failed. However, it seems that
3906 on at least one OS (Solaris 9) it is possible to have an interface that is
3907 included in the list of interfaces, but for which you get a failure error
3908 for this call. This happens when the interface is not "plumbed" into a
3909 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3910 failure of the "get flags" call assumes that the interface is down.
3912 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3913 makes life easier for people who are doing arithmetic on fields extracted
3914 from dates, where you often get leading zeros that should not be
3915 interpreted as octal.
3917 64. Added qualify_domain to the redirect router, to override the global
3920 65. If a pathologically long header line contained very many addresses (the
3921 report of this problem mentioned 10 000) and each of them was rewritten,
3922 Exim could use up a very large amount of memory. (It kept on making new
3923 copies of the header line as it rewrote, and never released the old ones.)
3924 At the expense of a bit more processing, the header rewriting function has
3925 been changed so that it no longer eats memory in this way.
3927 66. The generation of the Received: header has been moved from the time that a
3928 message starts to be received, to the time that it finishes. The timestamp
3929 in the Received: header should now be very close to that of the <= log
3930 line. There are two side-effects of this change:
3932 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3933 logged header lines no longer include the local Received: line, because
3934 it has not yet been created. The same applies to a copy of the message
3935 that is returned to a non-SMTP sender when a message is rejected.
3937 (b) When a filter file is tested using -bf, no additional Received: header
3938 is added to the test message. After some thought, I decided that this
3941 This change does not affect the value of $received_for. It is still set
3942 after address rewriting, but before local_scan() is called.
3944 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3946 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3947 gave an unhelpful panic error message, and a defer error. I have managed to
3948 change this behaviour so that it now rejects any supplied certificate,
3949 which seems right, as the list of acceptable certificates is empty.
3951 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3952 gave an unhelpful defer error. I have not managed to make this reject any
3953 supplied certificates, but the error message it gives is "no certificate
3954 supplied", which is not helpful.
3956 70. exigrep's output now also includes lines that are not associated with any
3957 message, but which match the given pattern. Implemented by a patch from
3958 Martin Sluka, which also tidied up the Perl a bit.
3960 71. Recipient callout verification, like sender verification, was using <> in
3961 the MAIL FROM command. This isn't really the right thing, since the actual
3962 sender may affect whether the remote host accepts the recipient or not. I
3963 have changed it to use the actual sender in the callout; this means that
3964 the cache record is now keyed on a recipient/sender pair, not just the
3965 recipient address. There doesn't seem to be a real danger of callout loops,
3966 since a callout by the remote host to check the sender would use <>.
3967 [SEE ABOVE: changed after hitting problems.]
3969 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3970 temporary errors. However, in the case of such a code being given after
3971 the end of a data transmission (i.e. after ".") Exim was failing to write
3972 a retry record for the message. (Yes, there was some broken host that was
3973 actually sending 8xx at this point.)
3975 73. An unknown lookup type in a host list could cause Exim to panic-die when
3976 the list was checked. (An example that provoked this was putting <; in the
3977 middle of a list instead of at the start.) If this happened during a DATA
3978 ACL check, a -D file could be left lying around. This kind of configuration
3979 error no longer causes Exim to die; instead it causes a defer errror. The
3980 incident is still logged to the main and panic logs.
3982 74. Buglet left over from Exim 3 conversion. The message "too many messages
3983 in one connection" was written to the rejectlog but not the mainlog, except
3984 when address rewriting (yes!) was being logged.
3986 75. Added write_rejectlog option.
3988 76. When a system filter was run not as root (that is, when system_filter_user
3989 was set), the values of the $n variables were not being returned to the
3990 main process; thus, they were not subsequently available in the $sn
3993 77. Added +return_path_on_delivery log selector.
3995 78. A connection timeout was being treated differently from recipients deferred
3996 when testing hosts_max_try with a message that was older than the host's
3997 retry timeout. (The host should not be counted, thus allowing all hosts to
3998 be tried at least once before bouncing.) This may have been the cause of an
3999 occasionally reported bug whereby a message would remain on the queue
4000 longer than the retry timeout, but would be bounced if a delivery was
4001 forced. I say "may" because I never totally pinned down the problem;
4002 setting up timeout/retry tests is difficult. See also the next item.
4004 79. The ultimate address timeout was not being applied to errors that involved
4005 a combination of host plus message (for example, a timeout on a MAIL
4006 command). When an address resolved to a number of possible hosts, and they
4007 were not all tried for each delivery (e.g. because of hosts_max_try), a
4008 message could remain on the queue longer than the retry timeout.
4010 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4013 81. Fixed an obscure SMTP outgoing bug which required at least the following
4014 conditions: (a) there was another message waiting for the same server;
4015 (b) the server returned 5xx to all RCPT commands in the first message so
4016 that the message was not completed; (c) the server dropped the connection
4017 or gave a negative response to the RSET that Exim sends to abort the
4018 transaction. The observed case was a dropped connection after DATA that had
4019 been sent in pipelining mode. That is, the server had advertised PIPELINING
4020 but was not implementing it correctly. The effect of the bug was incorrect
4021 behaviour, such as trying another host, and this could lead to a crash.
4027 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4028 and daemon.c were passed as pointers to ints; they should have been
4029 pointers to socklen_t variables (which are typically unsigned ints).
4031 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4034 3. Fixed a really odd bug that affected only the testing scheme; patching a
4035 certain fixed string in the binary changed the value of another string that
4036 happened to be identical to the end of the original first string.
4038 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4039 name", it returns that address as the IP address. On some operating
4040 systems (e.g. Solaris), it also passes back the IP address string as the
4041 "host name". However, on others (e.g. Linux), it passes back an empty
4042 string. Exim wasn't checking for this, and was changing the host name to an
4043 empty string, assuming it had been canonicized.
4045 5. Although rare, it is permitted to have more than one PTR record for a given
4046 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4047 all the names associated with an address, because they do in Solaris.
4048 However, it seems that they do not in Linux for data that comes from the
4049 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4050 I found this out when I moved to a new Linux workstation and tried to run
4051 the Exim test suite.
4053 To get round this problem I have changed the code so that it now does its
4054 own call to the DNS to look up PTR records when searching for a host name.
4055 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4056 addresses that are only in /etc/hosts are still found.
4058 This behaviour is, however, controlled by an option called host_lookup_
4059 order, which defaults to "bydns:byaddr". If people want to use the other
4060 order, or indeed, just use one or the other means of lookup, they can
4061 specify it in this variable.
4063 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4064 some operating systems, this comes back from gethostbyaddr() as an empty
4065 string, and this is what Exim used to test for. However, it seems that in
4066 other systems, "." is yielded. Exim now tests for this case too.
4068 7. The values of check_spool_space and check_log_space are now held internally
4069 as a number of kilobytes instead of an absolute number of bytes. If a
4070 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4071 kilobyte. This means that much larger values can be stored.
4073 8. Exim monitor: an attempt to get the action menu when not actually pointing
4074 at a message produces an empty menu entitled "No message selected". This
4075 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4076 no entries in it ("Shell widget menu has zero width and/or height"). So I
4077 have added a single, blank menu entry in this case.
4079 9. Added ${quote_local_part.
4081 10. MIME decoding is now applied to the contents of Subject: header lines when
4084 11. Now that a reference to $sender_host_address automatically causes a reverse
4085 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4086 host lookup before query-style lookups in lists that might use this
4087 variable. This has therefore been abolished, and the "net-" prefix is no
4088 longer necessary for query-style lookups.
4090 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4091 have been a typo for LFLAGS, so it has been changed.
4093 13. The install script calls Exim with "-C /dev/null" in order to find the
4094 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4095 to be output. Howeve, since Exim outputs its version number before the
4096 error, it didn't break the script. It just looked ugly. I fixed this by
4097 always allowing "-C /dev/null" if the caller is root.
4099 14. Ignore overlarge ACL variable number when reading spool file - insurance
4100 against a later release with more variables having written the file.
4102 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4103 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4104 this, as well as the form without the "IPv6" on the front.
4106 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4107 OS/Makefile-Darwin file.
4109 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4110 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4112 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4113 variable. However, some OS use ints for inodes. I've added cast to long int
4114 to get rid of the compiler warning.
4116 19. I had forgotten to lock out "/../" in configuration file names when
4117 ALT_CONFIG_PREFIX was set.
4119 20. Routers used for verification do not need to specify transports. However,
4120 if such a router generated a host list, and callout was configured, Exim
4121 crashed, because it could not find a port number from the (non-existent)
4122 transport. It now assumes port 25 in this circumstance.
4124 21. Added the -t option to exigrep.
4126 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4127 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4128 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4129 them is set, LOOKUP_LSEARCH is forced.
4131 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4133 24. Added sender and host information to the "rejected by local_scan()" log
4134 line; previously there was no indication of these.
4136 25. Added .include_if_exists.
4138 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4139 Linux. It turns out that not all file systems support this. Apparently some
4140 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4141 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4142 not supported on the file descriptor, is now ignored when Exim is trying to
4143 sync a directory. This applies only to Linux.
4145 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4147 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4148 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4151 29. Added support for maildirsize files from supplied patch (modified a bit).
4153 30. The use of :fail: followed by an empty string could lead Exim to respond to
4154 sender verification failures with (e.g.):
4156 550 Verification failed for <xxx>
4157 550 Sender verify failed
4159 where the first response line was missing the '-' that indicates it is not
4160 the final line of the response.
4162 31. The loop for finding the name of the user that called Exim had a hardwired
4163 limit of 10; it now uses the value of finduser_retries, which is used for
4164 all other user lookups.
4166 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4169 33. Exim was neglecting to zero errno before one call of strtol() when
4170 expanding a string and expecting an integer value. On some systems this
4171 resulted in spurious "integer overflow" errors. Also, it was casting the
4172 result into an int without checking.
4174 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4175 did not work. The code looks as if it has *never* worked, though it appears
4176 to have been documented since at least releast 1.62. I have made it work.
4178 35. The "timeout_DNS" error in retry rules, also documented since at least
4179 1.62, also never worked. As it isn't clear exactly what this means, and
4180 clearly it isn't a major issue, I have abolished the feature by treating it
4181 as "timeout", and writing a warning to the main and panic logs.
4183 36. The display of retry rules for -brt wasn't always showing the error code
4186 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4187 timeout_connect_A, timeout_connect_MX.
4189 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4190 to the empty sender.
4192 39. The daemon was not analysing the content of -oX till after it had closed
4193 stderr and disconnected from the controlling terminal. This meant that any
4194 syntax errors were only noted on the panic log, and the return code from
4195 the command was 0. By re-arranging the code a little, I've made the
4196 decoding happen first, so such errors now appear on stderr, and the return
4197 code is 1. However, the actual setting up of the sockets still happens in
4198 the disconnected process, so errors there are still only recorded on the
4201 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4202 connections (as happens on some IP stacks) was logged at start up time as
4203 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4204 it from "IPv6 and IPv4", which means that two separate sockets are being
4207 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4208 says whether AF_INET or AF_INET6 was passed as an argument.
4210 42. Exiwhat output was messed up when time zones were included in log
4213 43. Exiwhat now gives more information about the daemon's listening ports,
4214 and whether -tls-on-connect was used.
4216 44. The "port" option of the smtp transport is now expanded.
4218 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4219 silently ignored. Now an error message is written to the main and panic
4222 46. There's a new ACL modifier called "logwrite" which writes to a log file
4223 as soon as it is encountered.
4225 47. Added $local_user_uid and $local_user_gid at routing time.
4227 48. Exim crashed when trying to verify a sender address that was being
4230 49. Exim was recognizing only a space character after ".include". It now also
4231 recognizes a tab character.
4233 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4234 extracting the relevant information from the specification. The man page no
4235 longer contains scrambled data for the -d option, and I've added a section
4236 at the front about calling Exim under different names.
4238 51. Added "extra_headers" argument to the "mail" command in filter files.
4240 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4243 53. Installed eximstats 1.29.
4245 54. Added transport_filter_timeout as a generic transport option.
4247 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4248 Cc: header lines. This was required by RFC 822, but it not required by RFC
4251 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4252 incoming messages that did not have them. Now it does so only if the
4253 message originates locally, that is, if there is no associated remote host
4254 address. When Resent- header lines are present, this applies to the Resent-
4255 lines rather than the non-Resent- lines.
4257 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4258 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4260 58. Messages for configuration errors now include the name of the main
4261 configuration file - useful now that there may be more than one file in a
4262 list (.included file names were always shown).
4264 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4265 for those rare installations that do not start the daemon as root or run it
4266 setuid root. I've cut out the call to initgroups() if the daemon is not
4269 60. The Exim user and group can now be bound into the binary as text strings
4270 that are looked up at the start of Exim's processing.
4272 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4274 62. Added $mailstore_basename variable.
4276 63. Installed patch to sieve.c from Michael Haardt.
4278 64. When Exim failed to open the panic log after failing to open the main log,
4279 the original message it was trying to log was written to stderr and debug
4280 output, but if they were not available (the usual case in production), it
4281 was lost. Now it is written to syslog before the two lines that record the
4282 failures to open the logs.
4284 65. Users' Exim filters run in subprocesses under the user's uid. It is
4285 possible for a "deliver" command or an alias in a "personal" command to
4286 provoke an address rewrite. If logging of address rewriting is configured,
4287 this fails because the process is not running as root or exim. There may be
4288 a better way of dealing with this, but for the moment (because 4.30 needs
4289 to be released), I have disabled address rewrite logging when running a
4290 filter in a non-root, non-exim process.
4296 1. The buildconfig auxiliary program wasn't quoting the value set for
4297 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4298 not defined. This bug was masked in 4.22 by the effect that was fixed in
4301 2. Some messages that were rejected after a message id was allocated were
4302 shown as "incomplete" by exigrep. It no longer does this for messages that
4303 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4305 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4306 have allow_domain_literals set, the ID did not get logged in the <= line.
4307 Domain literals are now always recognized in Message-ID: header lines.
4309 4. The first argument for a ${extract expansion item is the key name or field
4310 number. Leading and trailing spaces in this item were not being ignored,
4311 causing some misleading effects.
4313 5. When deliver_drop_privilege was set, single queue runner processes started
4314 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4315 same command in the process it spins off) were not dropping privilege.
4317 6. When the daemon running as "exim" started a queue runner, it always
4318 re-executed Exim in the spun-off process. This is a waste of effort when
4319 deliver_drop_privilege is set. The new process now just calls the
4320 queue-runner function directly.
4326 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4329 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4330 ignored. Though the use of -oP was forcing the writing of a pid file, it
4331 was always written to the default place.
4333 3. If the message "no IP address found for host xxxx" is generated during
4334 incoming verification, it is now followed by identification of the incoming
4335 connection (so you can more easily find what provoked it).
4337 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4339 5. Added some features to "harden" Exim a bit more against certain attacks:
4341 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4342 be put in Local/Makefile. This is like the never_users runtime option,
4343 but it cannot be overridden. The default setting is "root".
4345 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4346 prefix string with which any file named in a -C command line option
4349 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4350 is retained for -C and -D only if the caller of Exim is root. Without
4351 it, the exim user may also use -C and -D and retain privilege.
4353 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4354 command line option is disabled.
4356 6. Macro names set by the -D option must start with an upper case letter, just
4357 like macro names defined in the configuration file.
4359 7. Added "dereference=" facility to LDAP.
4361 8. Two instances of the typo "uknown" in the source files are fixed.
4363 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4364 the Configure-Makefile script screwed up while processing it.
4366 10. Incorporated PCRE 4.4.
4368 11. The SMTP synchronization check was not operating right at the start of an
4369 SMTP session. For example, it could not catch a HELO sent before the client
4370 waited for the greeting. There is now a check for outstanding input at the
4371 point when the greeting is written. Because of the duplex, asynchronous
4372 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4373 way, but not yet received, when the check is performed.
4375 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4376 on TCP/IP sockets, because this apparently causes some broken clients to
4379 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4380 unchanged) from the Cygwin maintainer.
4382 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4383 was supported instead of testing for mbx. Effectively a typo.
4385 15. The spa authenticator server code was not checking that the input it
4386 received was valid base64.
4388 16. The debug output line for the "set" modifier in ACLs was not showing the
4389 name of the variable that was being set.
4391 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4393 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4394 lookup on its own. Something else was needed to trigger the lookup. For
4395 example, a match in host_lookup or the need for a host name in a host list.
4396 Now, if $sender_host_name is referenced and the host name has not yet been
4397 looked up, a lookup is performed. If the lookup fails, the variable remains
4398 empty, and $host_lookup_failed is set to "1".
4400 19. Added "eqi" as a case-independent comparison operator.
4402 20. The saslauthd authentication condition could segfault if neither service
4403 nor realm was specified.
4405 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4406 error message that was logged was misleading, and incoming SMTP
4407 connections were dropped. The message is now more accurate, and temporary
4408 errors are given to SMTP connections.
4410 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4411 (except RSET) with a 421 error, until QUIT is received. However, it was
4412 failing to send a response to QUIT.
4414 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4415 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4416 ACL is now run after the helo_try_verify_hosts code.
4418 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4419 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4420 case-independent, so other case variants are also recognized). Apparently
4421 some systems use these upper case variants.
4423 25. If more than two messages were waiting for the same host, and a transport
4424 filter was specified for the transport, Exim sent two messages over the
4425 same TCP/IP connection, and then failed with "socket operation on non-
4426 socket" when it tried to send the third.
4428 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4430 27. The extern definition of crypt16() in expand.c was not being excluded when
4431 the OS had its own crypt16() function.
4433 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4434 as a preferred synonym for return_size_limit, both as an option and as an
4437 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4439 30. Changed the default configuration ACL to relax the local part checking rule
4440 for addresses that are not in any local domains. For these addresses,
4441 slashes and pipe symbols are allowed within local parts, but the sequence
4442 /../ is explicitly forbidden.
4444 31. SPA server authentication was not clearing the challenge buffer before
4447 32. log_message in a "warn" ACL statement was writing to the reject log as
4448 well as to the main log, which contradicts the documentation and doesn't
4449 seem right (because no rejection is happening). So I have stopped it.
4451 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4452 However, I am unable to do any testing of this.
4454 34. Fixed an infelicity in the appendfile transport. When checking directories
4455 for a mailbox, to see if any needed to be created, it was accidentally
4456 using path names with one or more superfluous leading slashes; tracing
4457 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4459 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4460 contents are added to the log line that is written for every discarded
4461 recipient. (Previously a log_message setting was ignored.)
4463 36. The ${quote: operator now quotes the string if it is empty.
4465 37. The install script runs exim in order to find its version number. If for
4466 some reason other than non-existence or emptiness, which it checks, it
4467 could not run './exim', it was installing it with an empty version number,
4468 i.e. as "exim-". This error state is now caught, and the installation is
4471 38. An argument was missing from the function that creates an error message
4472 when Exim fails to connect to the socket for saslauthd authentication.
4473 This could cause Exim to crash, or give a corrupted message.
4475 39. Added isip, isip4, and isip6 to ${if conditions.
4477 40. The ACL variables $acl_xx are now saved with the message, and can be
4478 accessed later in routers, transports, and filters.
4480 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4481 strings in the file are not string-expanded.
4483 42. If a MAIL command specified a SIZE value that was too large to fit into an
4484 int variable, the check against message_size_limit failed. Such values are
4485 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4486 day this will have to be increased, but I don't think I want to be around
4487 when emails are that large.
4494 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4495 iconv() is not standard in FreeBSD.
4497 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4498 IPv6 enabled. The observed symptom was a segmentation fault on return from
4499 the function os_common_find_running_interfaces() in src/os.c.
4501 3. In the check_special_case() function in daemon.c I had used "errno" as an
4502 argument name, which causes warnings on some systems. This was basically a
4503 typo, since it was named "eno" in the comments!
4505 4. The code that waits for the clock to tick (at a resolution of some fraction
4506 of a second) so as to ensure message-id uniqueness was always waiting for
4507 at least one whole tick, when it could have waited for less. [This is
4508 almost certainly not relevant at current processor speeds, where it is
4509 unlikely to ever wait at all. But we try to future-proof.]
4511 5. The function that sleeps for a time interval that includes fractions of a
4512 second contained a race. It did not block SIGALRM between setting the
4513 timer, and suspending (a couple of lines later). If the interval was short
4514 and the sigsuspend() was delayed until after it had expired, the suspension
4515 never ended. On busy systems this could lead to processes getting stuck for
4518 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4519 process, before it forks any delivery processes. The open lookup caching
4520 mechanism meant that the open file or database connection was passed into
4521 the delivery process. The problem was that delivery processes always tidy
4522 up cached lookup data. This could cause a problem for the next delivery
4523 process started by the queue runner, because the external queue runner
4524 process does not know about the closure. So the next delivery process
4525 still has data in the lookup cache. In the case of a file lookup, there was
4526 no problem because closing a file descriptor in a subprocess doesn't affect
4527 the parent. However, if the lookup was caching a connection to a database,
4528 the connection was closed, and the second delivery process was likely to
4529 see errors such as "PGSQL: query failed: server closed the connection
4530 unexpectedly". The problem has been fixed by closing all cached lookups
4531 in a queue runner before running a delivery process.
4533 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4534 seem to have the "const" qualifier which it has on other OS. I've
4537 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4538 *of the same type* (client or server) with the same public name that an
4539 error should be diagnosed.
4541 9. When Exim looked up a host name for an IP address, but failed to find the
4542 original IP address when looking up the host name (a safety check), it
4543 output the message "<ip address> does not match any IP for NULL", which was
4544 confusing, to say the least. The bug was that the host name should have
4545 appeared instead of "NULL".
4547 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4548 user that is built into the binary, and the -C or -D options is used, root
4549 privilege is dropped before the configuration file is read. In addition,
4550 logging is switched to stderr instead of the normal log files. If the
4551 configuration then re-defines the Exim user, the unprivileged environment
4552 is probably not what is expected, so Exim logs a panic warning message (but
4555 However, if deliver_drop_privilege is set, the unprivileged state may well
4556 be exactly what is intended, so the warning has been cut out in that case,
4557 and Exim is allowed to try to write to its normal log files.
4563 1. smtp_return_error_details was not giving details for temporary sender
4564 or receiver verification errors.
4566 2. Diagnose a configuration error if two authenticators have the same public
4569 3. Exim used not to create the message log file for a message until the first
4570 delivery attempt. This could be confusing when incoming messages were held
4571 for policy or load reasons. The message log file is now created at the time
4572 the message is received, and an initial "Received" line is written to it.
4574 4. The automatically generated man page for command line options had a minor
4575 bug that caused no ill effects; however, a more serious problem was that
4576 the procedure for building the man page automatically didn't always
4577 operate. Consequently, release 4.20 contains an out-of-date version. This
4578 shouldn't happen again.
4580 5. When building Exim with embedded Perl support, the script that builds the
4581 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4582 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4584 6. The freeze_tell option was not being used for messages that were frozen on
4585 arrival, either by an ACL or by local_scan().
4587 7. Added the smtp_incomplete_transaction log selector.
4589 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4590 was accepting AUTH without a new EHLO.
4592 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4593 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4596 10. Logging of TCP/IP connections (when configured) now happens in the main
4597 daemon process instead of the child process, so that the TCP/IP connection
4598 count is more accurate (but it can never be perfect).
4600 11. The use of "drop" in a nested ACL was not being handled correctly in the
4601 outer ACL. Now, if condition failure induced by the nested "drop" causes
4602 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4603 or "require"), the connection is dropped.
4605 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4606 that yield "discard" can now be used with an "accept" or a "discard" verb,
4607 but an error is generated for any others (because I can't see a useful way
4608 to define what should happen).
4610 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4611 are now processed in the same way as lines in the Exim configuration file.
4612 In particular, continuation lines are supported.
4614 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4616 15. Added -ti meaning -t -i.
4618 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4619 domains, and warn by logging if others are found.
4621 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4622 returned from GIFCONF when Exim is trying to find the list of interfaces on
4623 a host. The code in os.c has been modified to copy each ifreq to an aligned
4624 structure in all cases.
4626 Also, in some cases, the returned ifreq's were being copied to a 'struct
4627 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4628 means the last couple of bytes of an IPv6 address could be chopped if the
4629 ifreq contained only a normal sockaddr (14 bytes storage).
4631 18. Named domain lists were not supported in the hosts_treat_as_local option.
4632 An entry such as +xxxx was not recognized, and was treated as a literal
4635 19. Ensure that header lines added by a DATA ACL are included in the reject log
4636 if the ACL subsequently rejects the message.
4638 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4639 MD5 (which is deprecated).
4641 21. When testing a filter file using -bf, Exim was writing a message when it
4642 took the sender from a "From " line in the message, but it was not doing so
4643 when it took $return_path from a Return-Path: header line. It now does.
4645 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4646 with a valid header line field name (a series of printing characters
4647 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4649 23. Changed "disc" in the source to "disk" to conform to the documentation and
4650 the book and for uniformity.
4652 24. Ignore Sendmail's -Ooption=value command line item.
4654 25. When execve() failed while trying to run a command in a pipe transport,
4655 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4656 could be confused with a return value of 69 from the command itself. This
4657 has been changed to 127, the value the shell returns if it is asked to run
4658 a non-existent command. The wording for the related log line suggests a
4659 non-existent command as the problem.
4661 26. If received_header_text expands to an empty string, do not add a Received:
4662 header line to the message. (Well, it adds a token one on the spool, but
4663 marks it "old" so that it doesn't get used or transmitted.)
4665 27. Installed eximstats 1.28 (addition of -nt option).
4667 28. There was no check for failure on the call to getsockname() in the daemon
4668 code. This can fail if there is a shortage of resources on the system, with
4669 ENOMEM, for example. A temporary error is now given on failure.
4671 29. Contrary to the C standard, it seems that in some environments, the
4672 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4673 program. Exim now does this explicitly; it affects the formatting of
4674 timestamps using strftime().
4676 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4677 complaints. I've now initialized all the variables, to avoid this.
4679 32. Header lines added by a system filter were not being "seen" during
4680 transport-time rewrites.
4682 33. The info_callback() function passed to OpenSSL is set up with type void
4683 (*)(SSL *, int, int), as described somewhere. However, when calling the
4684 function (actually a macro) that sets it up, the type void(*)() is
4685 expected. I've put in a cast to prevent warnings from picky compilers.
4687 34. If a DNS black list lookup found a CNAME record, but there were no A
4688 records associated with the domain it pointed at, Exim crashed.
4690 35. If a DNS black list lookup returned more than one A record, Exim ignored
4691 all but the first. It now scans all returned addresses if a particular IP
4692 value is being sought. In this situation, the contents of the
4693 $dnslist_value variable are a list of all the addresses, separated by a
4696 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4697 to accept a host name and all its aliases if the forward lookup for any of
4698 them yielded the IP address of the incoming connection. Now it accepts only
4699 those names whose forward lookup yields the correct IP address. Any other
4700 names are discarded. This closes a loophole whereby a rogue DNS
4701 administrator could create reverse DNS records to break through a
4702 wildcarded host restriction in an ACL.
4704 37. If a user filter or a system filter that ran in a subprocess used any of
4705 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4706 the wrong values were passed to the pipe command ($thisaddress had the
4707 value of $0, $0 had the value of $1, etc). This bug was introduced by
4708 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4710 38. Improved the line breaking for long SMTP error messages from ACLs.
4711 Previously, if there was no break point between 40 and 75 characters, Exim
4712 left the rest of the message alone. Two changes have been made: (a) I've
4713 reduced the minimum length to 35 characters; (b) if it can't find a break
4714 point between 35 and 75 characters, it looks ahead and uses the first one
4715 that it finds. This may give the occasional overlong line, but at least the
4716 remaining text gets split now.
4718 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4719 file descriptors might be low, and that setting 1000 would always raise it.
4720 It turns out that in some environments, the limit is already over 1000 and
4721 that lowering it causes trouble. So now Exim takes care not to decrease it.
4723 40. When delivering a message, the value of $return_path is set to $sender_
4724 address at the start of routing (routers may change the value). By an
4725 oversight, this default was not being set up when an address was tested by
4726 -bt or -bv, which affected the outcome if any router or filter referred to
4729 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4730 log only when "message" or "log_message" are set. However, if one of the
4731 conditions was an address verification, or a call to a nested ACL, the
4732 messages generated by the underlying test were being passed through. This
4733 no longer happens. The underlying message is available in $acl_verify_
4734 message for both "message" and "log_message" expansions, so it can be
4735 passed through if needed.
4737 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4738 new expansion $bh_ to give the encoded byte string without charset
4739 translation. Translation happens only if iconv() is available; HAVE_ICONV
4740 indicates this at build time. HEADERS_CHARSET gives the charset to
4741 translate to; headers_charset can change it in the configuration, and
4742 "headers charset" can change it in an individual filter file.
4744 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4745 that creates RFC 2047 encoded "words" labels them as that charset instead
4746 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4747 expansion operator; (ii) when Exim creates a From: line for a local
4748 message; (iii) when a header line is rewritten to include a "phrase" part.
4750 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4751 buggy, causing it to skip the first lines of messages whose message ID
4752 ended in 'D'. This would not have bitten before Exim release 4.14, because
4753 message IDs were unlikely to end in 'D' before then. The effect was to have
4754 incorrect size information for certain domains.
4756 45. #include "config.h" was missing at the start of the crypt16.c module. This
4757 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4760 46. If there was a timeout during a "random" callout check, Exim treated it as
4761 a failure of the random address, and carried on sending RSET and the real
4762 address. If the delay was just some slowness somewhere, the response to the
4763 original RCPT would be taken as a response to RSET and so on, causing
4764 mayhem of various kinds.
4766 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4767 when I implemented it. It didn't allow for the fact that some option values
4768 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4769 the right test for positive values.
4771 48. Domain names in DNS records are case-independent. Exim always looks them up
4772 in lower case. Some resolvers return domain names in exactly the case they
4773 appear in the zone file, that is, they may contain uppercase letters. Not
4774 all resolvers do this - some return always lower case. Exim was treating a
4775 change of case by a resolver as a change of domain, similar to a widening
4776 of a domain abbreviation. This triggered its re-routing code and so it was
4777 trying to route what was effectively the same domain again. This normally
4778 caused routing to fail (because the router wouldn't handle the domain
4779 twice). Now Exim checks for this case specially, and just changes the
4780 casing of the domain that it ultimately uses when it transmits the message
4783 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4786 50. If a filter generated a file delivery with a non-absolute name (possible if
4787 no home directory exists for the router), the forbid_file option was not
4790 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4791 the existing equality matching.
4793 52. Exim was using ints instead of ino_t variables in some places where it was
4794 dealing with inode numbers.
4796 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4797 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4798 called TMPDIR, and if it finds it is different, it changes its value.
4800 54. The smtp_printf() function is now made available to local_scan() so
4801 additional output lines can be written before returning. There is also an
4802 smtp_fflush() function to enable the detection of a dropped connection.
4803 The variables smtp_input and smtp_batched_input are exported to
4806 55. Changed the default runtime configuration: the message "Unknown user"
4807 has been removed from the ACL, and instead placed on the localuser router,
4808 using the cannot_route_message feature. This means that any verification
4809 failures that generate their own messages won't get overridden. Similarly,
4810 the "Unrouteable address" message that was in the ACL for unverifiable
4811 relay addresses has also been removed.
4813 56. Added hosts_avoid_esmtp to the smtp transport.
4815 57. The exicyclog script was not checking for the esoteric option
4816 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4817 will work only if exicyclog is run under the appropriate euid.
4819 58. Following a discussion on the list, the rules by which Exim recognises line
4820 endings on incoming messages have been changed. The -dropcr and drop_cr
4821 options are now no-ops, retained only for backwards compatibility. The
4822 following line terminators are recognized: LF CRLF CR. However, special
4823 processing applies to CR:
4825 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4826 nor a local message in the state where . is a terminator.
4828 (ii) If a bare CR is encountered in a header line, an extra space is added
4829 after the line terminator so as not to end the header. The reasoning
4830 behind this is that bare CRs in header lines are most likely either
4831 to be mistakes, or people trying to play silly games.
4833 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4834 was being incorrectly given as 18 bytes larger than it should have been.
4835 This is a VOB (very old bug).
4837 60. This may never have affected anything current, but just in case it has:
4838 When the local host is found other than at the start of a list of hosts,
4839 the local host, those with the same MX, and any that follow, are discarded.
4840 When the list in question was part of a longer list of hosts, the following
4841 hosts (not currently being processed) were also being discarded. This no
4842 longer happens. I'm not sure if this situation could ever has previously
4845 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4848 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4849 header. This is something that is recommended in a new Internet Draft, and
4850 is something that is documented as being done by Sendmail. There are two
4851 possible values. For messages generated by the autoreply transport, Exim
4854 Auto-Submitted: auto-replied
4856 whereas for all other generated messages (e.g. bounces) it adds
4858 Auto-Submitted: auto-generated
4860 63. The "personal" condition in filters now includes a test for the
4861 Auto-Submitted: header. If it contains the string "auto-" the message it
4862 not considered personal.
4864 64. Added rcpt_include_affixes as a generic transport option.
4866 65. Added queue_only_override (default true).
4868 66. Added the syslog_duplication option.
4870 67. If what should have been the first header line of a message consisted of
4871 a space followed by a colon, Exim was mis-interpreting it as a header line.
4872 It isn't of course - it is syntactically invalid and should therefore be
4873 treated as the start of the message body. The misbehaviour could have
4874 caused a number of strange effects, including loss of data in subsequent
4875 header lines, and spool format errors.
4877 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4878 client host had authenticated. This control can now be exercised by an ACL
4879 for more flexibility.
4881 69. By default, callouts do not happen when testing with -bh. There is now a
4882 variant, -bhc, which does actually run the callout code, including
4883 consulting and updating the callout cache.
4885 70. Added support for saslauthd authentication, courtesy of Alexander
4888 71. If statvfs() failed on the spool or log directories while checking their
4889 size for availability, Exim confusingly gave the error "space shortage".
4890 Furthermore, in debugging mode it crashed with a floating point exception.
4891 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4892 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4893 serious problem, Exim now writes to the main and panic logs when this
4894 happens, with details of the failure. It then refuses to accept the
4895 incoming message, giving the message "spool directory problem" or "log
4896 directory problem" with a 421 code for SMTP messages.
4898 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4899 0, 1, and 2 exist, because some OS complain for execs without them (see
4900 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4901 descriptors. However, the code omitted to check that the open succeeded,
4902 causing mysterious errors if for some reason the permissions on /dev/null
4903 got screwed. Now Exim writes a message to the main and panic logs, and
4904 bombs out if it can't open /dev/null.
4906 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4907 interact so that it is all more flexible. It is supposed to remain
4908 backwards compatible. Also added extra_local_interfaces.
4910 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4911 to bomb out with an assertion failure - to the client this appears as a
4912 connection drop. This problem occurs in the part of the code that was taken
4913 from the Samba project. Fortunately, the assertion is in a very simple
4914 function, so I have fixed this by reproducing the function inline in the
4915 one place where it is called, and arranging for authentication to fail
4916 instead of killing the process with assert().
4918 75. The SPA client code was not working when the server requested OEM rather
4919 than Unicode encoding.
4921 76. Added code to make require_files with a specific uid setting more usable in
4922 the case where statting the file as root fails - usually a non-root-mounted
4923 NFS file system. When this happens and the failure is EACCES, Exim now
4924 forks a subprocess and does the per-uid checking as the relevant uid.
4926 77. Added process_log_path.
4928 78. If log_file_path was not explicitly set, a setting of check_log_space or
4929 check_log_inodes was ignored.
4931 79. If a space check for the spool or log partitions fails, the incident is now
4932 logged. Of course, in the latter case the data may get lost...
4934 80. Added the %p formatting code to string_format() so that it can be used to
4935 print addresses in debug_print(). Adjusted all the address printing in the
4936 debugging in store.c to use %p rather than %d.
4938 81. There was a concern that a line of code in smtp_in.c could overflow a
4939 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4940 initially expressed, the concern was not well-founded, because trailing
4941 spaces are removed early. However, if the trailing spaces were followed by
4942 a NULL, they did not get removed, so the overflow was possible. Two fixes
4945 (a) I re-wrote the offending code in a cleaner fashion.
4946 (b) If an incoming SMTP command contains a NULL character, it is rejected
4949 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4950 runs initgroups(), so that if the Exim user is in any additional groups,
4951 they will be used during message reception.
4957 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob