1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.287 2006/02/08 14:28:51 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 The code for finding all the local interface addresses on a FreeBSD
10 system running IPv6 was broken. This may well have applied to all BSD
11 systems, as well as to others that have similar system calls. The broken
12 code found IPv4 interfaces correctly, but gave incorrect values for the
13 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
14 that it would not match correctly against @[] and not recognize the IPv6
17 PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
20 PH/03 Added disable_ipv6.
22 PH/04 Changed $reply_address to use the raw form of the headers instead of the
23 decoded form, because it is most often used to construct To: headers
24 lines in autoreplies, and the decoded form may well be syntactically
25 invalid. However, $reply_address has leading white space removed, and all
26 newlines turned into spaces so that the autoreply transport does not
29 PH/05 If group was specified without a user on a router, and no group or user
30 was specified on a transport, the group from the router was ignored.
32 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
33 for visible compile-time settings that can be used to change these
34 numbers, for those that want even more. Backwards compatibility with old
35 spool files has been maintained. However, going back to a previous Exim
36 release will lost any variables that are in spool files.
38 PH/07 Two small changes when running in the test harness: increase delay when
39 passing a TCP/IP connection to a new process, in case the original
40 process has to generate a bounce, and remove special handling of
41 127.0.0.2 (sic), which is no longer necessary.
43 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
44 be the same on different OS.
46 PH/09 Moved a debug statement in filter processing to avoid a race problem when
49 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
50 whether --show-vars was specified or not
52 JJ/02 exipick: Added support for new ACL variable spool format introduced
55 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
56 syntactically invalid From: or Reply-to: line, and a filter used this to
57 generate an autoreply, and therefore failed to obtain an address for the
58 autoreply, Exim could try to deliver to a non-existent relative file
59 name, causing unrelated and misleading errors. What now happens is that
60 it logs this as a hard delivery error, but does not attempt to create a
63 PH/11 The exinext utility has a -C option for testing purposes, but although
64 the given file was scanned by exinext itself; it wasn't being passed on
67 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
68 an end-of-file indication when reading a command response.
70 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
71 compiled. In many other places in Exim, IPv6 addresses are always
72 recognized, so I have changed this. It also means that IPv4 domain
73 literals of the form [IPV4:n.n.n.n] are now always recognized.
75 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
76 used if the router is not running as root, for example, when verifying at
77 ACL time, or when using -bh. The debugging output from this situation was
78 non-existent - all you got was a failure to exec. I have made two
81 (a) Failures to set uid/gid, the current directory, or a process leader
82 in a subprocess such as that created by queryprogram now generate
83 suitable debugging ouput when -d is set.
85 (b) The queryprogram router detects when it is not running as root,
86 outputs suitable debugging information if -d is set, and then runs
87 the subprocess without attempting to change uid/gid.
89 PH/15 Minor change to Makefile for building test_host (undocumented testing
92 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
93 additional section of a DNS packet that returns MX or SRV records.
94 Instead, it always explicitly searches for A/AAAA records. This avoids
95 major problems that occur when a DNS server includes only records of one
96 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
97 fixed another bug: if SRV records were looked up and the corresponding
98 address records were *not* found in the additional section, the port
99 values from the SRV records were lost.
101 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
102 using the correct key (the original address) when searching the retry
103 rules in order to find which one to use for generating the retry hint.
110 PH/01 Two changes to the default runtime configuration:
112 (1) Move the checks for relay_from_hosts and authenticated clients from
113 after to before the (commented out) DNS black list checks.
115 (2) Add control=submission to the relay_from_hosts and authenticated
116 clients checks, on the grounds that messages accepted by these
117 statements are most likely to be submissions.
119 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
121 (1) Generate an error if the third argument for the ${prvs expansion is
124 (2) Treat a missing third argument of ${prvscheck as if it were an empty
127 (3) Reset the variables that are obtained from the first argument of
128 ${prvscheck and used in the second argument before leaving the code,
129 because their memory is reclaimed, so using them afterwards may do
132 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
133 one (it's much easier than Tom thought :-).
135 (5) Because of (4), we can now allow for the use of $prvscheck_result
136 inside the third argument.
138 PH/03 For some reason, the default setting of PATH when running a command from
139 a pipe transport was just "/usr/bin". I have changed it to
142 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
143 anything to be listed in the output from -bV.
145 PH/05 When a filter generated an autoreply, the entire To: header line was
146 quoted in the delivery log line, like this:
148 => >A.N.Other <ano@some.domain> <original@ddress> ...
150 This has been changed so that it extracts the operative address. There
151 may be more than one such address. If so, they are comma-separated, like
154 => >ano@some.domain,ona@other.domain <original@ddress> ...
156 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
157 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
158 not being looked up in the rDNS to get a host name, Exim was showing the
159 IP address twice in Received: lines, even though the IP addresses were
160 identical. For example:
162 Received: from [1.2.3.4] (helo=[1.2.3.4])
164 However, if the real host name was known, it was omitting the HELO data
165 if it matched the actual IP address. This has been tidied up so that it
166 doesn't show the same IP address twice.
168 PH/07 When both +timestamp and +memory debugging was on, the value given by
169 $tod_xxx expansions could be wrong, because the tod_stamp() function was
170 called by the debug printing, thereby overwriting the timestamp buffer.
171 Debugging no longer uses the tod_stamp() function when +timestamp is set.
173 PH/08 When the original message was included in an autoreply transport, it
174 always said "this is a copy of the message, including all the headers",
175 even if body_only or headers_only was set. It now gives an appropriate
178 PH/09 Applied a patch from the Sieve maintainer which:
180 o fixes some comments
181 o adds the (disabled) notify extension core
182 o adds some debug output for the result of if/elsif tests
183 o points to the current vacation draft in the documentation
184 and documents the missing references header update
188 o fixes a bug in processing the envelope test (when testing
189 multiple envelope elements, the last element determinted the
192 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
193 Electronic Mail") by including:
195 Auto-submitted: auto-generated
197 in the messages that it generates (bounce messages and others, such as
198 warnings). In the case of bounce messages for non-SMTP mesages, there was
199 also a typo: it was using "Auto_submitted" (underscore instead of
200 hyphen). Since every message generated by Exim is necessarily in response
201 to another message, thes have all been changed to:
203 Auto-Submitted: auto-replied
205 in accordance with these statements in the RFC:
207 The auto-replied keyword:
209 - SHOULD be used on messages sent in direct response to another
210 message by an automatic process,
212 - MUST NOT be used on manually-generated messages,
214 - MAY be used on Delivery Status Notifications (DSNs) and Message
215 Disposition Notifications (MDNs),
217 - MUST NOT be used on messages generated by automatic or periodic
218 processes, except for messages which are automatic responses to
221 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
222 to the default Received: header definition.
224 PH/12 Added log selector acl_warn_skipped (default on).
226 PH/13 After a successful wildlsearch lookup, discard the values of numeric
227 variables because (a) they are in the wrong storage pool and (b) even if
228 they were copied, it wouldn't work properly because of the caching.
230 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
231 checking when decoding. Apparently there are clients that generate
232 overlong encoded strings. Why am I not surprised?
234 PH/15 If the first argument of "${if match_address" was not empty, but did not
235 contain an "@" character, Exim crashed. Now it writes a panic log message
236 and treats the condition as false.
238 PH/16 In autoreply, treat an empty string for "once" the same as unset.
240 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
241 extension "envelope-auth". The code is finished and in agreement with
242 other implementations, but there is no documentation so far and in fact,
243 nobody wrote the draft yet. This extension is currently #undef'ed, thus
244 not changing the active code.
246 Print executed "if" and "elsif" statements when debugging is used. This
247 helps a great deal to understand what a filter does.
249 Document more things not specified clearly in RFC3028. I had all this
250 sorted out, when out of a sudden new issues came to my mind. Oops."
252 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
255 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
256 canonical form (as documented). However, after a host name lookup from
257 the IP address, check_host() was doing a simple string comparison with
258 addresses acquired from the DNS when checking that the found name did
259 have the original IP as one of its addresses. Since any found IPv6
260 addresses are likely to be in abbreviated form, the comparison could
261 fail. Luckily, there already exists a function for doing the comparison
262 by converting both addresses to binary, so now that is used instead of
265 PH/20 There was another similar case to PH/19, when a complete host name was
266 given in a host list; looking up its IP address could give an abbreviated
267 form, whereas the current host's name might or might not be abbreviated.
268 The same fix has been applied.
274 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
275 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
278 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
279 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
281 PH/03 Typo: missing ".o" in src/pcre/Makefile.
283 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
284 header line, restrict the check to what is listed in RFCs 2369 and 2929.
285 Also, for "Auto-Submitted", treat anything other than "no" as
286 non-personal, in accordance with RFC 3834. (Previously it treated
287 anything starting "auto-" as non-personal.)
289 TF/01 The control=submission/name=... option had a problem with syntax
290 errors if the name included a slash character. The /name= option
291 now slurps the rest of the string, so it can include any characters
292 but it must come last in the list of options (after /sender_retain
295 PH/05 Some modifications to the interface to the fake nameserver for the new
303 TK/01 Added the "success_on_redirect" address verification option. See
304 NewStuff for rationale and an example.
306 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
308 PH/02 Patch to exigrep to allow it to work on syslog lines.
310 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
311 fread() to skip over the body file's header line, because in Cygwin the
312 header line is locked and is inaccessible.
314 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
315 co-exist for some time) to make it clear that it is the Exim ID that is
316 referenced, not the Message-ID: header line.
318 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
319 string_format() function, because snprintf() does not exist on all
322 PH/06 The use of forbid_filter_existstest now also locks out the use of the
323 ${stat: expansion item.
325 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
326 protocol synchronization error", to keep the pedants happy.
328 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
329 well as for IRIX systems, when gcc is being used. See the host.c source
332 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
334 PH/10 Named domain lists were not working if used in a queue_smtp_domains
337 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
338 transport and to the smtp transport in LMTP mode.
340 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
342 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
343 run a filter in a subprocess. This could lead to confusion in subsequent
344 lookups in the parent process. There should also be a search_tidyup() at
345 the end of the subprocess.
347 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
348 only if the host matched helo_try_verify_hosts, which caused the
349 verification to occur when the EHLO/HELO command was issued. The ACL just
350 tested the remembered result. Now, if a previous verification attempt has
351 not happened, "verify = helo" does it there and then.
353 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
355 TK/03 Fix log output including CR from clamd.
357 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
358 exist provoked a memory error which could cause a segfault.
360 PH/15 Installed PCRE 6.2
362 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
364 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
365 of the problem. Specifically, suggested +O2 rather than +O1 for the
368 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
370 PH/20 If a delivery was routed to a non-standard port by means of an SRV
371 record, the port was not correctly logged when the outgoing_port log
372 selector was set (it logged the transort's default port).
374 PH/21 Added support for host-specific ports to manualroute, queryprogram,
375 fallback_hosts, and "hosts" in the smtp transport.
377 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
378 host errors such as "Connection refused".
380 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
381 authentication with radiusclient 0.4.9:
383 - Error returned from rc_read_config was caught wrongly
384 - Username/password not passed on to radius server due to wrong length.
386 The presumption is that some radiusclient API changes for 4.51/PH/17
387 were not taken care of correctly. The code is still untested by me (my
388 Linux distribution still has 0.3.2 of radiusclient), but it was
389 contributed by a Radius user.
391 PH/24 When doing a callout, the value of $domain wasn't set correctly when
392 expanding the "port" option of the smtp transport.
394 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
395 while reading a MIME header. Thanks to Tom Hughes for a patch.
397 PH/24 Include config.h inside local_scan.h so that configuration settings are
400 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
401 that in an ACL for RCPT (for example), you can examine exactly what was
404 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
405 commands, but it was not correctly comparing the address with the actual
406 client host address. Thus, it would show the EHLO address in Received:
407 header lines when this was not necessary.
409 PH/27 Added the % operator to ${eval:}.
411 PH/28 Exim tries to create and chdir to its spool directory when it starts;
412 it should be ignoring failures (because with -C, for example, it has lost
413 privilege). It wasn't ignoring creation failures other than "already
416 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
419 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
420 because an input line was too long, either on its own, or by virtue of
421 too many continuations, the temporary file was not being removed, and the
422 return code was incorrect.
424 PH/31 Missing "BOOL" in function definition in filtertest.c.
426 PH/32 Applied Sieve patches from the maintainer.
428 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
430 PH/33 Added "verify = not_blind".
432 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
433 Local/Makefile (with some defaults set). These are used in built scripts
434 such as exicyclog, but they have never been used in the exim_install
435 script (though there are many overriding facilities there). I have
436 arranged that the exim_install script now takes note of these two
439 PH/35 Installed configuration files for Dragonfly.
441 PH/36 When a locally submitted message by a trusted user did not contain a
442 From: header, and the sender address was obtained from -f or from an SMTP
443 MAIL command, and the trusted user did not use -F to supply a sender
444 name, $originator_name was incorrectly used when constructing a From:
445 header. Furthermore, $originator_name was used for submission mode
446 messages from external hosts without From: headers in a similar way,
447 which is clearly wrong.
449 PH/37 Added control=suppress_local_fixups.
451 PH/38 When log_selector = +received_sender was set, and the addition of the
452 sender made the log line's construction buffer exactly full, or one byte
453 less than full, an overflow happened when the terminating "\n" was
456 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
457 when the result of a list match is failure because a DNS lookup failed.
459 PH/40 RM_COMMAND is now used in the building process.
461 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
462 the "build-* directories that it finds.
464 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
465 address in a domain literal was a prefix of an interface address.
467 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
468 when verifying a sender address, unless rewrite_headers is false.
470 PH/44 Wrote a long comment about why errors_to addresses are verified as
471 recipients, not senders.
473 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
474 the ratelimit ACL was added.
476 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
478 PH/46 Added extra information about PostgreSQL errors to the error string.
480 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
481 avoiding the need to install special zones in a real server. This is
482 backwards compatible; if it can't find the fake resolver, it drops back.
483 Thus, both old and new test suites can be run.
485 TF/02 Added util/ratelimit.pl
487 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
488 clock is set back in time.
490 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
491 Candler <B.Candler@pobox.com>.
493 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
494 OK for addresses that are the result of redirections.
496 PH/48 A number of further additions for the benefit of the new test suite,
497 including a fake gethostbyname() that interfaces to the fake DNS resolver
500 TF/06 The fix for widen_domains has also been applied to qualify_single and
501 search_parents which are the other dnslookup options that can cause
504 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
507 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
509 TF/07 Exim produced the error message "an SRV record indicated no SMTP
510 service" if it encountered an MX record with an empty target hostname.
511 The message is now "an MX or SRV record indicated no SMTP service".
513 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
514 if the DNS of the sending site is misconfigured. This is quite a
515 common situation. This change restores the behaviour of treating a
516 helo verification defer as a failure.
518 PH/51 If self=fail was set on a router, the bounce message did not include the
519 actual error message.
525 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
527 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
528 command itself ended in error, the underlying message about the transport
529 filter timeout was being overwritten with the pipe command error. Now the
530 underlying error message should be appended to the second error message.
532 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
534 PH/02 Reduce the amount of output that "make" produces by default. Full output
535 can still be requested.
537 PH/03 The warning log line about a condition test deferring for a "warn" verb
538 was being output only once per connection, rather than after each
539 occurrence (because it was using the same function as for successful
540 "warn" verbs). This seems wrong, so I have changed it.
542 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
543 it should not have, which might have caused a crash in the right
544 circumstances, but probably never did.
546 PH/04 Installed a modified version of Tony Finch's patch to make submission
547 mode fix the return path as well as the Sender: header line, and to
548 add a /name= option so that you can make the user's friendly name appear
551 TF/03 Added the control = fakedefer ACL modifier.
553 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
554 Mark Lowes for thorough testing.
556 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
558 TK/03 Merged latest SRS patch from Miles Wilton.
560 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
561 with the definition in sysexits.h (which is #included earlier).
562 Fortunately, Exim does not actually use EX_OK. The code used to try to
563 preserve the sysexits.h value, by assumimg that macro definitions were
564 scanned for macro replacements. I have been disabused of this notion,
565 so now the code just undefines EX_OK before #including unistd.h.
567 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
568 in the smtp transport. When a block could not be written in a single
569 write() function, the timeout was being re-applied to each part-write.
570 This seems wrong - if the receiver was accepting one byte at a time it
571 would take for ever. The timeout is now adjusted when this happens. It
572 doesn't have to be particularly precise.
574 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
575 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
577 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
578 without a domain if the check to <postmaster@domain> fails.
580 SC/01 Eximstats: added -xls and the ability to specify output files
581 (patch written by Frank Heydlauf).
583 SC/02 Eximstats: use FileHandles for outputing results.
585 SC/03 Eximstats: allow any combination of xls, txt, and html output.
587 SC/04 Eximstats: fixed display of large numbers with -nvr option
589 SC/05 Eximstats: fixed merging of reports with empty tables.
591 SC/06 Eximstats: added the -include_original_destination flag
593 SC/07 Eximstats: removed tabs and trailing whitespace.
595 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
597 TK/06 MBOX spool code: Add real "From " MBOX separator line
598 so the .eml file is really in mbox format (even though
599 most programs do not really care). Patch from Alex Miller.
601 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
602 The latter is generated from $received_to and is only set if the
603 message has one envelope recipient. SA can use these headers,
604 obviously out-of-the-box. Patch from Alex Miller.
606 PH/08 The ${def test on a variable was returning false if the variable's
607 value was "0", contrary to what the specification has always said!
608 The result should be true unless the variable is empty.
610 PH/09 The syntax error of a character other than { following "${if
611 def:variable_name" (after optional whitespace) was not being diagnosed.
612 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
613 accidental colon was present, for example, could give incorrect results.
615 PH/10 Tidied the code in a number of places where the st_size field of a stat()
616 result is used (not including appendfile, where other changes are about
619 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
620 This involved changing a lot of size variables from int to off_t. It
621 should work with maildirs and everything.
623 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
624 spamd dying while we are connected to it.
626 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
629 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
630 function for BDB 4.3.
632 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
634 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
635 the sources for PCRE have been re-arranged and more files are now
638 PH/15 The code I had for printing potentially long long variables in PH/11
639 above was not the best (it lost precision). The length of off_t variables
640 is now inspected at build time, and an appropriate printing format (%ld
641 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
642 to be "long long int" or "long int". This is needed for the internal
643 formatting function string_vformat().
645 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
646 the configuration file to be ":syslog", then the script "guesses" where
647 the logs files are, rather than using the compiled in default. In our
648 case the guess is not the same as the compiled default, so the script
649 suddenly stopped working when I started to use syslog. The patch checks
650 to see if log_file_path is "". If so, it attempts to read it from exim
651 with no configuration file to get the compiled in version, before it
652 falls back to the previous guessing code."
654 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
655 implementing BATV in an Exim configuration. See NewStuff for the gory
658 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
659 Makefile that are specific to HP-UX.
661 PH/18 If the "use_postmaster" option was set for a recipient callout together
662 with the "random" option, the postmaster address was used as the MAIL
663 FROM address for the random test, but not for the subsequent recipient
664 test. It is now used for both.
666 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
667 patch removes a few documentation additions to RFC 3028, because the
668 latest draft now contains them. It adds the new en;ascii-case comparator
669 and a new error check for 8bit text in MIME parts. Comparator and
670 require names are now matched exactly. I enabled the subaddress
671 extension, but it is not well tested yet (read: it works for me)."
673 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
674 rework some of the code of TK/09 above to avoid the hardwired use of
675 "%lld" and "long long". Replaced the call to snprintf() with a call to
678 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
679 records point to non-existent hosts", "retry timeout exceeded", and
680 "retry time not reached for any host after a long failure period".
682 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
683 experimental DomainKeys support:
685 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
686 (2) On an error such as an illegally used "control", the wrong name for
687 the control was given.
689 These problems did NOT occur unless DomainKeys support was compiled.
691 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
693 PH/24 Added ${if match_ip condition.
695 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
696 databases so that it will be absolutely obvious if a crash occurs in the
697 DB library. This is a regular occurrence (often caused by mis-matched
700 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
701 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
702 picked up on a user's system that detects such things. There doesn't seem
703 to be a gcc warning option for this - only an attribute that has to be
704 put on the function's prototype. It seems that in Fedora Core 4 they have
705 set this on a number of new functions. No doubt there will be more in due
708 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
709 specify a transport. However, if an address that was verified by such a
710 router was the subject of a callout, Exim crashed because it tried to
711 read the rcpt_include_affixes from the non-existent transport. Now it
712 just assumes that the setting of that option is false. This bug was
713 introduced by 4.51/PH/31.
715 PH/28 Changed -d+all to exclude +memory, because that information is very
716 rarely of interest, but it makes the output a lot bigger. People tend to
717 do -d+all out of habit.
719 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
720 code in os-type was giving problems when libc.so lives in lib64, like on
723 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
724 aren't the modern standard, and it seems that some systems' include files
725 don't always have them. Exim was already checking for some of the newer
726 ones like T_AAAA, and defining it itself. I've added checks for all the
727 record types that Exim uses.
729 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
730 not automatically generating a new one, as it is supposed to. This
731 prevented TLS from working. If the file did exist, but contained invalid
732 data, a new version was generated, as expected. It was only the case of a
733 non-existent file that was broken.
735 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
736 with a change in libdomainkeys > 0.64.
738 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
739 from DNS. If the selector record carries the flag, it now has
740 precedence over the domain-wide flag.
742 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
744 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
745 the use of a port name that isn't defined in /etc/services) occurred, the
746 message was deferred as in a normal delivery, and thus remained on the
747 spool, instead of being failed because of the mua_wrapper setting. This
748 is now fixed, and I tidied up some of the mua_wrapper messages at the
751 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
752 the messages in a hash of arrays rather than using individual hashes.
753 This is a bit cleaner and results in dramatic memory savings, albeit
754 at a slight CPU cost.
756 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
757 as requested by Marc Sherman.
759 SC/10 Eximstats: added histograms for user specified patterns as requested
762 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
764 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
765 fopen() in the content-scanning modules that did not already have it.
771 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
772 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
774 TK/02 Fix ACL "control" statement not being available in MIME ACL.
776 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
778 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
779 to test Sieve filters that use "vacation".
781 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
782 that changes the way the GnuTLS parameters are stored in the cache file.
783 The new format can be generated externally. For backward compatibility,
784 if the data in the cache doesn't make sense, Exim assumes it has read an
785 old-format file, and it generates new data and writes a new file. This
786 means that you can't go back to an older release without removing the
789 PH/03 A redirect router that has both "unseen" and "one_time" set does not
790 work if there are any delivery delays because "one_time" forces the
791 parent to be marked "delivered", so its unseen clone is never tried
792 again. For this reason, Exim now forbids the simultaneous setting of
795 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
796 redirected to themselves ("homonym" addresses). Read the long ChangeLog
797 entry if you want to know the details. The fix, however, neglected to
798 consider the case when local delivery batching is involved. The test for
799 "previously delivered" was not happening when checking to see if an
800 address could be batched with a previous (undelivered) one; under
801 certain circumstances this could lead to multiple deliveries to the same
804 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
805 in its include files, and this causes problems building Exim.
807 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
808 header_syntax) but Exim was just ignoring anything given after a slash.
809 In particular, this caused confusion with an attempt to use "verify =
810 reverse_host_lookup/defer_ok". An error is now given when options are
811 supplied for verify items that do not have them. (Maybe reverse_host_
812 lookup should have a defer_ok option, but that's a different point.)
814 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
815 defined by RFC 821) to 2048, because there were problems with some AUTH
816 commands, and RFC 1869 says the size should be increased for extended
817 SMTP commands that take arguments.
819 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
822 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
823 "unknown" error; now it says that the functionality isn't in the binary.
825 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
826 an address' error message when a string expansion fails (syntax or
827 whatever). Otherwise the password may appear in the log. Following change
828 PH/42 below, there is no longer a chance of it appearing in a bounce
831 PH/11 Installed exipick version 20050225.0 from John Jetmore.
833 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
834 of its addresses was ever tried. (Bugzilla bug #2.)
836 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
837 the result incorrectly in the debug output. (It correctly added a newline
838 to what was transported.)
840 TF/01 Added $received_time.
842 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
843 commented out examples of how to interface to a virus scanner and to
844 SpamAssassin. Also added commented examples of av_scanner and
845 spamd_address settings.
847 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
848 and controls are allowed in which ACLs. There were a couple of minor
849 errors. Some of the entries in the conditions table (which is a table of
850 where they are NOT allowed) were getting very unwieldy; rewrote them as a
851 negation of where the condition IS allowed.
853 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
855 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
856 header file does not have a version number, so I've had to invent a new
857 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
858 API. The code is untested by me (my Linux distribution still has 0.3.2 of
859 radiusclient), but it was contributed by a Radius user.
861 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
862 files or directories, for OpenSSL.
864 PH/19 When an Exim process that is running as root has to create an Exim log
865 file, it does so in a subprocess that runs as exim:exim so as to get the
866 ownership right at creation (otherwise, other Exim processes might see
867 the file with the wrong ownership). There was no test for failure of this
868 fork() call, which would lead to the process getting stuck as it waited
869 for a non-existent subprocess. Forks do occasionally fail when resources
870 run out. I reviewed all the other calls to fork(); they all seem to check
873 PH/20 When checking for unexpected SMTP input at connect time (before writing
874 the banner), Exim was not dealing correctly with a non-positive return
875 from the read() function. If the client had disconnected by this time,
876 the result was a log entry for a synchronization error with an empty
877 string after "input=" when read() returned zero. If read() returned -1
878 (an event I could not check), uninitialized data bytes were printed.
879 There were reports of junk text (parts of files, etc) appearing after
882 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
884 PH/22 Added support for macro redefinition, and (re)definition in between
885 driver and ACL definitions.
887 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
888 forgetting to use the resulting value; it was using the unexpanded value.
890 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
891 hadn't been configured. The fix is from Juergen Kreileder, who
892 understands it better than I do:
894 "Here's what I see happening with three configured cyrus_sasl
895 authenticators configured (plain, login, cram-md5):
897 On startup auth_cyrus_sasl_init() gets called for each of these.
898 This means three calls to sasl_listmech() without a specified mech_list.
899 => SASL tests which mechs of all available mechs actually work
900 => three warnings about OTP not working
901 => the returned list contains: plain, login, cram-md5, digest-md5, ...
903 With the patch, sasl_listmech() also gets called three times. But now
904 SASL's mech_list option is set to the server_mech specified in the the
905 authenticator. Or in other words, the answer from sasl_listmech()
906 gets limited to just the mech you're testing for (which is different
908 => the return list contains just 'plain' or 'login', 'cram-md5' or
909 nothing depending on the value of ob->server_mech.
911 I've just tested the patch: Authentication still works fine,
912 unavailable mechs specified in the exim configuration are still
913 caught, and the auth.log warnings about OTP are gone."
915 PH/25 When debugging is enabled, the contents of the command line are added
916 to the debugging output, even when log_selector=+arguments is not
919 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
920 answer is "GNU", and only if the return is "GNU/something" is the answer
923 PH/27 $acl_verify_message is now set immediately after the failure of a
924 verification in an ACL, and so is available in subsequent modifiers. In
925 particular, the message can be preserved by coding like this:
927 warn !verify = sender
928 set acl_m0 = $acl_verify_message
930 Previously, $acl_verify_message was set only while expanding "message"
931 and "log_message" when a very denied access.
933 PH/28 Modified OS/os.c-Linux with
935 -#ifndef OS_LOAD_AVERAGE
936 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
938 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
939 nomenclature these days.)
941 PH/29 Installed patch from the Sieve maintainer that adds the options
942 sieve_useraddress and sieve_subaddress to the redirect router.
944 PH/30 In these circumstances:
945 . Two addresses routed to the same list of hosts;
946 . First host does not offer TLS;
947 . First host accepts first address;
948 . First host gives temporary error to second address;
949 . Second host offers TLS and a TLS session is established;
950 . Second host accepts second address.
951 Exim incorrectly logged both deliveries with the TLS parameters (cipher
952 and peerdn, if requested) that were in fact used only for the second
955 PH/31 When doing a callout as part of verifying an address, Exim was not paying
956 attention to any local part prefix or suffix that was matched by the
957 router that accepted the address. It now behaves in the same way as it
958 does for delivery: the affixes are removed from the local part unless
959 rcpt_include_affixes is set on the transport.
961 PH/32 Add the sender address, as F=<...>, to the log line when logging a
962 timeout during the DATA phase of an incoming message.
964 PH/33 Sieve envelope tests were broken for match types other than :is. I have
965 applied a patch sanctioned by the Sieve maintainer.
967 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
968 the uid or gid is negative. A case of a negative gid caused this to be
969 noticed. The fix allows for either to be negative.
971 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
972 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
975 PH/36 The change PH/12 above was broken. Fixed it.
977 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
978 the grounds that routing the same address twice would always produce the
979 same answer. This might have been true once, but it is certainly no
980 longer true now. Routing a child address may depend on the previous
981 routing that produced that child. Some complicated redirection strategies
982 went wrong when messages had multiple recipients, and made Exim's
983 behaviour dependent on the order in which the addresses were given.
985 I have moved the duplicate checking until after the routing is complete.
986 Exim scans the addresses that are assigned to local and remote
987 transports, and removes any duplicates. This means that more work will be
988 done, as duplicates will always all be routed, but duplicates are
989 presumably rare, so I don't expect this is of any significance.
991 For deliveries to pipes, files, and autoreplies, the duplicate checking
992 still happens during the routing process, since they are not going to be
995 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
996 It corrects a timeout issue with spamd. This is Ian's comment: "The
997 background is that sometimes spamd either never reads data from a
998 connection it has accepted, or it never writes response data. The exiscan
999 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1000 blindly assumes that writes won't block so it may never time out."
1002 PH/39 Allow G after quota size as well as K and M.
1004 PH/40 The value set for $authenticated_id in an authenticator may not contain
1005 binary zeroes or newlines because the value is written to log lines and
1006 to spool files. There was no check on this. Now the value is run through
1007 the string_printing() function so that such characters are converted to
1008 printable escape sequences.
1010 PH/41 $message_linecount is a new variable that contains the total number of
1011 lines in the message. Compare $body_linecount, which is the count for the
1014 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1015 bounce and delay warning messages, except in certain special cases, which
1018 (a) An SMTP error message from a remote host;
1019 (b) A message specified in a :fail: redirection;
1020 (c) A message specified in a "fail" command in a system filter;
1021 (d) A message specified in a FAIL return from the queryprogram router;
1022 (e) A message specified by the cannot_route_message router option.
1024 In these cases only, Exim does include the error details in bounce and
1025 warning messages. There are also a few cases where bland messages such
1026 as "unrouteable address" or "local delivery error" are given.
1028 PH/43 $value is now also set for the "else" part of a ${run expansion.
1030 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1031 being worked on, but at least Exim now implements the latest version to
1034 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1035 process to complete was treated as a delivery failure, a timeout while
1036 writing the message to the pipe was logged, but erroneously treated as a
1037 successful delivery. Such timeouts include transport filter timeouts. For
1038 consistency with the overall process timeout, these timeouts are now
1039 treated as errors, giving rise to delivery failures by default. However,
1040 there is now a new Boolean option for the pipe transport called
1041 timeout_defer, which, if set TRUE, converts the failures into defers for
1042 both kinds of timeout. A transport filter timeout is now identified in
1045 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1046 systems where "make" and "gmake" are different, calling "gmake" at top
1047 level broke things. I've arranged for the value of $(MAKE) to be passed
1048 from the Makefile to this script so that it can call the same version of
1052 A note about Exim versions 4.44 and 4.50
1053 ----------------------------------------
1055 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1056 changes of various kinds. As a consequence, a big documentation update was
1057 needed. This delayed the release for rather longer than seemed good, especially
1058 in the light of a couple of (minor) security issues. Therefore, the changes
1059 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1060 release. So 4.44 and 4.50 are in effect two different branches that both start
1063 I have left the 4.50 change log unchanged; it contains all the changes since
1064 4.43. The change log for 4.44 is below; many of its items are identical to
1065 those for 4.50. This seems to be the most sensible way to preserve the
1066 historical information.
1072 1. Minor wording change to the doc/README.SIEVE file.
1074 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1075 computation of the current number of files was incorrect.
1077 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1078 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1081 4. Give more explanation in the error message when the command for a transport
1082 filter fails to execute.
1084 5. There are several places where Exim runs a non-Exim command in a
1085 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1086 was being done only for the command run by the queryprogram router. It is
1087 now done for all such subprocesses. The other cases are: ${run, transport
1088 filters, and the commands run by the lmtp and pipe transports.
1090 6. Added CONFIGURE_GROUP build-time option.
1092 7. Some older OS have a limit of 256 on the maximum number of file
1093 descriptors. Exim was using setrlimit() to set 1000 as a large value
1094 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1095 systems. I've change it so that if it can't get 1000, it tries for 256.
1097 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1098 was an oversight, and furthermore, ever since the addition of extra
1099 controls (e.g. 4.43/32), the checks on when to allow different forms of
1100 "control" were broken. There should now be diagnostics for all cases when a
1101 control that does not make sense is encountered.
1103 9. Added the /retain_sender option to "control=submission".
1105 10. $recipients is now available in the predata ACL (oversight).
1107 11. Tidy the search cache before the fork to do a delivery from a message
1108 received from the command line. Otherwise the child will trigger a lookup
1109 failure and thereby defer the delivery if it tries to use (for example) a
1110 cached ldap connection that the parent has called unbind on.
1112 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1113 of $address_data from the recipient verification was clobbered by the
1114 sender verification.
1116 13. The value of address_data from a sender verification is now available in
1117 $sender_address_data in subsequent conditions in the ACL statement.
1119 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1121 15. Added a new option "connect=<time>" to callout options, to set a different
1124 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1125 was its contents. (It was OK if the option was not defined at all.)
1127 17. A "Completed" log line is now written for messages that are removed from
1128 the spool by the -Mrm option.
1130 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1131 information about exactly what failed.
1133 19. Added -dd to debug only the daemon process.
1135 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1136 handles timeouts, both on the server side and network timeouts. Renamed the
1137 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1139 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1140 It is now set to "smtps".
1142 22. $host_address is now set to the target address during the checking of
1143 ignore_target_hosts.
1145 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1146 being passed; this would have caused $sender_host_name to have been used if
1147 matching the list had actually called for a host name (not very likely,
1148 since this list is usually IP addresses). A host name is now passed as
1151 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1152 code that specifies a non-restarting handler (typically sigaction() in
1153 modern systems) in an attempt to fix a rare and obscure crash bug.
1155 25. Narrowed the window for a race in the daemon that could cause it to ignore
1156 SIGCHLD signals. This is not a major problem, because they are used only to
1157 wake it up if nothing else does.
1159 26. A malformed maildirsize file could cause Exim to calculate negative values
1160 for the mailbox size or file count. Odd effects could occur as a result.
1161 The maildirsize information is now recalculated if the size or filecount
1164 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1165 support for a long time. Removed HAVE_SYS_VFS_H.
1167 28. Installed the latest version of exipick from John Jetmore.
1169 29. In an address list, if the pattern was not a regular expression, an empty
1170 subject address (from a bounce message) matched only if the pattern was an
1171 empty string. Non-empty patterns were not even tested. This was the wrong
1172 because it is perfectly reasonable to use an empty address as part of a
1173 database query. An empty address is now tested by patterns that are
1174 lookups. However, all the other forms of pattern expect the subject to
1175 contain a local part and a domain, and therefore, for them, an empty
1176 address still always fails if the pattern is not itself empty.
1178 30. Exim went into a mad DNS loop when attempting to do a callout where the
1179 host was specified on an smtp transport, and looking it up yielded more
1180 than one IP address.
1182 31. Re-factored the code for checking spool and log partition space into a
1183 function that finds that data and another that does the check. The former
1184 is then used to implement four new variables: $spool_space, $log_space,
1185 $spool_inodes, and $log_inodes.
1187 32. The RFC2047 encoding function was originally intended for short strings
1188 such as real names; it was not keeping to the 75-character limit for
1189 encoded words that the RFC imposes. It now respects the limit, and
1190 generates multiple encoded words if necessary. To be on the safe side, I
1191 have increased the buffer size for the ${rfc2047: expansion operator from
1194 33. It is now permitted to omit both strings after an "if" condition; if the
1195 condition is true, the result is "true". As before, when the second string
1196 is omitted, a false condition yields an empty string. This makes it less
1197 cumbersome to write custom ACL and router conditions.
1199 34. Failure to deliver a bounce message always caused it to be frozen, even if
1200 there was an errors_to setting on the router. The errors_to setting is now
1203 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1204 canonical form (fully expanded) before being placed in
1205 $sender_host_address.
1207 36. The table in the code that translates DNS record types into text (T_A to
1208 "A" for instance) was missing entries for NS and CNAME. It is just possible
1209 that this could have caused confusion if both these types were looked up
1210 for the same domain, because the text type is used as part of Exim's
1211 per-process caching. But the chance of anyone hitting this buglet seems
1214 37. The dnsdb lookup has been extended in a number of ways.
1216 (1) There is a new type, "zns", which walks up the domain tree until it
1217 finds some nameserver records. It should be used with care.
1219 (2) There is a new type, "mxh", which is like "mx" except that it returns
1220 just the host names, not the priorities.
1222 (3) It is now possible to give a list of domains (or IP addresses) to be
1223 looked up. The behaviour when one of the lookups defers can be
1224 controlled by a keyword.
1226 (4) It is now possible to specify the separator character for use when
1227 multiple records are returned.
1229 38. The dnslists ACL condition has been extended: it is now possible to supply
1230 a list of IP addresses and/or domains to be looked up in a particular DNS
1233 39. Added log_selector=+queue_time_overall.
1235 40. When running the queue in the test harness, wait just a tad after forking a
1236 delivery process, to get repeatability of debugging output.
1238 41. Include certificate and key file names in error message when GnuTLS fails
1239 to set them up, because the GnuTLS error message doesn't include the name
1240 of the failing file when there is a problem reading it.
1242 42. Allow both -bf and -bF in the same test run.
1244 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1246 44. The "Exiscan patch" is now merged into the mainline Exim source.
1248 45. Sometimes the final signoff response after QUIT could fail to get
1249 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1250 < 0 before doing a fflush(). This bug looks as though it goes back to the
1251 introduction of TLS in release 3.20, but "sometimes" must have been rare
1252 because the tests only now provoked it.
1254 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1255 (this can affect the format of dates).
1257 47. exim_tidydb, when checking for the continued existence of a message for
1258 which it has found a message-specific retry record, was not finding
1259 messages that were in split spool directories. Consequently, it was
1260 deleting retry records that should have stayed in existence.
1262 48. Steve fixed some bugs in eximstats.
1264 49. The SPA authentication driver was not abandoning authentication and moving
1265 on to the next authenticator when an expansion was forced to fail,
1266 contradicting the general specification for all authenticators. Instead it
1267 was generating a temporary error. It now behaves as specified.
1269 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1270 (the order specifies the preference for clients). The order is now AES256,
1271 AES128, 3DES, ARCFOUR128.
1273 51. Small patch to Sieve code - explicitly set From: when generating an
1276 52. Exim crashed if a remote delivery caused a very long error message to be
1277 recorded - for instance if somebody sent an entire SpamAssassin report back
1278 as a large number of 550 error lines. This bug was coincidentally fixed by
1279 increasing the size of one of Exim's internal buffers (big_buffer) that
1280 happened as part of the Exiscan merge. However, to be on the safe side, I
1281 have made the code more robust (and fixed the comments that describe what
1284 53. Now that there can be additional text after "Completed" in log lines (if
1285 the queue_time_overall log selector is set), a one-byte patch to exigrep
1286 was needed to allow it to recognize "Completed" as not the last thing in
1289 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1290 patch that reportedly fixes this has been added. I am not expert enough to
1291 create a test for it. This is what the patch creator wrote:
1293 "I found a little strange behaviour of ldap code when working with
1294 Windows 2003 AD Domain, where users was placed in more than one
1295 Organization Units. When I tried to give exim partial DN, the exit code
1296 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1297 But simultaneously result of request was absolutely normal ldap result,
1298 so I produce this patch..."
1300 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1301 REFERENCE, so I have modified the code to exclude the patch when that macro
1304 55. Some experimental protocols are using DNS PTR records for new purposes. The
1305 keys for these records are domain names, not reversed IP addresses. The
1306 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1307 leaves it alone. Component reversal etc. now happens only for IP addresses.
1310 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1312 57. Double the size of the debug message buffer (to 2048) so that more of very
1313 long debug lines gets shown.
1315 58. The exicyclog utility now does better if the number of log files to keep
1316 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1318 59. Two changes related to the smtp_active_hostname option:
1320 (1) $smtp_active_hostname is now available as a variable.
1321 (2) The default for smtp_banner uses $smtp_active_hostname instead
1322 of $primary_hostname.
1324 60. The host_aton() function is supposed to be passed a string that is known
1325 to be a valid IP address. However, in the case of IPv6 addresses, it was
1326 not checking this. This is a hostage to fortune. Exim now panics and dies
1327 if the condition is not met. A case was found where this could be provoked
1328 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1329 components; fortuitously, this particular loophole had already been fixed
1330 by change 4.50/55 above.
1332 If there are any other similar loopholes, the new check in host_aton()
1333 itself should stop them being exploited. The report I received stated that
1334 data on the command line could provoke the exploit when Exim was running as
1335 exim, but did not say which command line option was involved. All I could
1336 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1337 running as the user.
1340 61. There was a buffer overflow vulnerability in the SPA authentication code
1341 (which came originally from the Samba project). I have added a test to the
1342 spa_base64_to_bits() function which I hope fixes it.
1345 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1346 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1348 63. The daemon start-up calls getloadavg() while still root for those OS that
1349 need the first call to be done as root, but it missed one case: when
1350 deliver_queue_load_max is set with deliver_drop_privilege. This is
1351 necessary for the benefit of the queue runner, because there is no re-exec
1352 when deliver_drop_privilege is set.
1354 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1355 This has been fixed.
1357 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1358 was in use, was not putting the data itself into the right store pool;
1359 consequently, it could be overwritten for a subsequent message in the same
1360 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1363 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1365 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1366 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1367 the function were treating the return as a boolean value, which happened to
1368 work because 0=false and not-0=true, but is not correct code.
1370 68. The host_aton() function was not handling scoped IPv6 addresses (those
1371 with, for example, "%eth0" on the end) correctly.
1373 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1374 negated items (that is, ~something) in unsigned ints. Some compilers
1375 apparently mutter when there is no cast.
1377 70. If an address verification called from an ACL failed, and did not produce a
1378 user-specific message (i.e. there was only a "system" message), nothing was
1379 put in $acl_verify_message. In this situation, it now puts the system
1382 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1383 session; change 4.31/43 added the unwanted input to the log line - except
1384 that it did not do this in the start of session case. It now does.
1386 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1387 This is wrong and can cause the other end to generate a synchronization
1388 error if it is another Exim or anything else that does the synchronization
1389 check. A QUIT command is no longer sent after a timeout.
1391 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1392 during host lookups.
1394 74. The defer_ok option of callout verification was not working if it was used
1395 when verifying addresses in header lines, that is, for this case:
1397 verify = header_sender/callout=defer_ok
1399 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1400 those file descriptors could be used for SMTP connections. If anything
1401 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1402 could be sent to the SMTP client, causing chaos. The daemon now opens
1403 stdin, stdout, and stderr to /dev/null when it puts itself into the
1406 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1407 log by default. The user can override this with suitable Perl magic.
1409 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1410 the log message when discard triggers, was not working for the DATA ACL or
1411 for the non-SMTP ACL.
1413 78. Error message wording change in sieve.c.
1415 79. If smtp_accept_max_per_host was set, the number of connections could be
1416 restricted to fewer than expected, because the daemon was trying to set up
1417 a new connection before checking whether the processes handling previous
1418 connections had finished. The check for completed processes is now done
1419 earlier. On busy systems, this bug wouldn't be noticed because something
1420 else would have woken the daemon, and it would have reaped the completed
1423 80. If a message was submitted locally by a user whose login name contained one
1424 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1425 It caused a spool format error. I have fixed the spool reading code. A
1426 related problem was that the "from" clause in the Received: line became
1427 illegal because of the space(s). It is now covered by ${quote_local_part.
1429 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1432 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1434 83. Patch from Sieve maintainer to fix unterminated string problem in
1435 "vacation" handling.
1437 84. Some minor changes to the Linux configuration files to help with other
1438 OS variants using glibc.
1440 85. One more patch for Sieve to update vacation handling to latest spec.
1443 ----------------------------------------------------
1444 See the note above about the 4.44 and 4.50 releases.
1445 ----------------------------------------------------
1451 1. Change 4.43/35 introduced a bug that caused file counts to be
1452 incorrectly computed when quota_filecount was set in an appendfile
1455 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1456 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1459 3. Give more explanation in the error message when the command for a transport
1460 filter fails to execute.
1462 4. There are several places where Exim runs a non-Exim command in a
1463 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1464 was being done only for the command run by the queryprogram router. It is
1465 now done for all such subprocesses. The other cases are: ${run, transport
1466 filters, and the commands run by the lmtp and pipe transports.
1468 5. Some older OS have a limit of 256 on the maximum number of file
1469 descriptors. Exim was using setrlimit() to set 1000 as a large value
1470 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1471 systems. I've change it so that if it can't get 1000, it tries for 256.
1473 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1474 was an oversight, and furthermore, ever since the addition of extra
1475 controls (e.g. 4.43/32), the checks on when to allow different forms of
1476 "control" were broken. There should now be diagnostics for all cases when a
1477 control that does not make sense is encountered.
1479 7. $recipients is now available in the predata ACL (oversight).
1481 8. Tidy the search cache before the fork to do a delivery from a message
1482 received from the command line. Otherwise the child will trigger a lookup
1483 failure and thereby defer the delivery if it tries to use (for example) a
1484 cached ldap connection that the parent has called unbind on.
1486 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1487 of $address_data from the recipient verification was clobbered by the
1488 sender verification.
1490 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1491 was its contents. (It was OK if the option was not defined at all.)
1493 11. A "Completed" log line is now written for messages that are removed from
1494 the spool by the -Mrm option.
1496 12. $host_address is now set to the target address during the checking of
1497 ignore_target_hosts.
1499 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1500 being passed; this would have caused $sender_host_name to have been used if
1501 matching the list had actually called for a host name (not very likely,
1502 since this list is usually IP addresses). A host name is now passed as
1505 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1506 code that specifies a non-restarting handler (typically sigaction() in
1507 modern systems) in an attempt to fix a rare and obscure crash bug.
1509 15. Narrowed the window for a race in the daemon that could cause it to ignore
1510 SIGCHLD signals. This is not a major problem, because they are used only to
1511 wake it up if nothing else does.
1513 16. A malformed maildirsize file could cause Exim to calculate negative values
1514 for the mailbox size or file count. Odd effects could occur as a result.
1515 The maildirsize information is now recalculated if the size or filecount
1518 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1519 support for a long time. Removed HAVE_SYS_VFS_H.
1521 18. Updated exipick to current release from John Jetmore.
1523 19. Allow an empty sender to be matched against a lookup in an address list.
1524 Previously the only cases considered were a regular expression, or an
1527 20. Exim went into a mad DNS lookup loop when doing a callout where the
1528 host was specified on the transport, if the DNS lookup yielded more than
1531 21. The RFC2047 encoding function was originally intended for short strings
1532 such as real names; it was not keeping to the 75-character limit for
1533 encoded words that the RFC imposes. It now respects the limit, and
1534 generates multiple encoded words if necessary. To be on the safe side, I
1535 have increased the buffer size for the ${rfc2047: expansion operator from
1538 22. Failure to deliver a bounce message always caused it to be frozen, even if
1539 there was an errors_to setting on the router. The errors_to setting is now
1542 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1543 canonical form (fully expanded) before being placed in
1544 $sender_host_address.
1546 24. Updated eximstats to version 1.33
1548 25. Include certificate and key file names in error message when GnuTLS fails
1549 to set them up, because the GnuTLS error message doesn't include the name
1550 of the failing file when there is a problem reading it.
1552 26. Expand error message when OpenSSL has problems setting up cert/key files.
1555 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1556 (this can affect the format of dates).
1558 28. exim_tidydb, when checking for the continued existence of a message for
1559 which it has found a message-specific retry record, was not finding
1560 messages that were in split spool directories. Consequently, it was
1561 deleting retry records that should have stayed in existence.
1563 29. eximstats updated to version 1.35
1564 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1565 1.35 - bugfix such that pie charts by volume are generated correctly
1567 30. The SPA authentication driver was not abandoning authentication and moving
1568 on to the next authenticator when an expansion was forced to fail,
1569 contradicting the general specification for all authenticators. Instead it
1570 was generating a temporary error. It now behaves as specified.
1572 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1573 (the order specifies the preference for clients). The order is now AES256,
1574 AES128, 3DES, ARCFOUR128.
1576 31. Small patch to Sieve code - explicitly set From: when generating an
1579 32. Exim crashed if a remote delivery caused a very long error message to be
1580 recorded - for instance if somebody sent an entire SpamAssassin report back
1581 as a large number of 550 error lines. This bug was coincidentally fixed by
1582 increasing the size of one of Exim's internal buffers (big_buffer) that
1583 happened as part of the Exiscan merge. However, to be on the safe side, I
1584 have made the code more robust (and fixed the comments that describe what
1587 33. Some experimental protocols are using DNS PTR records for new purposes. The
1588 keys for these records are domain names, not reversed IP addresses. The
1589 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1590 leaves it alone. Component reversal etc. now happens only for IP addresses.
1593 34. The host_aton() function is supposed to be passed a string that is known
1594 to be a valid IP address. However, in the case of IPv6 addresses, it was
1595 not checking this. This is a hostage to fortune. Exim now panics and dies
1596 if the condition is not met. A case was found where this could be provoked
1597 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1598 components; fortuitously, this particular loophole had already been fixed
1599 by change 4.50/55 or 4.44/33 above.
1601 If there are any other similar loopholes, the new check in host_aton()
1602 itself should stop them being exploited. The report I received stated that
1603 data on the command line could provoke the exploit when Exim was running as
1604 exim, but did not say which command line option was involved. All I could
1605 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1606 running as the user.
1609 35. There was a buffer overflow vulnerability in the SPA authentication code
1610 (which came originally from the Samba project). I have added a test to the
1611 spa_base64_to_bits() function which I hope fixes it.
1614 36. The daemon start-up calls getloadavg() while still root for those OS that
1615 need the first call to be done as root, but it missed one case: when
1616 deliver_queue_load_max is set with deliver_drop_privilege. This is
1617 necessary for the benefit of the queue runner, because there is no re-exec
1618 when deliver_drop_privilege is set.
1620 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1621 was in use, was not putting the data itself into the right store pool;
1622 consequently, it could be overwritten for a subsequent message in the same
1623 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1626 38. Sometimes the final signoff response after QUIT could fail to get
1627 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1628 < 0 before doing a fflush(). This bug looks as though it goes back to the
1629 introduction of TLS in release 3.20, but "sometimes" must have been rare
1630 because the tests only now provoked it.
1636 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1637 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1638 Now it returns an integer. A number of calls were still expecting a T/F
1639 return. Fortuitously, in all cases, the tests worked in OK situations,
1640 which is the norm. However, things would have gone wrong on any write
1641 failures on the smtp file descriptor. This function is used when sending
1642 messages over SMTP and also when doing verify callouts.
1644 2. When Exim is called to do synchronous delivery of a locally submitted
1645 message (the -odf or -odi options), it no longer closes stderr before doing
1648 3. Implemented the mua_wrapper option.
1650 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1652 5. Implemented the functions header_remove(), header_testname(),
1653 header_add_at_position(), and receive_remove_recipient(), and exported them
1656 6. If an ACL "warn" statement specified the addition of headers, Exim already
1657 inserted X-ACL-Warn: at the start if there was no header name. However, it
1658 was not making this test for the second and subsequent header lines if
1659 there were newlines in the string. This meant that an invalid header could
1660 be inserted if Exim was badly configured.
1662 7. Allow an ACL "warn" statement to add header lines at the start or after all
1663 the Received: headers, as well as at the end.
1665 8. Added the rcpt_4xx retry error code.
1667 9. Added postmaster_mailfrom=xxx to callout verification option.
1669 10. Added mailfrom=xxxx to the callout verification option, for verify=
1672 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1673 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1675 12. Inserted some casts to stop certain compilers warning when using pointer
1676 differences as field lengths or precisions in printf-type calls (mostly
1677 affecting debugging statements).
1679 13. Added optional readline() support for -be (dynamically loaded).
1681 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1682 same clock tick as a message's arrival, so that its received time was the
1683 same as the "first fail" time on the retry record, and that message
1684 remained on the queue past the ultimate address timeout, every queue runner
1685 would try a delivery (because it was past the ultimate address timeout) but
1686 after another failure, the ultimate address timeout, which should have then
1687 bounced the address, did not kick in. This was a "< instead of <=" error;
1688 in most cases the first failure would have been in the next clock tick
1689 after the received time, and all would be well.
1691 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1692 being recognized when the domain list was tested by the match_domain
1693 condition in an expansion string.
1695 16. Added the ${str2b64: operator.
1697 17. Exim was always calling setrlimit() to set a large limit for the number of
1698 processes, without checking whether the existing limit was already
1699 adequate. (It did check for the limit on file descriptors.) Furthermore,
1700 errors from getrlimit() and setrlimit() were being ignored. Now they are
1701 logged to the main and panic logs, but Exim does carry on, to try to do its
1702 job under whatever limits there are.
1704 18. Imported PCRE 5.0.
1706 19. Trivial typo in log message " temporarily refused connection" (the leading
1709 20. If the log selector return_path_on_delivery was set and an address was
1710 redirected to /dev/null, the delivery process crashed because it assumed
1711 that a return path would always be set for a "successful" delivery. In this
1712 case, the whole delivery is bypassed as an optimization, and therefore no
1715 21. Internal re-arrangement: the function for sending a challenge and reading
1716 a response while authentication was assuming a zero-terminated challenge
1717 string. It's now changed to take a pointer and a length, to allow for
1718 binary data in such strings.
1720 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1722 23. Exim was not respecting finduser_retries when seeking the login of the
1723 uid under which it was called; it was always trying 10 times. (The default
1724 setting of finduser_retries is zero.) Also, it was sleeping after the final
1725 failure, which is pointless.
1727 24. Implemented tls_on_connect_ports.
1729 25. Implemented acl_smtp_predata.
1731 26. If the domain in control=submission is set empty, Exim assumes that the
1732 authenticated id is a complete email address when it generates From: or
1733 Sender: header lines.
1735 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1736 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1737 chown and chgrp in /bin and hostname in /usr/bin.
1739 28. Exim was keeping the "process log" file open after each use, just as it
1740 does for the main log. This opens the possibility of it remaining open for
1741 long periods when the USR1 signal hits a daemon. Occasional processlog
1742 errors were reported, that could have been caused by this. Anyway, it seems
1743 much more sensible not to leave this file open at all, so that is what now
1746 29. The long-running daemon process does not normally write to the log once it
1747 has entered its main loop, and it closes the log before doing so. This is
1748 so that log files can straightforwardly be renamed and moved. However,
1749 there are a couple of unusual error situations where the daemon does write
1750 log entries, and I had neglected to close the log afterwards.
1752 30. The text of an SMTP error response that was received during a remote
1753 delivery was being truncated at 512 bytes. This is too short for some of
1754 the long messages that one sometimes sees. I've increased the limit to
1757 31. It is now possible to make retry rules that apply only when a message has a
1758 specific sender, in particular, an empty sender.
1760 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1761 it possible to be selective about when SMTP synchronization is enforced.
1763 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1765 32. Implemented hosts_connection_nolog.
1767 33. Added an ACL for QUIT.
1769 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1772 35. Added mailbox_size and mailbox_filecount to appendfile.
1774 36. Added control = no_multiline_responses to ACLs.
1776 37. There was a bug in the logic of the code that waits for the clock to tick
1777 in the case where the clock went backwards by a substantial amount such
1778 that the microsecond fraction of "now" was more than the microsecond
1779 fraction of "then" (but the whole seconds number was less).
1781 38. Added support for the libradius Radius client library this is found on
1782 FreeBSD (previously only the radiusclient library was supported).
1788 1. When certain lookups returned multiple values in the form name=value, the
1789 quoting of the values was not always being done properly. Specifically:
1790 (a) If the value started with a double quote, but contained no whitespace,
1792 (b) If the value contained whitespace other than a space character (i.e.
1793 tabs or newlines or carriage returns) it was not quoted.
1794 This fix has been applied to the mysql and pgsql lookups by writing a
1795 separate quoting function and calling it from the lookup code. The fix
1796 should probably also be applied to nisplus, ibase and oracle lookups, but
1797 since I cannot test any of those, I have not disturbed their existing code.
1799 2. A hit in the callout cache for a specific address caused a log line with no
1800 reason for rejecting RCPT. Now it says "Previous (cached) callout
1801 verification failure".
1803 3. There was an off-by-one bug in the queryprogram router. An over-long
1804 return line was truncated at 256 instead of 255 characters, thereby
1805 overflowing its buffer with the terminating zero. As well as fixing this, I
1806 have increased the buffer size to 1024 (and made a note to document this).
1808 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1809 when Exim is waiting for an SMTP response from a remote server, Exim
1810 restarts its select() call on the socket, thereby resetting its timeout.
1811 This is not a problem when such interrupts are rare. Somebody set up a cron
1812 job to run exiwhat every 2 minutes, which is less than the normal select()
1813 timeout (5 or 10 minutes). This meant that the select() timeout never
1814 kicked in because it was always reset. I have fixed this by comparing the
1815 time when an interrupt arrives with the time at the start of the first call
1816 to select(). If more time than the timeout has elapsed, the interrupt is
1817 treated as a timeout.
1819 5. Some internal re-factoring in preparation for the addition of Sieve
1820 extensions (by MH). In particular, the "personal" test is moved to a
1821 separate function, and given an option for scanning Cc: and Bcc: (which is
1822 not set for Exim filters).
1824 6. When Exim created an email address using the login of the caller as the
1825 local part (e.g. when creating a From: or Sender: header line), it was not
1826 quoting the local part when it contained special characters such as @.
1828 7. Installed new OpenBSD configuration files.
1830 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1831 try to make them clearer.
1833 9. Callout options, other than the timeout value, were being ignored when
1834 verifying sender addresses in header lines. For example, when using
1836 verify = header_sender/callout=no_cache
1838 the cache was (incorrectly) being used.
1840 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1841 only the Cygwin environment.
1843 11. When return_path_on_delivery was set as a log selector, if different remote
1844 addresses in the same message used different return paths and parallel
1845 remote delivery occurred, the wrong values would sometimes be logged.
1846 (Whenever a remote delivery process finished, the return path value from
1847 the most recently started remote delivery process was logged.)
1849 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1850 lines when AUTH and/or TLS are in use. This is the "received protocol"
1851 field. Exim used to use "asmtp" for authenticated SMTP, without any
1852 indication (in the protocol name) for TLS use. Now it follows the RFC and
1853 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1854 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1855 names appear in log lines as well as in Received: header lines.
1857 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1858 extensions, and comparison tests, and to fix some bugs.
1860 14. Changes to the "personal" filter test:
1862 (1) The test was buggy in that it was just doing the equivalent of
1863 "contains" tests on header lines. For example, if a user's address was
1864 anne@some.where, the "personal" test would incorrectly be true for
1866 To: susanne@some.where
1868 This test is now done by extracting each address from the header in turn,
1869 and checking the entire address. Other tests that are part of "personal"
1870 are now done using regular expressions (for example, to check local parts
1871 of addresses in From: header lines).
1873 (2) The list of non-personal local parts in From: addresses has been
1874 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1875 taken from the Sieve specification recommendations.
1877 (3) If the message contains any header line starting with "List-" it is
1878 treated as non-personal.
1880 (4) The test for "circular" in the Subject: header line has been removed
1881 because it now seems ill-conceived.
1883 15. Minor typos in src/EDITME comments corrected.
1885 16. Installed latest exipick from John Jetmore.
1887 17. If headers_add on a router specified a text string that was too long for
1888 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1889 of string_sprintf() is now avoided.
1891 18. $message_body_size was not set (it was always zero) when running the DATA
1892 ACL and the local_scan() function.
1894 19. For the "mail" command in an Exim filter, no default was being set for
1895 the once_repeat time, causing a random time value to be used if "once" was
1896 specified. (If the value happened to be <= 0, no repeat happened.) The
1897 default is now 0s, meaning "never repeat". The "vacation" command was OK
1898 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1899 (I found it when inspecting the code).
1901 20. There is now an overall timeout for performing a callout verification. It
1902 defaults to 4 times the callout timeout, which applies to individual SMTP
1903 commands during the callout. The overall timeout applies when there is more
1904 than one host that can be tried. The timeout is checked before trying the
1905 next host. This prevents very long delays if there are a large number of
1906 hosts and all are timing out (e.g. when the network connections are timing
1907 out). The value of the overall timeout can be changed by specifying an
1908 additional sub-option for "callout", called "maxwait". For example:
1910 verify = sender/callout=5s,maxwait=20s
1912 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1913 to the end before writing, but this should make it even safer).
1915 22. Exim was forgetting that it had advertised PIPELINING for the second and
1916 subsequent messages on an SMTP connection. It was also not resetting its
1917 memory on STARTTLS and an internal HELO.
1919 23. When Exim logs an SMTP synchronization error within a session, it now
1920 records whether PIPELINING has been advertised or not.
1922 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1923 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1924 rather than long int.
1926 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1928 26. Added the never_mail option to autoreply.
1934 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1935 crash if the getsockname() call failed; this can happen if a connection is
1936 closed very soon after it is established. The problem was simply in the
1937 order in which certain operations were done, causing Exim to try to write
1938 to the SMTP stream before it had set up the file descriptor. The bug has
1939 been fixed by making things happen in the correct order.
1945 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1946 before the connection was closed, thus losing the rejection response.
1948 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1949 some early Solaris releases, but causes trouble in current releases where
1950 socklen_t is defined.
1952 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1955 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1958 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1959 that releasing the top of it at the end releases what was used for sub-
1960 expansions (unless the block got too big). However, discard this block if
1961 the first thing is a variable or header, so that we can use its block when
1962 it is dynamic (useful for very large $message_headers, for example).
1964 6. Lookups now cache *every* query, not just the most recent. A new, separate
1965 store pool is used for this. It can be recovered when all lookup caches are
1966 flushed. Lookups now release memory at the end of their result strings.
1967 This has involved some general refactoring of the lookup sources.
1969 7. Some code has been added to the store_xxx() functions to reduce the amount
1970 of flapping under certain conditions.
1972 8. log_incoming_interface used to affect only the <= reception log lines. Now
1973 it causes the local interface and port to be added to several more SMTP log
1974 lines, for example "SMTP connection from", and rejection lines.
1976 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1978 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1980 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1981 could be overwritten at the end of the current message (or the start of a
1982 new message if it was set in a HELO ACL). The value is now preserved for
1983 the duration of the SMTP connection.
1985 12. If a transport had a headers_rewrite setting, and a matching header line
1986 contained an unqualified address, that address was qualified, even if it
1987 did not match any rewriting rules. The underlying bug was that the values
1988 of the flags that permit the existence of unqualified sender and recipient
1989 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1990 non-local messages, and by -bnq for local messages) were not being
1991 preserved with the message after it was received.
1993 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1994 "next input=" as part of the text comprising the host identity instead of
1995 the correct text. The code was using the same buffer for two different
1996 strings. However, depending on which order the printing function evaluated
1997 its arguments, the bug did not always show up. Under Linux, for example, my
1998 test suite worked just fine.
2000 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2001 doesn't work with some older versions of Perl. It has been changed to "my",
2002 which in any case is probably the better facility to use.
2004 15. A really picky compiler found some instances of statements for creating
2005 error messages that either had too many or two few arguments for the format
2008 16. The size of the buffer for calls to the DNS resolver has been increased
2009 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2010 for addresses that have a lot of PTR records. This alleviates a problem; it
2011 does not fully solve it.
2013 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2014 buffer now truncates the list and logs the incident, which is the same
2015 action as happens when Exim is looking up a host name and its aliases.
2016 Previously in this situation something unpredictable would happen;
2017 sometimes it was "internal error: store_reset failed".
2019 18. If a server dropped the connection unexpectedly when an Exim client was
2020 using GnuTLS and trying to read a response, the client delivery process
2021 crashed while trying to generate an error log message.
2023 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2024 string, for example:
2026 warn message = H1: something\nH2: something
2028 the text was added as a single header line from Exim's point of view
2029 though it ended up OK in the delivered message. However, searching for the
2030 second and subsequent header lines using $h_h2: did not work. This has been
2031 fixed. Similarly, if a system filter added multiple headers in this way,
2032 the routers could not see them.
2034 20. Expanded the error message when iplsearch is called with an invalid key to
2035 suggest using net-iplsearch in a host list.
2037 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2038 ACLs are no longer actually imposed (and a message to that effect is
2041 22. If a "gecos" field in a passwd entry contained escaped characters, in
2042 particular, if it contained a \" sequence, Exim got it wrong when building
2043 a From: or a Sender: header from that name. A second bug also caused
2044 incorrect handling when an unquoted " was present following a character
2045 that needed quoting.
2047 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2048 was not being matched caselessly.
2050 24. Arranged for all hyphens in the exim.8 source to be escaped with
2053 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2054 properly. Recipient callout cache records were still being keyed to include
2055 the sender, even when use_sender was set false. This led to far more
2056 callouts that were necessary. The sender is no longer included in the key
2057 when use_sender is false.
2059 26. Added "control = submission" modifier to ACLs.
2061 27. Added the ${base62d: operator to decode base 62 numbers.
2063 28. dnsdb lookups can now access SRV records.
2065 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2066 the configuration file.
2068 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2069 (-v) mode. This makes the output for a verbose queue run more intelligible.
2071 31. Added a use_postmaster feature to recipient callouts.
2073 32. Added the $body_zerocount variable, containing the number of binary zero
2074 bytes in the message body.
2076 33. The time of last modification of the "new" subdirectory is now used as the
2077 "mailbox time last read" when there is a quota error for a maildir
2080 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2082 35. Added +ignore_unknown as a special item in host lists.
2084 36. Code for decoding IPv6 addresses in host lists is now included, even if
2085 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2086 address was recognized as an IP address, but was then not correctly decoded
2087 into binary, causing unexpected and incorrect effects when compared with
2094 1. Very minor rewording of debugging text in manualroute to say "list of
2095 hosts" instead of "hostlist".
2097 2. If verify=header_syntax was set, and a header line with an unqualified
2098 address (no domain) and a large number of spaces between the end of the
2099 name and the colon was received, the reception process suffered a buffer
2100 overflow, and (when I tested it) crashed. This was caused by some obsolete
2101 code that should have been removed. The fix is to remove it!
2103 3. When running in the test harness, delay a bit after writing a bounce
2104 message to get a bit more predictability in the log output.
2106 4. Added a call to search_tidyup() just before forking a reception process. In
2107 theory, someone could use a lookup in the expansion of smtp_accept_max_
2108 per_host which, without the tidyup, could leave open a database connection.
2110 5. Added the variables $recipient_data and $sender_data which get set from a
2111 lookup success in an ACL "recipients" or "senders" condition, or a router
2112 "senders" option, similar to $domain_data and $local_part_data.
2114 6. Moved the writing of debug_print from before to after the "senders" test
2117 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2118 problems for message scanning, either using a data ACL, or using
2119 local_scan() because the Received: header was not generated till after they
2120 were called (in order to set the time as the time of reception completion).
2121 I have revised the way this works. The header is now generated after the
2122 body is received, but before the ACL or local_scan() are called. After they
2123 are run, the timestamp in the header is updated.
2129 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2130 before starting a queue runner without re-exec. This happened only when
2131 deliver_drop_privilege was set or when the Exim user was set to root. The
2132 effect of the bug was that timeouts during subsequent deliveries caused
2133 crashes instead of being properly handled. The handler is now left at its
2134 default (and expected) setting.
2136 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2137 message, again when deliver_drop_privilege is set or Exim is run as root.
2138 The bug described in (1) was not present in this case, but the tidying up
2139 of the other signals was missing. I have made the two cases consistent.
2141 3. The ignore_target_hosts setting on a manualroute router was being ignored
2142 for hosts that were looked up using the /MX notation.
2144 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2147 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2148 operated on the sender address. After changing the $sender_address to <>
2149 for the sender address verify, Exim was re-instated it as the original
2150 (before rewriting) address, but remembering that it had rewritten it, so it
2151 wasn't rewriting it again. This bug also had the effect of breaking the
2152 sender address verification caching when the sender address was rewritten.
2154 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2155 This has been changed so that if the ip literal address matches
2156 ignore_target_hosts, the router declines.
2158 7. Added expansion conditions match_domain, match_address, and match_local_
2159 part (NOT match_host).
2161 8. The placeholder for the Received: header didn't have a length field set.
2163 9. Added code to Exim itself and to exim_lock to test for a specific race
2164 condition that could lead to file corruption when using MBX delivery. The
2165 issue is with the lockfile that is created in /tmp. If this file is removed
2166 after a process has opened it but before that process has acquired a lock,
2167 there is the potential for a second process to recreate the file and also
2168 acquire a lock. This could lead to two Exim processes writing to the file
2169 at the same time. The added code performs the same test as UW imapd; it
2170 checks after acquiring the lock that its file descriptor still refers to
2171 the same named file.
2173 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2174 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2177 11. Added the smtp_active_hostname option. If used, this will typically be made
2178 to depend on the incoming interface address. Because $interface_address is
2179 not set up until the daemon has forked a reception process, error responses
2180 that can happen earlier (such as "too many connections") no longer contain
2183 12. If an expansion in a condition on a "warn" statement fails because a lookup
2184 defers, the "warn" statement is abandoned, and the next ACL statement is
2185 processed. Previously this caused the whole ACL to be aborted.
2187 13. Added the iplsearch lookup type.
2189 14. Added ident_timeout as a log selector.
2191 15. Added tls_certificate_verified as a log selector.
2193 16. Added a global option tls_require_ciphers (compare the smtp transport
2194 option of the same name). This controls incoming TLS connections.
2196 17. I finally figured out how to make tls_require_ciphers do a similar thing
2197 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2198 before starting the TLS session.
2200 18. Tabs are now shown as \t in -bP output.
2202 19. If the log selector return_path_on_delivery was set, Exim crashed when
2203 bouncing a message because it had too many Received: header lines.
2205 20. If two routers both had headers_remove settings, and the first one included
2206 a superfluous trailing colon, the final name in the first list and the
2207 first name in the second list were incorrectly joined into one item (with a
2208 colon in the middle).
2214 1. Added -C and -D options to the exinext utility, mainly to make it easier
2215 to include in the automated testing, but these could be helpful when
2216 multiple configurations are in use.
2218 2. The exinext utility was not formatting the output nicely when there was
2219 an alternate port involved in the retry record key, nor when there was a
2220 message id as well (for retries that were specific to a specific message
2221 and a specific host). It was also confused by IPv6 addresses, because of
2222 the additional colons they contain. I have fixed the IPv4 problem, and
2223 patched it up to do a reasonable job for IPv6.
2225 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2226 delivery, the log line now contains "pipelined" if PIPELINING was used.
2228 4. An SMTP transport process used to panic and die if the bind() call to set
2229 an explicit outgoing interface failed. This has been changed; it is now
2230 treated in the same way as a connect() failure.
2232 5. A reference to $sender_host_name in the part of a conditional expansion
2233 that was being skipped was still causing a DNS lookup. This no longer
2236 6. The def: expansion condition was not recognizing references to header lines
2237 that used bh_ and bheader_.
2239 7. Added the _cache feature to named lists.
2241 8. The code for checking quota_filecount in the appendfile transport was
2242 allowing one more file than it should have been.
2244 9. For compatibility with Sendmail, the command line option
2252 and sets the incoming protocol and host name (for trusted callers). The
2253 host name and its colon can be omitted when only the protocol is to be set.
2254 Note the Exim already has two private options, -pd and -ps, that refer to
2255 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2256 "s", but I don't think that's a major issue.
2258 10. A number of refactoring changes to the code, none of which should affect
2261 (a) The number of logging options was getting close to filling up the
2262 32-bit word that was used as a bit map. I have split them into two classes:
2263 those that are passed in the argument to log_write(), and those that are
2264 only ever tested independently outside of that function. These are now in
2265 separate 32-bit words, so there is plenty of room for expansion again.
2266 There is no change in the user interface or the logging behaviour.
2268 (b) When building, for example, log lines, the code previously used a
2269 macro that called string_cat() twice, in order to add two strings. This is
2270 not really sufficiently general. Furthermore, there was one instance where
2271 it was actually wrong because one of the argument was used twice, and in
2272 one call a function was used. (As it happened, calling the function twice
2273 did not affect the overall behaviour.) The macro has been replaced by a
2274 function that can join an arbitrary number of extra strings onto a growing
2277 (c) The code for expansion conditions now uses a table and a binary chop
2278 instead of a serial search (which was left over from when there were very
2279 few conditions). Also, it now recognizes conditions like "pam" even when
2280 the relevant support is not compiled in: a suitably worded error message is
2281 given if an attempt is made to use such a condition.
2283 11. Added ${time_interval:xxxxx}.
2285 12. A bug was causing one of the ddress fields not to be passed back correctly
2286 from remote delivery subprocesses. The field in question was not being
2287 subsequently used, so this caused to problems in practice.
2289 13. Added new log selectors queue_time and deliver_time.
2291 14. Might have fixed a bug in maildirsizefile handling that threw up
2292 "unexpected character" debug warnings, and recalculated the data
2293 unnecessarily. In any case, I expanded the warning message to give more
2296 15. Added the message "Restricted characters in address" to the statements in
2297 the default ACL that block characters like @ and % in local parts.
2299 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2300 Three changes have been made:
2302 (a) There was a serious bug; a negative response to MAIL caused the whole
2303 recipient domain to be cached as invalid, thereby blocking all messages
2304 to all local parts at the same domain, from all senders. This bug has
2305 been fixed. The domain is no longer cached after a negative response to
2306 MAIL if the sender used is not empty.
2308 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2311 (c) A new callout option, "use_sender" has been added for people who want
2312 the modified behaviour.
2318 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2321 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2322 indeed breaks things for older releases.
2324 3. Added additional logging to the case where there is a problem reading data
2325 from a filter that is running in a subprocess using a pipe, in order to
2326 try to track down a specific problem.
2328 4. Testing facility fudge: when running in the test harness and attempting
2329 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2330 getting "No route to host". Convert this to a timeout.
2332 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2335 6. Some OS don't have socklen_t but use size_t instead. This affects the
2336 fifth argument of getsockopt() amongst other things. This is now
2337 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2338 can be set for individual OS. I have set it for SunOS5, OSF1, and
2339 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2340 some earlier ones do not.
2342 7. Change 4.30/15 was not doing the test caselessly.
2344 8. The standard form for an IPv6 address literal was being rejected by address
2345 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2346 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2347 this, as well as the form without the "IPv6" on the front (but only when
2348 address literals are enabled, of course).
2350 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2352 10. Exim crashed if a message with an empty sender address specified by -f
2353 encountered a router with an errors_to setting. This could be provoked only
2354 by a command such as
2358 where an empty string was supplied; "<>" did not hit this bug.
2360 11. Installed PCRE release 4.5.
2362 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2363 remained set. It is now erased.
2365 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2366 times from message ids (which are base 36 rather than the normal 62).
2368 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2369 were being counted as actual protocol errors, and logged if the log
2370 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2371 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2372 and DATA following a set of rejected RCPTs do not count as protocol errors.
2373 In other words, Exim assumes they were pipelined, though this may not
2374 actually be the case. Of course, in all cases the client gets an
2375 appropriate error code.
2377 15. If a lookup fails in an ACL condition, a message about the failure may
2378 be available; it is used if testing the ACL cannot continue, because most
2379 such messages specify what the cause of the deferral is. However, some
2380 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2381 that caused an old message to be retained and used if a later statement
2382 caused a defer, replacing the real cause of the deferral.
2384 16. If an IP address had so many PTR records that the DNS lookup buffer
2385 was not large enough to hold them, Exim could crash while trying to process
2386 the truncated data. It now detects and logs this case.
2388 17. Further to 4.21/58, another change has been made: if (and only if) the
2389 first line of a message (the first header line) ends with CRLF, a bare LF
2390 in a subsequent header line has a space inserted after it, so as not to
2391 terminate the header.
2393 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2394 unnecessarily, used atoi() instead of strtol(), and didn't check the
2395 termination when getting file sizes from file names by regex.
2397 19. Completely re-implemented the support for maildirsize files, in the light
2398 of a number of problems with the previous contributed implementation
2399 (4.30/29). In particular:
2401 . If the quota is zero, the maildirsize file is maintained, but no quota is
2404 . If the maildir directory does not exist, it is created before any attempt
2405 to write a maildirsize file.
2407 . The quota value in the file is just a cache; if the quota is changed in
2408 the transport, the new value overrides.
2410 . A regular expression is available for excluding directories from the
2413 20. The autoreply transport checks the characters in options that define the
2414 message's headers; it allows continued headers, but it was checking with
2415 isspace() after an embedded newline instead of explicitly looking for a
2418 21. If all the "regular" hosts to which an address was routed had passed their
2419 expiry times, and had not reached their retry times, the address was
2420 bounced, even if fallback hosts were defined. Now Exim should go on to try
2423 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2424 equivalent code in the SMTP transport. Some hosts send humungous responses
2425 to HELO/EHLO, more than 1024 it seems.
2427 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2428 gives compiler warnings in some environments. I've now done it "properly",
2431 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2432 (because of problems with the built-in one) was declared to return uschar *
2433 instead of char *, causing compiler failure.
2435 25. Fixed a file descriptor leak when processing alias/forward files.
2437 26. Fixed a minor format string issue in dbfn.c.
2439 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2441 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2442 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2445 29. When a sender address is verified, it is cached, to save repeating the test
2446 when there is more than one recipient in a message. However, when the
2447 verification involves a callout, it is possible for different callout
2448 options to be set for different recipients. It is too complicated to keep
2449 track of this in the cache, so now Exim always runs a verification when a
2450 callout is required, relying on the callout cache for the optimization.
2451 The overhead is duplication of the address routing, but this should not be
2454 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2455 to be verified with callout=postmaster, and the main callout worked but the
2456 postmaster check failed, the verification correctly failed. However, if a
2457 subsequent RCPT command asked for sender verification *without* the
2458 postmaster check, incorrect caching caused this verification also to fail,
2461 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2462 it was not caching the DNS options (qualify_single, search_parents) that
2463 were used when the lookup failed. A subsequent lookup with different
2464 options therefore always gave the same answer, though there were cases
2465 where it should not have. (Example: a "domains = !$mx_any" option on a
2466 dnslookup router: the "domains" option is always processed without any
2467 widening, but the router might have qualify_single set.) Now Exim uses the
2468 cached value only when the same options are set.
2470 32. Added John Jetmore's "exipick" utility to the distribution.
2472 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2473 than a timeout (e.g. a certificate is required, and is not provided), an
2474 Exim server now closes the connection immediately. Previously it waited for
2475 the client to close - but if the client is SSL, it seems that they each
2476 wait for each other, leading to a delay before one of them times out.
2478 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2479 maintained 0.8.x compatibility because I don't think many are using it, and
2480 it is clearly obsolete.
2482 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2485 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2486 client certificate was expired. A simple patch fixes this, though I don't
2487 understand the full logic of why the verify callback is called multiple
2490 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2491 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2492 which causes problems with some clients (such as the Certicom SSL Plus
2493 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2494 disables the coutermeasure allowing Eudora to connect."
2496 38. Exim was not checking that a write() to a log file succeeded. This could
2497 lead to Bad Things if a log got too big, in particular if it hit a file
2498 size limit. Exim now panics and dies if it cannot write to a log file, just
2499 as it does if it cannot open a log file.
2501 39. Modified OS/Makefile-Linux so that it now contains
2503 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2505 The two -D definitions ensure that Exim is compiled with large file
2506 support, which makes it possible to handle log files that are bigger than
2509 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2510 instance) a domain was checked against a named list that involved a lookup,
2511 causing $domain_data to be set, then another domain was checked against the
2512 same list, then the first domain was re-checked, the value of $domain_data
2513 after the final check could be wrong. In particular, if the second check
2514 failed, it could be set empty. This bug probably also applied to
2517 41. The strip_trailing_dot option was not being applied to the address given
2518 with the -f command-line option.
2520 42. The code for reading a message's header from the spool was incrementing
2521 $received_count, but never initializing it. This meant that the value was
2522 incorrect (doubled) while delivering a message in the same process in which
2523 it was received. In the most common configuration of Exim, this never
2524 happens - a fresh exec is done - but it can happen when
2525 deliver_drop_privilege is set.
2527 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2528 it now includes up to 150 characters of the unexpected data in the log
2531 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2532 and building data strings. The size of both of these buffers was 10 000
2533 bytes - far larger than anybody would *ever* want, thought I. Needless to
2534 say, somebody hit the limit. I have increased the maximum line length to
2535 20 000 and the maximum data length of concatenated lines to 100 000. I have
2536 also fixed two bugs, because there was no checking on these buffers. Tsk,
2537 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2538 buffer is too small.
2540 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2541 lsearch lookups. Now it does.
2543 46. When parsing a route_list item in a manualroute router, a fixed-length
2544 buffer was used for the list of hosts. I made this 1024 bytes long,
2545 thinking that nobody would ever have a list of hosts that long. Wrong.
2546 Somebody had a whole pile of complicated expansion conditions, and the
2547 string was silently truncated, leading to an expansion error. It turns out
2548 that it is easier to change to an unlimited length (owing to other changes
2549 that have happened since this code was originally written) than to build
2550 structure for giving a limitation error. The length of the item that
2551 expands into the list of hosts is now unlimited.
2553 47. The lsearch lookup could not handle data where the length of text line was
2554 more than 4095 characters. Such lines were truncated, leading to shortened
2555 data being returned. It should now handle lines of any length.
2557 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2558 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2561 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2562 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2563 no longer happens in comment lines. A list of replacements is now placed
2564 at the head of all of the source files, except those whose only change is
2565 to replace PERL_COMMAND in the very first #! line.
2567 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2568 messages on the queue, with a bottom-up merge sort, using code contributed
2569 by Michael Haardt. This should make operations like -bp somewhat faster on
2570 large queues. It won't affect queue runners, except when queue_run_in_order
2573 51. Installed eximstats 1.31 in the distribution.
2575 52. Added support for SRV lookups to the dnslookup router.
2577 53. If an ACL referred to $message_body or $message_body_end, the value was not
2578 reset for any messages that followed in the same SMTP session.
2580 54. The store-handling optimization for building very long strings was not
2581 differentiating between the different store pools. I don't think this
2582 actually made any difference in practice, but I've tidied it.
2584 55. While running the routers to verify a sender address, $sender_address
2585 was still set to the sender address. This is wrong, because when routing to
2586 send a bounce to the sender, it would be empty. Therefore, I have changed
2587 it so that, while verifying a sender address, $sender_address is set to <>.
2588 (There is no change to what happens when verifying a recipient address.)
2590 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2591 target A or AAAA records (if not already returned) without resetting the
2592 qualify_single or search_parents options of the DNS resolver. These are
2593 inappropriate in this case because the targets of MX and SRV records must
2594 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2595 target that, when qualified, matched something in the local domain. These
2596 two options are now turned off when doing these lookups.
2598 57. It seems that at least some releases of Reiserfs (which does not have the
2599 concept of a fixed number of inodes) returns zero and not -1 for the
2600 number of available inodes. This interacted badly with check_spool_inodes,
2601 which assumed that -1 was the "no such thing" setting. What I have done is
2602 to check that the total number of inodes is greater than zero before doing
2603 the test of how many are available.
2605 58. When a "warn" ACL statement has a log_message modifier, the message is
2606 remembered, and not repeated. This is to avoid a lot of repetition when a
2607 message has many recipients that cause the same warning to be written.
2608 Howewer, Exim was preserving the list of already written lines for an
2609 entire SMTP session, which doesn't seem right. The memory is now reset if a
2610 new message is started.
2612 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2613 debugging output unless log_rewrite was also set.
2615 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2616 of (int)(handle) when we know that handle contains (void *)(-1).
2618 61. The Exim daemon panic-logs an error return when it closes the incoming
2619 connection. However "connection reset by peer" seems to be common, and
2620 isn't really an error worthy of noting specially, so that particular error
2623 62. When Exim is trying to find all the local interfaces, it used to panic and
2624 die if the ioctl to get the interface flags failed. However, it seems that
2625 on at least one OS (Solaris 9) it is possible to have an interface that is
2626 included in the list of interfaces, but for which you get a failure error
2627 for this call. This happens when the interface is not "plumbed" into a
2628 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2629 failure of the "get flags" call assumes that the interface is down.
2631 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2632 makes life easier for people who are doing arithmetic on fields extracted
2633 from dates, where you often get leading zeros that should not be
2634 interpreted as octal.
2636 64. Added qualify_domain to the redirect router, to override the global
2639 65. If a pathologically long header line contained very many addresses (the
2640 report of this problem mentioned 10 000) and each of them was rewritten,
2641 Exim could use up a very large amount of memory. (It kept on making new
2642 copies of the header line as it rewrote, and never released the old ones.)
2643 At the expense of a bit more processing, the header rewriting function has
2644 been changed so that it no longer eats memory in this way.
2646 66. The generation of the Received: header has been moved from the time that a
2647 message starts to be received, to the time that it finishes. The timestamp
2648 in the Received: header should now be very close to that of the <= log
2649 line. There are two side-effects of this change:
2651 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2652 logged header lines no longer include the local Received: line, because
2653 it has not yet been created. The same applies to a copy of the message
2654 that is returned to a non-SMTP sender when a message is rejected.
2656 (b) When a filter file is tested using -bf, no additional Received: header
2657 is added to the test message. After some thought, I decided that this
2660 This change does not affect the value of $received_for. It is still set
2661 after address rewriting, but before local_scan() is called.
2663 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2665 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2666 gave an unhelpful panic error message, and a defer error. I have managed to
2667 change this behaviour so that it now rejects any supplied certificate,
2668 which seems right, as the list of acceptable certificates is empty.
2670 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2671 gave an unhelpful defer error. I have not managed to make this reject any
2672 supplied certificates, but the error message it gives is "no certificate
2673 supplied", which is not helpful.
2675 70. exigrep's output now also includes lines that are not associated with any
2676 message, but which match the given pattern. Implemented by a patch from
2677 Martin Sluka, which also tidied up the Perl a bit.
2679 71. Recipient callout verification, like sender verification, was using <> in
2680 the MAIL FROM command. This isn't really the right thing, since the actual
2681 sender may affect whether the remote host accepts the recipient or not. I
2682 have changed it to use the actual sender in the callout; this means that
2683 the cache record is now keyed on a recipient/sender pair, not just the
2684 recipient address. There doesn't seem to be a real danger of callout loops,
2685 since a callout by the remote host to check the sender would use <>.
2686 [SEE ABOVE: changed after hitting problems.]
2688 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2689 temporary errors. However, in the case of such a code being given after
2690 the end of a data transmission (i.e. after ".") Exim was failing to write
2691 a retry record for the message. (Yes, there was some broken host that was
2692 actually sending 8xx at this point.)
2694 73. An unknown lookup type in a host list could cause Exim to panic-die when
2695 the list was checked. (An example that provoked this was putting <; in the
2696 middle of a list instead of at the start.) If this happened during a DATA
2697 ACL check, a -D file could be left lying around. This kind of configuration
2698 error no longer causes Exim to die; instead it causes a defer errror. The
2699 incident is still logged to the main and panic logs.
2701 74. Buglet left over from Exim 3 conversion. The message "too many messages
2702 in one connection" was written to the rejectlog but not the mainlog, except
2703 when address rewriting (yes!) was being logged.
2705 75. Added write_rejectlog option.
2707 76. When a system filter was run not as root (that is, when system_filter_user
2708 was set), the values of the $n variables were not being returned to the
2709 main process; thus, they were not subsequently available in the $sn
2712 77. Added +return_path_on_delivery log selector.
2714 78. A connection timeout was being treated differently from recipients deferred
2715 when testing hosts_max_try with a message that was older than the host's
2716 retry timeout. (The host should not be counted, thus allowing all hosts to
2717 be tried at least once before bouncing.) This may have been the cause of an
2718 occasionally reported bug whereby a message would remain on the queue
2719 longer than the retry timeout, but would be bounced if a delivery was
2720 forced. I say "may" because I never totally pinned down the problem;
2721 setting up timeout/retry tests is difficult. See also the next item.
2723 79. The ultimate address timeout was not being applied to errors that involved
2724 a combination of host plus message (for example, a timeout on a MAIL
2725 command). When an address resolved to a number of possible hosts, and they
2726 were not all tried for each delivery (e.g. because of hosts_max_try), a
2727 message could remain on the queue longer than the retry timeout.
2729 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2732 81. Fixed an obscure SMTP outgoing bug which required at least the following
2733 conditions: (a) there was another message waiting for the same server;
2734 (b) the server returned 5xx to all RCPT commands in the first message so
2735 that the message was not completed; (c) the server dropped the connection
2736 or gave a negative response to the RSET that Exim sends to abort the
2737 transaction. The observed case was a dropped connection after DATA that had
2738 been sent in pipelining mode. That is, the server had advertised PIPELINING
2739 but was not implementing it correctly. The effect of the bug was incorrect
2740 behaviour, such as trying another host, and this could lead to a crash.
2746 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2747 and daemon.c were passed as pointers to ints; they should have been
2748 pointers to socklen_t variables (which are typically unsigned ints).
2750 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2753 3. Fixed a really odd bug that affected only the testing scheme; patching a
2754 certain fixed string in the binary changed the value of another string that
2755 happened to be identical to the end of the original first string.
2757 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2758 name", it returns that address as the IP address. On some operating
2759 systems (e.g. Solaris), it also passes back the IP address string as the
2760 "host name". However, on others (e.g. Linux), it passes back an empty
2761 string. Exim wasn't checking for this, and was changing the host name to an
2762 empty string, assuming it had been canonicized.
2764 5. Although rare, it is permitted to have more than one PTR record for a given
2765 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2766 all the names associated with an address, because they do in Solaris.
2767 However, it seems that they do not in Linux for data that comes from the
2768 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2769 I found this out when I moved to a new Linux workstation and tried to run
2770 the Exim test suite.
2772 To get round this problem I have changed the code so that it now does its
2773 own call to the DNS to look up PTR records when searching for a host name.
2774 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2775 addresses that are only in /etc/hosts are still found.
2777 This behaviour is, however, controlled by an option called host_lookup_
2778 order, which defaults to "bydns:byaddr". If people want to use the other
2779 order, or indeed, just use one or the other means of lookup, they can
2780 specify it in this variable.
2782 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2783 some operating systems, this comes back from gethostbyaddr() as an empty
2784 string, and this is what Exim used to test for. However, it seems that in
2785 other systems, "." is yielded. Exim now tests for this case too.
2787 7. The values of check_spool_space and check_log_space are now held internally
2788 as a number of kilobytes instead of an absolute number of bytes. If a
2789 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2790 kilobyte. This means that much larger values can be stored.
2792 8. Exim monitor: an attempt to get the action menu when not actually pointing
2793 at a message produces an empty menu entitled "No message selected". This
2794 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2795 no entries in it ("Shell widget menu has zero width and/or height"). So I
2796 have added a single, blank menu entry in this case.
2798 9. Added ${quote_local_part.
2800 10. MIME decoding is now applied to the contents of Subject: header lines when
2803 11. Now that a reference to $sender_host_address automatically causes a reverse
2804 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2805 host lookup before query-style lookups in lists that might use this
2806 variable. This has therefore been abolished, and the "net-" prefix is no
2807 longer necessary for query-style lookups.
2809 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2810 have been a typo for LFLAGS, so it has been changed.
2812 13. The install script calls Exim with "-C /dev/null" in order to find the
2813 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2814 to be output. Howeve, since Exim outputs its version number before the
2815 error, it didn't break the script. It just looked ugly. I fixed this by
2816 always allowing "-C /dev/null" if the caller is root.
2818 14. Ignore overlarge ACL variable number when reading spool file - insurance
2819 against a later release with more variables having written the file.
2821 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2822 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2823 this, as well as the form without the "IPv6" on the front.
2825 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2826 OS/Makefile-Darwin file.
2828 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2829 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2831 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2832 variable. However, some OS use ints for inodes. I've added cast to long int
2833 to get rid of the compiler warning.
2835 19. I had forgotten to lock out "/../" in configuration file names when
2836 ALT_CONFIG_PREFIX was set.
2838 20. Routers used for verification do not need to specify transports. However,
2839 if such a router generated a host list, and callout was configured, Exim
2840 crashed, because it could not find a port number from the (non-existent)
2841 transport. It now assumes port 25 in this circumstance.
2843 21. Added the -t option to exigrep.
2845 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2846 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2847 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2848 them is set, LOOKUP_LSEARCH is forced.
2850 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2852 24. Added sender and host information to the "rejected by local_scan()" log
2853 line; previously there was no indication of these.
2855 25. Added .include_if_exists.
2857 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2858 Linux. It turns out that not all file systems support this. Apparently some
2859 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2860 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2861 not supported on the file descriptor, is now ignored when Exim is trying to
2862 sync a directory. This applies only to Linux.
2864 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2866 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2867 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2870 29. Added support for maildirsize files from supplied patch (modified a bit).
2872 30. The use of :fail: followed by an empty string could lead Exim to respond to
2873 sender verification failures with (e.g.):
2875 550 Verification failed for <xxx>
2876 550 Sender verify failed
2878 where the first response line was missing the '-' that indicates it is not
2879 the final line of the response.
2881 31. The loop for finding the name of the user that called Exim had a hardwired
2882 limit of 10; it now uses the value of finduser_retries, which is used for
2883 all other user lookups.
2885 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2888 33. Exim was neglecting to zero errno before one call of strtol() when
2889 expanding a string and expecting an integer value. On some systems this
2890 resulted in spurious "integer overflow" errors. Also, it was casting the
2891 result into an int without checking.
2893 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2894 did not work. The code looks as if it has *never* worked, though it appears
2895 to have been documented since at least releast 1.62. I have made it work.
2897 35. The "timeout_DNS" error in retry rules, also documented since at least
2898 1.62, also never worked. As it isn't clear exactly what this means, and
2899 clearly it isn't a major issue, I have abolished the feature by treating it
2900 as "timeout", and writing a warning to the main and panic logs.
2902 36. The display of retry rules for -brt wasn't always showing the error code
2905 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2906 timeout_connect_A, timeout_connect_MX.
2908 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2909 to the empty sender.
2911 39. The daemon was not analysing the content of -oX till after it had closed
2912 stderr and disconnected from the controlling terminal. This meant that any
2913 syntax errors were only noted on the panic log, and the return code from
2914 the command was 0. By re-arranging the code a little, I've made the
2915 decoding happen first, so such errors now appear on stderr, and the return
2916 code is 1. However, the actual setting up of the sockets still happens in
2917 the disconnected process, so errors there are still only recorded on the
2920 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2921 connections (as happens on some IP stacks) was logged at start up time as
2922 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2923 it from "IPv6 and IPv4", which means that two separate sockets are being
2926 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2927 says whether AF_INET or AF_INET6 was passed as an argument.
2929 42. Exiwhat output was messed up when time zones were included in log
2932 43. Exiwhat now gives more information about the daemon's listening ports,
2933 and whether -tls-on-connect was used.
2935 44. The "port" option of the smtp transport is now expanded.
2937 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2938 silently ignored. Now an error message is written to the main and panic
2941 46. There's a new ACL modifier called "logwrite" which writes to a log file
2942 as soon as it is encountered.
2944 47. Added $local_user_uid and $local_user_gid at routing time.
2946 48. Exim crashed when trying to verify a sender address that was being
2949 49. Exim was recognizing only a space character after ".include". It now also
2950 recognizes a tab character.
2952 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2953 extracting the relevant information from the specification. The man page no
2954 longer contains scrambled data for the -d option, and I've added a section
2955 at the front about calling Exim under different names.
2957 51. Added "extra_headers" argument to the "mail" command in filter files.
2959 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2962 53. Installed eximstats 1.29.
2964 54. Added transport_filter_timeout as a generic transport option.
2966 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2967 Cc: header lines. This was required by RFC 822, but it not required by RFC
2970 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2971 incoming messages that did not have them. Now it does so only if the
2972 message originates locally, that is, if there is no associated remote host
2973 address. When Resent- header lines are present, this applies to the Resent-
2974 lines rather than the non-Resent- lines.
2976 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2977 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2979 58. Messages for configuration errors now include the name of the main
2980 configuration file - useful now that there may be more than one file in a
2981 list (.included file names were always shown).
2983 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2984 for those rare installations that do not start the daemon as root or run it
2985 setuid root. I've cut out the call to initgroups() if the daemon is not
2988 60. The Exim user and group can now be bound into the binary as text strings
2989 that are looked up at the start of Exim's processing.
2991 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2993 62. Added $mailstore_basename variable.
2995 63. Installed patch to sieve.c from Michael Haardt.
2997 64. When Exim failed to open the panic log after failing to open the main log,
2998 the original message it was trying to log was written to stderr and debug
2999 output, but if they were not available (the usual case in production), it
3000 was lost. Now it is written to syslog before the two lines that record the
3001 failures to open the logs.
3003 65. Users' Exim filters run in subprocesses under the user's uid. It is
3004 possible for a "deliver" command or an alias in a "personal" command to
3005 provoke an address rewrite. If logging of address rewriting is configured,
3006 this fails because the process is not running as root or exim. There may be
3007 a better way of dealing with this, but for the moment (because 4.30 needs
3008 to be released), I have disabled address rewrite logging when running a
3009 filter in a non-root, non-exim process.
3015 1. The buildconfig auxiliary program wasn't quoting the value set for
3016 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3017 not defined. This bug was masked in 4.22 by the effect that was fixed in
3020 2. Some messages that were rejected after a message id was allocated were
3021 shown as "incomplete" by exigrep. It no longer does this for messages that
3022 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3024 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3025 have allow_domain_literals set, the ID did not get logged in the <= line.
3026 Domain literals are now always recognized in Message-ID: header lines.
3028 4. The first argument for a ${extract expansion item is the key name or field
3029 number. Leading and trailing spaces in this item were not being ignored,
3030 causing some misleading effects.
3032 5. When deliver_drop_privilege was set, single queue runner processes started
3033 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3034 same command in the process it spins off) were not dropping privilege.
3036 6. When the daemon running as "exim" started a queue runner, it always
3037 re-executed Exim in the spun-off process. This is a waste of effort when
3038 deliver_drop_privilege is set. The new process now just calls the
3039 queue-runner function directly.
3045 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3048 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3049 ignored. Though the use of -oP was forcing the writing of a pid file, it
3050 was always written to the default place.
3052 3. If the message "no IP address found for host xxxx" is generated during
3053 incoming verification, it is now followed by identification of the incoming
3054 connection (so you can more easily find what provoked it).
3056 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3058 5. Added some features to "harden" Exim a bit more against certain attacks:
3060 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3061 be put in Local/Makefile. This is like the never_users runtime option,
3062 but it cannot be overridden. The default setting is "root".
3064 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3065 prefix string with which any file named in a -C command line option
3068 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3069 is retained for -C and -D only if the caller of Exim is root. Without
3070 it, the exim user may also use -C and -D and retain privilege.
3072 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3073 command line option is disabled.
3075 6. Macro names set by the -D option must start with an upper case letter, just
3076 like macro names defined in the configuration file.
3078 7. Added "dereference=" facility to LDAP.
3080 8. Two instances of the typo "uknown" in the source files are fixed.
3082 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3083 the Configure-Makefile script screwed up while processing it.
3085 10. Incorporated PCRE 4.4.
3087 11. The SMTP synchronization check was not operating right at the start of an
3088 SMTP session. For example, it could not catch a HELO sent before the client
3089 waited for the greeting. There is now a check for outstanding input at the
3090 point when the greeting is written. Because of the duplex, asynchronous
3091 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3092 way, but not yet received, when the check is performed.
3094 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3095 on TCP/IP sockets, because this apparently causes some broken clients to
3098 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3099 unchanged) from the Cygwin maintainer.
3101 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3102 was supported instead of testing for mbx. Effectively a typo.
3104 15. The spa authenticator server code was not checking that the input it
3105 received was valid base64.
3107 16. The debug output line for the "set" modifier in ACLs was not showing the
3108 name of the variable that was being set.
3110 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3112 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3113 lookup on its own. Something else was needed to trigger the lookup. For
3114 example, a match in host_lookup or the need for a host name in a host list.
3115 Now, if $sender_host_name is referenced and the host name has not yet been
3116 looked up, a lookup is performed. If the lookup fails, the variable remains
3117 empty, and $host_lookup_failed is set to "1".
3119 19. Added "eqi" as a case-independent comparison operator.
3121 20. The saslauthd authentication condition could segfault if neither service
3122 nor realm was specified.
3124 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3125 error message that was logged was misleading, and incoming SMTP
3126 connections were dropped. The message is now more accurate, and temporary
3127 errors are given to SMTP connections.
3129 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3130 (except RSET) with a 421 error, until QUIT is received. However, it was
3131 failing to send a response to QUIT.
3133 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3134 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3135 ACL is now run after the helo_try_verify_hosts code.
3137 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3138 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3139 case-independent, so other case variants are also recognized). Apparently
3140 some systems use these upper case variants.
3142 25. If more than two messages were waiting for the same host, and a transport
3143 filter was specified for the transport, Exim sent two messages over the
3144 same TCP/IP connection, and then failed with "socket operation on non-
3145 socket" when it tried to send the third.
3147 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3149 27. The extern definition of crypt16() in expand.c was not being excluded when
3150 the OS had its own crypt16() function.
3152 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3153 as a preferred synonym for return_size_limit, both as an option and as an
3156 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3158 30. Changed the default configuration ACL to relax the local part checking rule
3159 for addresses that are not in any local domains. For these addresses,
3160 slashes and pipe symbols are allowed within local parts, but the sequence
3161 /../ is explicitly forbidden.
3163 31. SPA server authentication was not clearing the challenge buffer before
3166 32. log_message in a "warn" ACL statement was writing to the reject log as
3167 well as to the main log, which contradicts the documentation and doesn't
3168 seem right (because no rejection is happening). So I have stopped it.
3170 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3171 However, I am unable to do any testing of this.
3173 34. Fixed an infelicity in the appendfile transport. When checking directories
3174 for a mailbox, to see if any needed to be created, it was accidentally
3175 using path names with one or more superfluous leading slashes; tracing
3176 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3178 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3179 contents are added to the log line that is written for every discarded
3180 recipient. (Previously a log_message setting was ignored.)
3182 36. The ${quote: operator now quotes the string if it is empty.
3184 37. The install script runs exim in order to find its version number. If for
3185 some reason other than non-existence or emptiness, which it checks, it
3186 could not run './exim', it was installing it with an empty version number,
3187 i.e. as "exim-". This error state is now caught, and the installation is
3190 38. An argument was missing from the function that creates an error message
3191 when Exim fails to connect to the socket for saslauthd authentication.
3192 This could cause Exim to crash, or give a corrupted message.
3194 39. Added isip, isip4, and isip6 to ${if conditions.
3196 40. The ACL variables $acl_xx are now saved with the message, and can be
3197 accessed later in routers, transports, and filters.
3199 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3200 strings in the file are not string-expanded.
3202 42. If a MAIL command specified a SIZE value that was too large to fit into an
3203 int variable, the check against message_size_limit failed. Such values are
3204 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3205 day this will have to be increased, but I don't think I want to be around
3206 when emails are that large.
3213 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3214 iconv() is not standard in FreeBSD.
3216 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3217 IPv6 enabled. The observed symptom was a segmentation fault on return from
3218 the function os_common_find_running_interfaces() in src/os.c.
3220 3. In the check_special_case() function in daemon.c I had used "errno" as an
3221 argument name, which causes warnings on some systems. This was basically a
3222 typo, since it was named "eno" in the comments!
3224 4. The code that waits for the clock to tick (at a resolution of some fraction
3225 of a second) so as to ensure message-id uniqueness was always waiting for
3226 at least one whole tick, when it could have waited for less. [This is
3227 almost certainly not relevant at current processor speeds, where it is
3228 unlikely to ever wait at all. But we try to future-proof.]
3230 5. The function that sleeps for a time interval that includes fractions of a
3231 second contained a race. It did not block SIGALRM between setting the
3232 timer, and suspending (a couple of lines later). If the interval was short
3233 and the sigsuspend() was delayed until after it had expired, the suspension
3234 never ended. On busy systems this could lead to processes getting stuck for
3237 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3238 process, before it forks any delivery processes. The open lookup caching
3239 mechanism meant that the open file or database connection was passed into
3240 the delivery process. The problem was that delivery processes always tidy
3241 up cached lookup data. This could cause a problem for the next delivery
3242 process started by the queue runner, because the external queue runner
3243 process does not know about the closure. So the next delivery process
3244 still has data in the lookup cache. In the case of a file lookup, there was
3245 no problem because closing a file descriptor in a subprocess doesn't affect
3246 the parent. However, if the lookup was caching a connection to a database,
3247 the connection was closed, and the second delivery process was likely to
3248 see errors such as "PGSQL: query failed: server closed the connection
3249 unexpectedly". The problem has been fixed by closing all cached lookups
3250 in a queue runner before running a delivery process.
3252 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3253 seem to have the "const" qualifier which it has on other OS. I've
3256 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3257 *of the same type* (client or server) with the same public name that an
3258 error should be diagnosed.
3260 9. When Exim looked up a host name for an IP address, but failed to find the
3261 original IP address when looking up the host name (a safety check), it
3262 output the message "<ip address> does not match any IP for NULL", which was
3263 confusing, to say the least. The bug was that the host name should have
3264 appeared instead of "NULL".
3266 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3267 user that is built into the binary, and the -C or -D options is used, root
3268 privilege is dropped before the configuration file is read. In addition,
3269 logging is switched to stderr instead of the normal log files. If the
3270 configuration then re-defines the Exim user, the unprivileged environment
3271 is probably not what is expected, so Exim logs a panic warning message (but
3274 However, if deliver_drop_privilege is set, the unprivileged state may well
3275 be exactly what is intended, so the warning has been cut out in that case,
3276 and Exim is allowed to try to write to its normal log files.
3282 1. smtp_return_error_details was not giving details for temporary sender
3283 or receiver verification errors.
3285 2. Diagnose a configuration error if two authenticators have the same public
3288 3. Exim used not to create the message log file for a message until the first
3289 delivery attempt. This could be confusing when incoming messages were held
3290 for policy or load reasons. The message log file is now created at the time
3291 the message is received, and an initial "Received" line is written to it.
3293 4. The automatically generated man page for command line options had a minor
3294 bug that caused no ill effects; however, a more serious problem was that
3295 the procedure for building the man page automatically didn't always
3296 operate. Consequently, release 4.20 contains an out-of-date version. This
3297 shouldn't happen again.
3299 5. When building Exim with embedded Perl support, the script that builds the
3300 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3301 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3303 6. The freeze_tell option was not being used for messages that were frozen on
3304 arrival, either by an ACL or by local_scan().
3306 7. Added the smtp_incomplete_transaction log selector.
3308 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3309 was accepting AUTH without a new EHLO.
3311 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3312 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3315 10. Logging of TCP/IP connections (when configured) now happens in the main
3316 daemon process instead of the child process, so that the TCP/IP connection
3317 count is more accurate (but it can never be perfect).
3319 11. The use of "drop" in a nested ACL was not being handled correctly in the
3320 outer ACL. Now, if condition failure induced by the nested "drop" causes
3321 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3322 or "require"), the connection is dropped.
3324 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3325 that yield "discard" can now be used with an "accept" or a "discard" verb,
3326 but an error is generated for any others (because I can't see a useful way
3327 to define what should happen).
3329 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3330 are now processed in the same way as lines in the Exim configuration file.
3331 In particular, continuation lines are supported.
3333 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3335 15. Added -ti meaning -t -i.
3337 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3338 domains, and warn by logging if others are found.
3340 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3341 returned from GIFCONF when Exim is trying to find the list of interfaces on
3342 a host. The code in os.c has been modified to copy each ifreq to an aligned
3343 structure in all cases.
3345 Also, in some cases, the returned ifreq's were being copied to a 'struct
3346 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3347 means the last couple of bytes of an IPv6 address could be chopped if the
3348 ifreq contained only a normal sockaddr (14 bytes storage).
3350 18. Named domain lists were not supported in the hosts_treat_as_local option.
3351 An entry such as +xxxx was not recognized, and was treated as a literal
3354 19. Ensure that header lines added by a DATA ACL are included in the reject log
3355 if the ACL subsequently rejects the message.
3357 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3358 MD5 (which is deprecated).
3360 21. When testing a filter file using -bf, Exim was writing a message when it
3361 took the sender from a "From " line in the message, but it was not doing so
3362 when it took $return_path from a Return-Path: header line. It now does.
3364 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3365 with a valid header line field name (a series of printing characters
3366 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3368 23. Changed "disc" in the source to "disk" to conform to the documentation and
3369 the book and for uniformity.
3371 24. Ignore Sendmail's -Ooption=value command line item.
3373 25. When execve() failed while trying to run a command in a pipe transport,
3374 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3375 could be confused with a return value of 69 from the command itself. This
3376 has been changed to 127, the value the shell returns if it is asked to run
3377 a non-existent command. The wording for the related log line suggests a
3378 non-existent command as the problem.
3380 26. If received_header_text expands to an empty string, do not add a Received:
3381 header line to the message. (Well, it adds a token one on the spool, but
3382 marks it "old" so that it doesn't get used or transmitted.)
3384 27. Installed eximstats 1.28 (addition of -nt option).
3386 28. There was no check for failure on the call to getsockname() in the daemon
3387 code. This can fail if there is a shortage of resources on the system, with
3388 ENOMEM, for example. A temporary error is now given on failure.
3390 29. Contrary to the C standard, it seems that in some environments, the
3391 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3392 program. Exim now does this explicitly; it affects the formatting of
3393 timestamps using strftime().
3395 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3396 complaints. I've now initialized all the variables, to avoid this.
3398 32. Header lines added by a system filter were not being "seen" during
3399 transport-time rewrites.
3401 33. The info_callback() function passed to OpenSSL is set up with type void
3402 (*)(SSL *, int, int), as described somewhere. However, when calling the
3403 function (actually a macro) that sets it up, the type void(*)() is
3404 expected. I've put in a cast to prevent warnings from picky compilers.
3406 34. If a DNS black list lookup found a CNAME record, but there were no A
3407 records associated with the domain it pointed at, Exim crashed.
3409 35. If a DNS black list lookup returned more than one A record, Exim ignored
3410 all but the first. It now scans all returned addresses if a particular IP
3411 value is being sought. In this situation, the contents of the
3412 $dnslist_value variable are a list of all the addresses, separated by a
3415 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3416 to accept a host name and all its aliases if the forward lookup for any of
3417 them yielded the IP address of the incoming connection. Now it accepts only
3418 those names whose forward lookup yields the correct IP address. Any other
3419 names are discarded. This closes a loophole whereby a rogue DNS
3420 administrator could create reverse DNS records to break through a
3421 wildcarded host restriction in an ACL.
3423 37. If a user filter or a system filter that ran in a subprocess used any of
3424 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3425 the wrong values were passed to the pipe command ($thisaddress had the
3426 value of $0, $0 had the value of $1, etc). This bug was introduced by
3427 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3429 38. Improved the line breaking for long SMTP error messages from ACLs.
3430 Previously, if there was no break point between 40 and 75 characters, Exim
3431 left the rest of the message alone. Two changes have been made: (a) I've
3432 reduced the minimum length to 35 characters; (b) if it can't find a break
3433 point between 35 and 75 characters, it looks ahead and uses the first one
3434 that it finds. This may give the occasional overlong line, but at least the
3435 remaining text gets split now.
3437 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3438 file descriptors might be low, and that setting 1000 would always raise it.
3439 It turns out that in some environments, the limit is already over 1000 and
3440 that lowering it causes trouble. So now Exim takes care not to decrease it.
3442 40. When delivering a message, the value of $return_path is set to $sender_
3443 address at the start of routing (routers may change the value). By an
3444 oversight, this default was not being set up when an address was tested by
3445 -bt or -bv, which affected the outcome if any router or filter referred to
3448 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3449 log only when "message" or "log_message" are set. However, if one of the
3450 conditions was an address verification, or a call to a nested ACL, the
3451 messages generated by the underlying test were being passed through. This
3452 no longer happens. The underlying message is available in $acl_verify_
3453 message for both "message" and "log_message" expansions, so it can be
3454 passed through if needed.
3456 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3457 new expansion $bh_ to give the encoded byte string without charset
3458 translation. Translation happens only if iconv() is available; HAVE_ICONV
3459 indicates this at build time. HEADERS_CHARSET gives the charset to
3460 translate to; headers_charset can change it in the configuration, and
3461 "headers charset" can change it in an individual filter file.
3463 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3464 that creates RFC 2047 encoded "words" labels them as that charset instead
3465 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3466 expansion operator; (ii) when Exim creates a From: line for a local
3467 message; (iii) when a header line is rewritten to include a "phrase" part.
3469 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3470 buggy, causing it to skip the first lines of messages whose message ID
3471 ended in 'D'. This would not have bitten before Exim release 4.14, because
3472 message IDs were unlikely to end in 'D' before then. The effect was to have
3473 incorrect size information for certain domains.
3475 45. #include "config.h" was missing at the start of the crypt16.c module. This
3476 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3479 46. If there was a timeout during a "random" callout check, Exim treated it as
3480 a failure of the random address, and carried on sending RSET and the real
3481 address. If the delay was just some slowness somewhere, the response to the
3482 original RCPT would be taken as a response to RSET and so on, causing
3483 mayhem of various kinds.
3485 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3486 when I implemented it. It didn't allow for the fact that some option values
3487 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3488 the right test for positive values.
3490 48. Domain names in DNS records are case-independent. Exim always looks them up
3491 in lower case. Some resolvers return domain names in exactly the case they
3492 appear in the zone file, that is, they may contain uppercase letters. Not
3493 all resolvers do this - some return always lower case. Exim was treating a
3494 change of case by a resolver as a change of domain, similar to a widening
3495 of a domain abbreviation. This triggered its re-routing code and so it was
3496 trying to route what was effectively the same domain again. This normally
3497 caused routing to fail (because the router wouldn't handle the domain
3498 twice). Now Exim checks for this case specially, and just changes the
3499 casing of the domain that it ultimately uses when it transmits the message
3502 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3505 50. If a filter generated a file delivery with a non-absolute name (possible if
3506 no home directory exists for the router), the forbid_file option was not
3509 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3510 the existing equality matching.
3512 52. Exim was using ints instead of ino_t variables in some places where it was
3513 dealing with inode numbers.
3515 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3516 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3517 called TMPDIR, and if it finds it is different, it changes its value.
3519 54. The smtp_printf() function is now made available to local_scan() so
3520 additional output lines can be written before returning. There is also an
3521 smtp_fflush() function to enable the detection of a dropped connection.
3522 The variables smtp_input and smtp_batched_input are exported to
3525 55. Changed the default runtime configuration: the message "Unknown user"
3526 has been removed from the ACL, and instead placed on the localuser router,
3527 using the cannot_route_message feature. This means that any verification
3528 failures that generate their own messages won't get overridden. Similarly,
3529 the "Unrouteable address" message that was in the ACL for unverifiable
3530 relay addresses has also been removed.
3532 56. Added hosts_avoid_esmtp to the smtp transport.
3534 57. The exicyclog script was not checking for the esoteric option
3535 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3536 will work only if exicyclog is run under the appropriate euid.
3538 58. Following a discussion on the list, the rules by which Exim recognises line
3539 endings on incoming messages have been changed. The -dropcr and drop_cr
3540 options are now no-ops, retained only for backwards compatibility. The
3541 following line terminators are recognized: LF CRLF CR. However, special
3542 processing applies to CR:
3544 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3545 nor a local message in the state where . is a terminator.
3547 (ii) If a bare CR is encountered in a header line, an extra space is added
3548 after the line terminator so as not to end the header. The reasoning
3549 behind this is that bare CRs in header lines are most likely either
3550 to be mistakes, or people trying to play silly games.
3552 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3553 was being incorrectly given as 18 bytes larger than it should have been.
3554 This is a VOB (very old bug).
3556 60. This may never have affected anything current, but just in case it has:
3557 When the local host is found other than at the start of a list of hosts,
3558 the local host, those with the same MX, and any that follow, are discarded.
3559 When the list in question was part of a longer list of hosts, the following
3560 hosts (not currently being processed) were also being discarded. This no
3561 longer happens. I'm not sure if this situation could ever has previously
3564 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3567 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3568 header. This is something that is recommended in a new Internet Draft, and
3569 is something that is documented as being done by Sendmail. There are two
3570 possible values. For messages generated by the autoreply transport, Exim
3573 Auto-Submitted: auto-replied
3575 whereas for all other generated messages (e.g. bounces) it adds
3577 Auto-Submitted: auto-generated
3579 63. The "personal" condition in filters now includes a test for the
3580 Auto-Submitted: header. If it contains the string "auto-" the message it
3581 not considered personal.
3583 64. Added rcpt_include_affixes as a generic transport option.
3585 65. Added queue_only_override (default true).
3587 66. Added the syslog_duplication option.
3589 67. If what should have been the first header line of a message consisted of
3590 a space followed by a colon, Exim was mis-interpreting it as a header line.
3591 It isn't of course - it is syntactically invalid and should therefore be
3592 treated as the start of the message body. The misbehaviour could have
3593 caused a number of strange effects, including loss of data in subsequent
3594 header lines, and spool format errors.
3596 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3597 client host had authenticated. This control can now be exercised by an ACL
3598 for more flexibility.
3600 69. By default, callouts do not happen when testing with -bh. There is now a
3601 variant, -bhc, which does actually run the callout code, including
3602 consulting and updating the callout cache.
3604 70. Added support for saslauthd authentication, courtesy of Alexander
3607 71. If statvfs() failed on the spool or log directories while checking their
3608 size for availability, Exim confusingly gave the error "space shortage".
3609 Furthermore, in debugging mode it crashed with a floating point exception.
3610 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3611 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3612 serious problem, Exim now writes to the main and panic logs when this
3613 happens, with details of the failure. It then refuses to accept the
3614 incoming message, giving the message "spool directory problem" or "log
3615 directory problem" with a 421 code for SMTP messages.
3617 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3618 0, 1, and 2 exist, because some OS complain for execs without them (see
3619 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3620 descriptors. However, the code omitted to check that the open succeeded,
3621 causing mysterious errors if for some reason the permissions on /dev/null
3622 got screwed. Now Exim writes a message to the main and panic logs, and
3623 bombs out if it can't open /dev/null.
3625 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3626 interact so that it is all more flexible. It is supposed to remain
3627 backwards compatible. Also added extra_local_interfaces.
3629 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3630 to bomb out with an assertion failure - to the client this appears as a
3631 connection drop. This problem occurs in the part of the code that was taken
3632 from the Samba project. Fortunately, the assertion is in a very simple
3633 function, so I have fixed this by reproducing the function inline in the
3634 one place where it is called, and arranging for authentication to fail
3635 instead of killing the process with assert().
3637 75. The SPA client code was not working when the server requested OEM rather
3638 than Unicode encoding.
3640 76. Added code to make require_files with a specific uid setting more usable in
3641 the case where statting the file as root fails - usually a non-root-mounted
3642 NFS file system. When this happens and the failure is EACCES, Exim now
3643 forks a subprocess and does the per-uid checking as the relevant uid.
3645 77. Added process_log_path.
3647 78. If log_file_path was not explicitly set, a setting of check_log_space or
3648 check_log_inodes was ignored.
3650 79. If a space check for the spool or log partitions fails, the incident is now
3651 logged. Of course, in the latter case the data may get lost...
3653 80. Added the %p formatting code to string_format() so that it can be used to
3654 print addresses in debug_print(). Adjusted all the address printing in the
3655 debugging in store.c to use %p rather than %d.
3657 81. There was a concern that a line of code in smtp_in.c could overflow a
3658 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3659 initially expressed, the concern was not well-founded, because trailing
3660 spaces are removed early. However, if the trailing spaces were followed by
3661 a NULL, they did not get removed, so the overflow was possible. Two fixes
3664 (a) I re-wrote the offending code in a cleaner fashion.
3665 (b) If an incoming SMTP command contains a NULL character, it is rejected
3668 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3669 runs initgroups(), so that if the Exim user is in any additional groups,
3670 they will be used during message reception.
3676 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob