1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.289 2006/02/09 14:50:58 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.
105 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
106 adding the default one. Similarly, if it contains a Reply-To: header, the
107 errors_reply_to option, if set, is not used.
109 PH/19 When calculating a retry time, Exim used to measure the "time since
110 failure" by looking at the "first failed" field in the retry record. Now
111 it does not use this if it is later than than the arrival time of the
112 message. Instead it uses the arrival time. This makes for better
113 behaviour in cases where some deliveries succeed, thus re-setting the
114 "first failed" field. An example is a quota failure for a huge message
115 when small messages continue to be delivered. Without this change, the
116 "time since failure" will always be short, possible causing more frequent
117 delivery attempts for the huge message than are intended.
124 PH/01 Two changes to the default runtime configuration:
126 (1) Move the checks for relay_from_hosts and authenticated clients from
127 after to before the (commented out) DNS black list checks.
129 (2) Add control=submission to the relay_from_hosts and authenticated
130 clients checks, on the grounds that messages accepted by these
131 statements are most likely to be submissions.
133 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
135 (1) Generate an error if the third argument for the ${prvs expansion is
138 (2) Treat a missing third argument of ${prvscheck as if it were an empty
141 (3) Reset the variables that are obtained from the first argument of
142 ${prvscheck and used in the second argument before leaving the code,
143 because their memory is reclaimed, so using them afterwards may do
146 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
147 one (it's much easier than Tom thought :-).
149 (5) Because of (4), we can now allow for the use of $prvscheck_result
150 inside the third argument.
152 PH/03 For some reason, the default setting of PATH when running a command from
153 a pipe transport was just "/usr/bin". I have changed it to
156 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
157 anything to be listed in the output from -bV.
159 PH/05 When a filter generated an autoreply, the entire To: header line was
160 quoted in the delivery log line, like this:
162 => >A.N.Other <ano@some.domain> <original@ddress> ...
164 This has been changed so that it extracts the operative address. There
165 may be more than one such address. If so, they are comma-separated, like
168 => >ano@some.domain,ona@other.domain <original@ddress> ...
170 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
171 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
172 not being looked up in the rDNS to get a host name, Exim was showing the
173 IP address twice in Received: lines, even though the IP addresses were
174 identical. For example:
176 Received: from [1.2.3.4] (helo=[1.2.3.4])
178 However, if the real host name was known, it was omitting the HELO data
179 if it matched the actual IP address. This has been tidied up so that it
180 doesn't show the same IP address twice.
182 PH/07 When both +timestamp and +memory debugging was on, the value given by
183 $tod_xxx expansions could be wrong, because the tod_stamp() function was
184 called by the debug printing, thereby overwriting the timestamp buffer.
185 Debugging no longer uses the tod_stamp() function when +timestamp is set.
187 PH/08 When the original message was included in an autoreply transport, it
188 always said "this is a copy of the message, including all the headers",
189 even if body_only or headers_only was set. It now gives an appropriate
192 PH/09 Applied a patch from the Sieve maintainer which:
194 o fixes some comments
195 o adds the (disabled) notify extension core
196 o adds some debug output for the result of if/elsif tests
197 o points to the current vacation draft in the documentation
198 and documents the missing references header update
202 o fixes a bug in processing the envelope test (when testing
203 multiple envelope elements, the last element determinted the
206 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
207 Electronic Mail") by including:
209 Auto-submitted: auto-generated
211 in the messages that it generates (bounce messages and others, such as
212 warnings). In the case of bounce messages for non-SMTP mesages, there was
213 also a typo: it was using "Auto_submitted" (underscore instead of
214 hyphen). Since every message generated by Exim is necessarily in response
215 to another message, thes have all been changed to:
217 Auto-Submitted: auto-replied
219 in accordance with these statements in the RFC:
221 The auto-replied keyword:
223 - SHOULD be used on messages sent in direct response to another
224 message by an automatic process,
226 - MUST NOT be used on manually-generated messages,
228 - MAY be used on Delivery Status Notifications (DSNs) and Message
229 Disposition Notifications (MDNs),
231 - MUST NOT be used on messages generated by automatic or periodic
232 processes, except for messages which are automatic responses to
235 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
236 to the default Received: header definition.
238 PH/12 Added log selector acl_warn_skipped (default on).
240 PH/13 After a successful wildlsearch lookup, discard the values of numeric
241 variables because (a) they are in the wrong storage pool and (b) even if
242 they were copied, it wouldn't work properly because of the caching.
244 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
245 checking when decoding. Apparently there are clients that generate
246 overlong encoded strings. Why am I not surprised?
248 PH/15 If the first argument of "${if match_address" was not empty, but did not
249 contain an "@" character, Exim crashed. Now it writes a panic log message
250 and treats the condition as false.
252 PH/16 In autoreply, treat an empty string for "once" the same as unset.
254 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
255 extension "envelope-auth". The code is finished and in agreement with
256 other implementations, but there is no documentation so far and in fact,
257 nobody wrote the draft yet. This extension is currently #undef'ed, thus
258 not changing the active code.
260 Print executed "if" and "elsif" statements when debugging is used. This
261 helps a great deal to understand what a filter does.
263 Document more things not specified clearly in RFC3028. I had all this
264 sorted out, when out of a sudden new issues came to my mind. Oops."
266 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
269 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
270 canonical form (as documented). However, after a host name lookup from
271 the IP address, check_host() was doing a simple string comparison with
272 addresses acquired from the DNS when checking that the found name did
273 have the original IP as one of its addresses. Since any found IPv6
274 addresses are likely to be in abbreviated form, the comparison could
275 fail. Luckily, there already exists a function for doing the comparison
276 by converting both addresses to binary, so now that is used instead of
279 PH/20 There was another similar case to PH/19, when a complete host name was
280 given in a host list; looking up its IP address could give an abbreviated
281 form, whereas the current host's name might or might not be abbreviated.
282 The same fix has been applied.
288 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
289 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
292 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
293 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
295 PH/03 Typo: missing ".o" in src/pcre/Makefile.
297 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
298 header line, restrict the check to what is listed in RFCs 2369 and 2929.
299 Also, for "Auto-Submitted", treat anything other than "no" as
300 non-personal, in accordance with RFC 3834. (Previously it treated
301 anything starting "auto-" as non-personal.)
303 TF/01 The control=submission/name=... option had a problem with syntax
304 errors if the name included a slash character. The /name= option
305 now slurps the rest of the string, so it can include any characters
306 but it must come last in the list of options (after /sender_retain
309 PH/05 Some modifications to the interface to the fake nameserver for the new
317 TK/01 Added the "success_on_redirect" address verification option. See
318 NewStuff for rationale and an example.
320 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
322 PH/02 Patch to exigrep to allow it to work on syslog lines.
324 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
325 fread() to skip over the body file's header line, because in Cygwin the
326 header line is locked and is inaccessible.
328 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
329 co-exist for some time) to make it clear that it is the Exim ID that is
330 referenced, not the Message-ID: header line.
332 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
333 string_format() function, because snprintf() does not exist on all
336 PH/06 The use of forbid_filter_existstest now also locks out the use of the
337 ${stat: expansion item.
339 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
340 protocol synchronization error", to keep the pedants happy.
342 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
343 well as for IRIX systems, when gcc is being used. See the host.c source
346 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
348 PH/10 Named domain lists were not working if used in a queue_smtp_domains
351 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
352 transport and to the smtp transport in LMTP mode.
354 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
356 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
357 run a filter in a subprocess. This could lead to confusion in subsequent
358 lookups in the parent process. There should also be a search_tidyup() at
359 the end of the subprocess.
361 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
362 only if the host matched helo_try_verify_hosts, which caused the
363 verification to occur when the EHLO/HELO command was issued. The ACL just
364 tested the remembered result. Now, if a previous verification attempt has
365 not happened, "verify = helo" does it there and then.
367 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
369 TK/03 Fix log output including CR from clamd.
371 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
372 exist provoked a memory error which could cause a segfault.
374 PH/15 Installed PCRE 6.2
376 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
378 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
379 of the problem. Specifically, suggested +O2 rather than +O1 for the
382 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
384 PH/20 If a delivery was routed to a non-standard port by means of an SRV
385 record, the port was not correctly logged when the outgoing_port log
386 selector was set (it logged the transort's default port).
388 PH/21 Added support for host-specific ports to manualroute, queryprogram,
389 fallback_hosts, and "hosts" in the smtp transport.
391 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
392 host errors such as "Connection refused".
394 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
395 authentication with radiusclient 0.4.9:
397 - Error returned from rc_read_config was caught wrongly
398 - Username/password not passed on to radius server due to wrong length.
400 The presumption is that some radiusclient API changes for 4.51/PH/17
401 were not taken care of correctly. The code is still untested by me (my
402 Linux distribution still has 0.3.2 of radiusclient), but it was
403 contributed by a Radius user.
405 PH/24 When doing a callout, the value of $domain wasn't set correctly when
406 expanding the "port" option of the smtp transport.
408 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
409 while reading a MIME header. Thanks to Tom Hughes for a patch.
411 PH/24 Include config.h inside local_scan.h so that configuration settings are
414 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
415 that in an ACL for RCPT (for example), you can examine exactly what was
418 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
419 commands, but it was not correctly comparing the address with the actual
420 client host address. Thus, it would show the EHLO address in Received:
421 header lines when this was not necessary.
423 PH/27 Added the % operator to ${eval:}.
425 PH/28 Exim tries to create and chdir to its spool directory when it starts;
426 it should be ignoring failures (because with -C, for example, it has lost
427 privilege). It wasn't ignoring creation failures other than "already
430 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
433 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
434 because an input line was too long, either on its own, or by virtue of
435 too many continuations, the temporary file was not being removed, and the
436 return code was incorrect.
438 PH/31 Missing "BOOL" in function definition in filtertest.c.
440 PH/32 Applied Sieve patches from the maintainer.
442 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
444 PH/33 Added "verify = not_blind".
446 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
447 Local/Makefile (with some defaults set). These are used in built scripts
448 such as exicyclog, but they have never been used in the exim_install
449 script (though there are many overriding facilities there). I have
450 arranged that the exim_install script now takes note of these two
453 PH/35 Installed configuration files for Dragonfly.
455 PH/36 When a locally submitted message by a trusted user did not contain a
456 From: header, and the sender address was obtained from -f or from an SMTP
457 MAIL command, and the trusted user did not use -F to supply a sender
458 name, $originator_name was incorrectly used when constructing a From:
459 header. Furthermore, $originator_name was used for submission mode
460 messages from external hosts without From: headers in a similar way,
461 which is clearly wrong.
463 PH/37 Added control=suppress_local_fixups.
465 PH/38 When log_selector = +received_sender was set, and the addition of the
466 sender made the log line's construction buffer exactly full, or one byte
467 less than full, an overflow happened when the terminating "\n" was
470 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
471 when the result of a list match is failure because a DNS lookup failed.
473 PH/40 RM_COMMAND is now used in the building process.
475 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
476 the "build-* directories that it finds.
478 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
479 address in a domain literal was a prefix of an interface address.
481 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
482 when verifying a sender address, unless rewrite_headers is false.
484 PH/44 Wrote a long comment about why errors_to addresses are verified as
485 recipients, not senders.
487 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
488 the ratelimit ACL was added.
490 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
492 PH/46 Added extra information about PostgreSQL errors to the error string.
494 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
495 avoiding the need to install special zones in a real server. This is
496 backwards compatible; if it can't find the fake resolver, it drops back.
497 Thus, both old and new test suites can be run.
499 TF/02 Added util/ratelimit.pl
501 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
502 clock is set back in time.
504 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
505 Candler <B.Candler@pobox.com>.
507 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
508 OK for addresses that are the result of redirections.
510 PH/48 A number of further additions for the benefit of the new test suite,
511 including a fake gethostbyname() that interfaces to the fake DNS resolver
514 TF/06 The fix for widen_domains has also been applied to qualify_single and
515 search_parents which are the other dnslookup options that can cause
518 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
521 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
523 TF/07 Exim produced the error message "an SRV record indicated no SMTP
524 service" if it encountered an MX record with an empty target hostname.
525 The message is now "an MX or SRV record indicated no SMTP service".
527 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
528 if the DNS of the sending site is misconfigured. This is quite a
529 common situation. This change restores the behaviour of treating a
530 helo verification defer as a failure.
532 PH/51 If self=fail was set on a router, the bounce message did not include the
533 actual error message.
539 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
541 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
542 command itself ended in error, the underlying message about the transport
543 filter timeout was being overwritten with the pipe command error. Now the
544 underlying error message should be appended to the second error message.
546 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
548 PH/02 Reduce the amount of output that "make" produces by default. Full output
549 can still be requested.
551 PH/03 The warning log line about a condition test deferring for a "warn" verb
552 was being output only once per connection, rather than after each
553 occurrence (because it was using the same function as for successful
554 "warn" verbs). This seems wrong, so I have changed it.
556 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
557 it should not have, which might have caused a crash in the right
558 circumstances, but probably never did.
560 PH/04 Installed a modified version of Tony Finch's patch to make submission
561 mode fix the return path as well as the Sender: header line, and to
562 add a /name= option so that you can make the user's friendly name appear
565 TF/03 Added the control = fakedefer ACL modifier.
567 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
568 Mark Lowes for thorough testing.
570 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
572 TK/03 Merged latest SRS patch from Miles Wilton.
574 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
575 with the definition in sysexits.h (which is #included earlier).
576 Fortunately, Exim does not actually use EX_OK. The code used to try to
577 preserve the sysexits.h value, by assumimg that macro definitions were
578 scanned for macro replacements. I have been disabused of this notion,
579 so now the code just undefines EX_OK before #including unistd.h.
581 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
582 in the smtp transport. When a block could not be written in a single
583 write() function, the timeout was being re-applied to each part-write.
584 This seems wrong - if the receiver was accepting one byte at a time it
585 would take for ever. The timeout is now adjusted when this happens. It
586 doesn't have to be particularly precise.
588 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
589 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
591 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
592 without a domain if the check to <postmaster@domain> fails.
594 SC/01 Eximstats: added -xls and the ability to specify output files
595 (patch written by Frank Heydlauf).
597 SC/02 Eximstats: use FileHandles for outputing results.
599 SC/03 Eximstats: allow any combination of xls, txt, and html output.
601 SC/04 Eximstats: fixed display of large numbers with -nvr option
603 SC/05 Eximstats: fixed merging of reports with empty tables.
605 SC/06 Eximstats: added the -include_original_destination flag
607 SC/07 Eximstats: removed tabs and trailing whitespace.
609 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
611 TK/06 MBOX spool code: Add real "From " MBOX separator line
612 so the .eml file is really in mbox format (even though
613 most programs do not really care). Patch from Alex Miller.
615 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
616 The latter is generated from $received_to and is only set if the
617 message has one envelope recipient. SA can use these headers,
618 obviously out-of-the-box. Patch from Alex Miller.
620 PH/08 The ${def test on a variable was returning false if the variable's
621 value was "0", contrary to what the specification has always said!
622 The result should be true unless the variable is empty.
624 PH/09 The syntax error of a character other than { following "${if
625 def:variable_name" (after optional whitespace) was not being diagnosed.
626 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
627 accidental colon was present, for example, could give incorrect results.
629 PH/10 Tidied the code in a number of places where the st_size field of a stat()
630 result is used (not including appendfile, where other changes are about
633 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
634 This involved changing a lot of size variables from int to off_t. It
635 should work with maildirs and everything.
637 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
638 spamd dying while we are connected to it.
640 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
643 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
644 function for BDB 4.3.
646 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
648 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
649 the sources for PCRE have been re-arranged and more files are now
652 PH/15 The code I had for printing potentially long long variables in PH/11
653 above was not the best (it lost precision). The length of off_t variables
654 is now inspected at build time, and an appropriate printing format (%ld
655 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
656 to be "long long int" or "long int". This is needed for the internal
657 formatting function string_vformat().
659 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
660 the configuration file to be ":syslog", then the script "guesses" where
661 the logs files are, rather than using the compiled in default. In our
662 case the guess is not the same as the compiled default, so the script
663 suddenly stopped working when I started to use syslog. The patch checks
664 to see if log_file_path is "". If so, it attempts to read it from exim
665 with no configuration file to get the compiled in version, before it
666 falls back to the previous guessing code."
668 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
669 implementing BATV in an Exim configuration. See NewStuff for the gory
672 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
673 Makefile that are specific to HP-UX.
675 PH/18 If the "use_postmaster" option was set for a recipient callout together
676 with the "random" option, the postmaster address was used as the MAIL
677 FROM address for the random test, but not for the subsequent recipient
678 test. It is now used for both.
680 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
681 patch removes a few documentation additions to RFC 3028, because the
682 latest draft now contains them. It adds the new en;ascii-case comparator
683 and a new error check for 8bit text in MIME parts. Comparator and
684 require names are now matched exactly. I enabled the subaddress
685 extension, but it is not well tested yet (read: it works for me)."
687 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
688 rework some of the code of TK/09 above to avoid the hardwired use of
689 "%lld" and "long long". Replaced the call to snprintf() with a call to
692 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
693 records point to non-existent hosts", "retry timeout exceeded", and
694 "retry time not reached for any host after a long failure period".
696 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
697 experimental DomainKeys support:
699 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
700 (2) On an error such as an illegally used "control", the wrong name for
701 the control was given.
703 These problems did NOT occur unless DomainKeys support was compiled.
705 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
707 PH/24 Added ${if match_ip condition.
709 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
710 databases so that it will be absolutely obvious if a crash occurs in the
711 DB library. This is a regular occurrence (often caused by mis-matched
714 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
715 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
716 picked up on a user's system that detects such things. There doesn't seem
717 to be a gcc warning option for this - only an attribute that has to be
718 put on the function's prototype. It seems that in Fedora Core 4 they have
719 set this on a number of new functions. No doubt there will be more in due
722 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
723 specify a transport. However, if an address that was verified by such a
724 router was the subject of a callout, Exim crashed because it tried to
725 read the rcpt_include_affixes from the non-existent transport. Now it
726 just assumes that the setting of that option is false. This bug was
727 introduced by 4.51/PH/31.
729 PH/28 Changed -d+all to exclude +memory, because that information is very
730 rarely of interest, but it makes the output a lot bigger. People tend to
731 do -d+all out of habit.
733 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
734 code in os-type was giving problems when libc.so lives in lib64, like on
737 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
738 aren't the modern standard, and it seems that some systems' include files
739 don't always have them. Exim was already checking for some of the newer
740 ones like T_AAAA, and defining it itself. I've added checks for all the
741 record types that Exim uses.
743 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
744 not automatically generating a new one, as it is supposed to. This
745 prevented TLS from working. If the file did exist, but contained invalid
746 data, a new version was generated, as expected. It was only the case of a
747 non-existent file that was broken.
749 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
750 with a change in libdomainkeys > 0.64.
752 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
753 from DNS. If the selector record carries the flag, it now has
754 precedence over the domain-wide flag.
756 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
758 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
759 the use of a port name that isn't defined in /etc/services) occurred, the
760 message was deferred as in a normal delivery, and thus remained on the
761 spool, instead of being failed because of the mua_wrapper setting. This
762 is now fixed, and I tidied up some of the mua_wrapper messages at the
765 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
766 the messages in a hash of arrays rather than using individual hashes.
767 This is a bit cleaner and results in dramatic memory savings, albeit
768 at a slight CPU cost.
770 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
771 as requested by Marc Sherman.
773 SC/10 Eximstats: added histograms for user specified patterns as requested
776 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
778 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
779 fopen() in the content-scanning modules that did not already have it.
785 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
786 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
788 TK/02 Fix ACL "control" statement not being available in MIME ACL.
790 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
792 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
793 to test Sieve filters that use "vacation".
795 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
796 that changes the way the GnuTLS parameters are stored in the cache file.
797 The new format can be generated externally. For backward compatibility,
798 if the data in the cache doesn't make sense, Exim assumes it has read an
799 old-format file, and it generates new data and writes a new file. This
800 means that you can't go back to an older release without removing the
803 PH/03 A redirect router that has both "unseen" and "one_time" set does not
804 work if there are any delivery delays because "one_time" forces the
805 parent to be marked "delivered", so its unseen clone is never tried
806 again. For this reason, Exim now forbids the simultaneous setting of
809 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
810 redirected to themselves ("homonym" addresses). Read the long ChangeLog
811 entry if you want to know the details. The fix, however, neglected to
812 consider the case when local delivery batching is involved. The test for
813 "previously delivered" was not happening when checking to see if an
814 address could be batched with a previous (undelivered) one; under
815 certain circumstances this could lead to multiple deliveries to the same
818 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
819 in its include files, and this causes problems building Exim.
821 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
822 header_syntax) but Exim was just ignoring anything given after a slash.
823 In particular, this caused confusion with an attempt to use "verify =
824 reverse_host_lookup/defer_ok". An error is now given when options are
825 supplied for verify items that do not have them. (Maybe reverse_host_
826 lookup should have a defer_ok option, but that's a different point.)
828 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
829 defined by RFC 821) to 2048, because there were problems with some AUTH
830 commands, and RFC 1869 says the size should be increased for extended
831 SMTP commands that take arguments.
833 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
836 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
837 "unknown" error; now it says that the functionality isn't in the binary.
839 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
840 an address' error message when a string expansion fails (syntax or
841 whatever). Otherwise the password may appear in the log. Following change
842 PH/42 below, there is no longer a chance of it appearing in a bounce
845 PH/11 Installed exipick version 20050225.0 from John Jetmore.
847 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
848 of its addresses was ever tried. (Bugzilla bug #2.)
850 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
851 the result incorrectly in the debug output. (It correctly added a newline
852 to what was transported.)
854 TF/01 Added $received_time.
856 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
857 commented out examples of how to interface to a virus scanner and to
858 SpamAssassin. Also added commented examples of av_scanner and
859 spamd_address settings.
861 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
862 and controls are allowed in which ACLs. There were a couple of minor
863 errors. Some of the entries in the conditions table (which is a table of
864 where they are NOT allowed) were getting very unwieldy; rewrote them as a
865 negation of where the condition IS allowed.
867 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
869 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
870 header file does not have a version number, so I've had to invent a new
871 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
872 API. The code is untested by me (my Linux distribution still has 0.3.2 of
873 radiusclient), but it was contributed by a Radius user.
875 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
876 files or directories, for OpenSSL.
878 PH/19 When an Exim process that is running as root has to create an Exim log
879 file, it does so in a subprocess that runs as exim:exim so as to get the
880 ownership right at creation (otherwise, other Exim processes might see
881 the file with the wrong ownership). There was no test for failure of this
882 fork() call, which would lead to the process getting stuck as it waited
883 for a non-existent subprocess. Forks do occasionally fail when resources
884 run out. I reviewed all the other calls to fork(); they all seem to check
887 PH/20 When checking for unexpected SMTP input at connect time (before writing
888 the banner), Exim was not dealing correctly with a non-positive return
889 from the read() function. If the client had disconnected by this time,
890 the result was a log entry for a synchronization error with an empty
891 string after "input=" when read() returned zero. If read() returned -1
892 (an event I could not check), uninitialized data bytes were printed.
893 There were reports of junk text (parts of files, etc) appearing after
896 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
898 PH/22 Added support for macro redefinition, and (re)definition in between
899 driver and ACL definitions.
901 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
902 forgetting to use the resulting value; it was using the unexpanded value.
904 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
905 hadn't been configured. The fix is from Juergen Kreileder, who
906 understands it better than I do:
908 "Here's what I see happening with three configured cyrus_sasl
909 authenticators configured (plain, login, cram-md5):
911 On startup auth_cyrus_sasl_init() gets called for each of these.
912 This means three calls to sasl_listmech() without a specified mech_list.
913 => SASL tests which mechs of all available mechs actually work
914 => three warnings about OTP not working
915 => the returned list contains: plain, login, cram-md5, digest-md5, ...
917 With the patch, sasl_listmech() also gets called three times. But now
918 SASL's mech_list option is set to the server_mech specified in the the
919 authenticator. Or in other words, the answer from sasl_listmech()
920 gets limited to just the mech you're testing for (which is different
922 => the return list contains just 'plain' or 'login', 'cram-md5' or
923 nothing depending on the value of ob->server_mech.
925 I've just tested the patch: Authentication still works fine,
926 unavailable mechs specified in the exim configuration are still
927 caught, and the auth.log warnings about OTP are gone."
929 PH/25 When debugging is enabled, the contents of the command line are added
930 to the debugging output, even when log_selector=+arguments is not
933 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
934 answer is "GNU", and only if the return is "GNU/something" is the answer
937 PH/27 $acl_verify_message is now set immediately after the failure of a
938 verification in an ACL, and so is available in subsequent modifiers. In
939 particular, the message can be preserved by coding like this:
941 warn !verify = sender
942 set acl_m0 = $acl_verify_message
944 Previously, $acl_verify_message was set only while expanding "message"
945 and "log_message" when a very denied access.
947 PH/28 Modified OS/os.c-Linux with
949 -#ifndef OS_LOAD_AVERAGE
950 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
952 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
953 nomenclature these days.)
955 PH/29 Installed patch from the Sieve maintainer that adds the options
956 sieve_useraddress and sieve_subaddress to the redirect router.
958 PH/30 In these circumstances:
959 . Two addresses routed to the same list of hosts;
960 . First host does not offer TLS;
961 . First host accepts first address;
962 . First host gives temporary error to second address;
963 . Second host offers TLS and a TLS session is established;
964 . Second host accepts second address.
965 Exim incorrectly logged both deliveries with the TLS parameters (cipher
966 and peerdn, if requested) that were in fact used only for the second
969 PH/31 When doing a callout as part of verifying an address, Exim was not paying
970 attention to any local part prefix or suffix that was matched by the
971 router that accepted the address. It now behaves in the same way as it
972 does for delivery: the affixes are removed from the local part unless
973 rcpt_include_affixes is set on the transport.
975 PH/32 Add the sender address, as F=<...>, to the log line when logging a
976 timeout during the DATA phase of an incoming message.
978 PH/33 Sieve envelope tests were broken for match types other than :is. I have
979 applied a patch sanctioned by the Sieve maintainer.
981 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
982 the uid or gid is negative. A case of a negative gid caused this to be
983 noticed. The fix allows for either to be negative.
985 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
986 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
989 PH/36 The change PH/12 above was broken. Fixed it.
991 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
992 the grounds that routing the same address twice would always produce the
993 same answer. This might have been true once, but it is certainly no
994 longer true now. Routing a child address may depend on the previous
995 routing that produced that child. Some complicated redirection strategies
996 went wrong when messages had multiple recipients, and made Exim's
997 behaviour dependent on the order in which the addresses were given.
999 I have moved the duplicate checking until after the routing is complete.
1000 Exim scans the addresses that are assigned to local and remote
1001 transports, and removes any duplicates. This means that more work will be
1002 done, as duplicates will always all be routed, but duplicates are
1003 presumably rare, so I don't expect this is of any significance.
1005 For deliveries to pipes, files, and autoreplies, the duplicate checking
1006 still happens during the routing process, since they are not going to be
1009 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1010 It corrects a timeout issue with spamd. This is Ian's comment: "The
1011 background is that sometimes spamd either never reads data from a
1012 connection it has accepted, or it never writes response data. The exiscan
1013 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1014 blindly assumes that writes won't block so it may never time out."
1016 PH/39 Allow G after quota size as well as K and M.
1018 PH/40 The value set for $authenticated_id in an authenticator may not contain
1019 binary zeroes or newlines because the value is written to log lines and
1020 to spool files. There was no check on this. Now the value is run through
1021 the string_printing() function so that such characters are converted to
1022 printable escape sequences.
1024 PH/41 $message_linecount is a new variable that contains the total number of
1025 lines in the message. Compare $body_linecount, which is the count for the
1028 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1029 bounce and delay warning messages, except in certain special cases, which
1032 (a) An SMTP error message from a remote host;
1033 (b) A message specified in a :fail: redirection;
1034 (c) A message specified in a "fail" command in a system filter;
1035 (d) A message specified in a FAIL return from the queryprogram router;
1036 (e) A message specified by the cannot_route_message router option.
1038 In these cases only, Exim does include the error details in bounce and
1039 warning messages. There are also a few cases where bland messages such
1040 as "unrouteable address" or "local delivery error" are given.
1042 PH/43 $value is now also set for the "else" part of a ${run expansion.
1044 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1045 being worked on, but at least Exim now implements the latest version to
1048 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1049 process to complete was treated as a delivery failure, a timeout while
1050 writing the message to the pipe was logged, but erroneously treated as a
1051 successful delivery. Such timeouts include transport filter timeouts. For
1052 consistency with the overall process timeout, these timeouts are now
1053 treated as errors, giving rise to delivery failures by default. However,
1054 there is now a new Boolean option for the pipe transport called
1055 timeout_defer, which, if set TRUE, converts the failures into defers for
1056 both kinds of timeout. A transport filter timeout is now identified in
1059 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1060 systems where "make" and "gmake" are different, calling "gmake" at top
1061 level broke things. I've arranged for the value of $(MAKE) to be passed
1062 from the Makefile to this script so that it can call the same version of
1066 A note about Exim versions 4.44 and 4.50
1067 ----------------------------------------
1069 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1070 changes of various kinds. As a consequence, a big documentation update was
1071 needed. This delayed the release for rather longer than seemed good, especially
1072 in the light of a couple of (minor) security issues. Therefore, the changes
1073 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1074 release. So 4.44 and 4.50 are in effect two different branches that both start
1077 I have left the 4.50 change log unchanged; it contains all the changes since
1078 4.43. The change log for 4.44 is below; many of its items are identical to
1079 those for 4.50. This seems to be the most sensible way to preserve the
1080 historical information.
1086 1. Minor wording change to the doc/README.SIEVE file.
1088 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1089 computation of the current number of files was incorrect.
1091 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1092 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1095 4. Give more explanation in the error message when the command for a transport
1096 filter fails to execute.
1098 5. There are several places where Exim runs a non-Exim command in a
1099 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1100 was being done only for the command run by the queryprogram router. It is
1101 now done for all such subprocesses. The other cases are: ${run, transport
1102 filters, and the commands run by the lmtp and pipe transports.
1104 6. Added CONFIGURE_GROUP build-time option.
1106 7. Some older OS have a limit of 256 on the maximum number of file
1107 descriptors. Exim was using setrlimit() to set 1000 as a large value
1108 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1109 systems. I've change it so that if it can't get 1000, it tries for 256.
1111 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1112 was an oversight, and furthermore, ever since the addition of extra
1113 controls (e.g. 4.43/32), the checks on when to allow different forms of
1114 "control" were broken. There should now be diagnostics for all cases when a
1115 control that does not make sense is encountered.
1117 9. Added the /retain_sender option to "control=submission".
1119 10. $recipients is now available in the predata ACL (oversight).
1121 11. Tidy the search cache before the fork to do a delivery from a message
1122 received from the command line. Otherwise the child will trigger a lookup
1123 failure and thereby defer the delivery if it tries to use (for example) a
1124 cached ldap connection that the parent has called unbind on.
1126 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1127 of $address_data from the recipient verification was clobbered by the
1128 sender verification.
1130 13. The value of address_data from a sender verification is now available in
1131 $sender_address_data in subsequent conditions in the ACL statement.
1133 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1135 15. Added a new option "connect=<time>" to callout options, to set a different
1138 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1139 was its contents. (It was OK if the option was not defined at all.)
1141 17. A "Completed" log line is now written for messages that are removed from
1142 the spool by the -Mrm option.
1144 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1145 information about exactly what failed.
1147 19. Added -dd to debug only the daemon process.
1149 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1150 handles timeouts, both on the server side and network timeouts. Renamed the
1151 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1153 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1154 It is now set to "smtps".
1156 22. $host_address is now set to the target address during the checking of
1157 ignore_target_hosts.
1159 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1160 being passed; this would have caused $sender_host_name to have been used if
1161 matching the list had actually called for a host name (not very likely,
1162 since this list is usually IP addresses). A host name is now passed as
1165 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1166 code that specifies a non-restarting handler (typically sigaction() in
1167 modern systems) in an attempt to fix a rare and obscure crash bug.
1169 25. Narrowed the window for a race in the daemon that could cause it to ignore
1170 SIGCHLD signals. This is not a major problem, because they are used only to
1171 wake it up if nothing else does.
1173 26. A malformed maildirsize file could cause Exim to calculate negative values
1174 for the mailbox size or file count. Odd effects could occur as a result.
1175 The maildirsize information is now recalculated if the size or filecount
1178 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1179 support for a long time. Removed HAVE_SYS_VFS_H.
1181 28. Installed the latest version of exipick from John Jetmore.
1183 29. In an address list, if the pattern was not a regular expression, an empty
1184 subject address (from a bounce message) matched only if the pattern was an
1185 empty string. Non-empty patterns were not even tested. This was the wrong
1186 because it is perfectly reasonable to use an empty address as part of a
1187 database query. An empty address is now tested by patterns that are
1188 lookups. However, all the other forms of pattern expect the subject to
1189 contain a local part and a domain, and therefore, for them, an empty
1190 address still always fails if the pattern is not itself empty.
1192 30. Exim went into a mad DNS loop when attempting to do a callout where the
1193 host was specified on an smtp transport, and looking it up yielded more
1194 than one IP address.
1196 31. Re-factored the code for checking spool and log partition space into a
1197 function that finds that data and another that does the check. The former
1198 is then used to implement four new variables: $spool_space, $log_space,
1199 $spool_inodes, and $log_inodes.
1201 32. The RFC2047 encoding function was originally intended for short strings
1202 such as real names; it was not keeping to the 75-character limit for
1203 encoded words that the RFC imposes. It now respects the limit, and
1204 generates multiple encoded words if necessary. To be on the safe side, I
1205 have increased the buffer size for the ${rfc2047: expansion operator from
1208 33. It is now permitted to omit both strings after an "if" condition; if the
1209 condition is true, the result is "true". As before, when the second string
1210 is omitted, a false condition yields an empty string. This makes it less
1211 cumbersome to write custom ACL and router conditions.
1213 34. Failure to deliver a bounce message always caused it to be frozen, even if
1214 there was an errors_to setting on the router. The errors_to setting is now
1217 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1218 canonical form (fully expanded) before being placed in
1219 $sender_host_address.
1221 36. The table in the code that translates DNS record types into text (T_A to
1222 "A" for instance) was missing entries for NS and CNAME. It is just possible
1223 that this could have caused confusion if both these types were looked up
1224 for the same domain, because the text type is used as part of Exim's
1225 per-process caching. But the chance of anyone hitting this buglet seems
1228 37. The dnsdb lookup has been extended in a number of ways.
1230 (1) There is a new type, "zns", which walks up the domain tree until it
1231 finds some nameserver records. It should be used with care.
1233 (2) There is a new type, "mxh", which is like "mx" except that it returns
1234 just the host names, not the priorities.
1236 (3) It is now possible to give a list of domains (or IP addresses) to be
1237 looked up. The behaviour when one of the lookups defers can be
1238 controlled by a keyword.
1240 (4) It is now possible to specify the separator character for use when
1241 multiple records are returned.
1243 38. The dnslists ACL condition has been extended: it is now possible to supply
1244 a list of IP addresses and/or domains to be looked up in a particular DNS
1247 39. Added log_selector=+queue_time_overall.
1249 40. When running the queue in the test harness, wait just a tad after forking a
1250 delivery process, to get repeatability of debugging output.
1252 41. Include certificate and key file names in error message when GnuTLS fails
1253 to set them up, because the GnuTLS error message doesn't include the name
1254 of the failing file when there is a problem reading it.
1256 42. Allow both -bf and -bF in the same test run.
1258 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1260 44. The "Exiscan patch" is now merged into the mainline Exim source.
1262 45. Sometimes the final signoff response after QUIT could fail to get
1263 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1264 < 0 before doing a fflush(). This bug looks as though it goes back to the
1265 introduction of TLS in release 3.20, but "sometimes" must have been rare
1266 because the tests only now provoked it.
1268 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1269 (this can affect the format of dates).
1271 47. exim_tidydb, when checking for the continued existence of a message for
1272 which it has found a message-specific retry record, was not finding
1273 messages that were in split spool directories. Consequently, it was
1274 deleting retry records that should have stayed in existence.
1276 48. Steve fixed some bugs in eximstats.
1278 49. The SPA authentication driver was not abandoning authentication and moving
1279 on to the next authenticator when an expansion was forced to fail,
1280 contradicting the general specification for all authenticators. Instead it
1281 was generating a temporary error. It now behaves as specified.
1283 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1284 (the order specifies the preference for clients). The order is now AES256,
1285 AES128, 3DES, ARCFOUR128.
1287 51. Small patch to Sieve code - explicitly set From: when generating an
1290 52. Exim crashed if a remote delivery caused a very long error message to be
1291 recorded - for instance if somebody sent an entire SpamAssassin report back
1292 as a large number of 550 error lines. This bug was coincidentally fixed by
1293 increasing the size of one of Exim's internal buffers (big_buffer) that
1294 happened as part of the Exiscan merge. However, to be on the safe side, I
1295 have made the code more robust (and fixed the comments that describe what
1298 53. Now that there can be additional text after "Completed" in log lines (if
1299 the queue_time_overall log selector is set), a one-byte patch to exigrep
1300 was needed to allow it to recognize "Completed" as not the last thing in
1303 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1304 patch that reportedly fixes this has been added. I am not expert enough to
1305 create a test for it. This is what the patch creator wrote:
1307 "I found a little strange behaviour of ldap code when working with
1308 Windows 2003 AD Domain, where users was placed in more than one
1309 Organization Units. When I tried to give exim partial DN, the exit code
1310 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1311 But simultaneously result of request was absolutely normal ldap result,
1312 so I produce this patch..."
1314 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1315 REFERENCE, so I have modified the code to exclude the patch when that macro
1318 55. Some experimental protocols are using DNS PTR records for new purposes. The
1319 keys for these records are domain names, not reversed IP addresses. The
1320 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1321 leaves it alone. Component reversal etc. now happens only for IP addresses.
1324 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1326 57. Double the size of the debug message buffer (to 2048) so that more of very
1327 long debug lines gets shown.
1329 58. The exicyclog utility now does better if the number of log files to keep
1330 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1332 59. Two changes related to the smtp_active_hostname option:
1334 (1) $smtp_active_hostname is now available as a variable.
1335 (2) The default for smtp_banner uses $smtp_active_hostname instead
1336 of $primary_hostname.
1338 60. The host_aton() function is supposed to be passed a string that is known
1339 to be a valid IP address. However, in the case of IPv6 addresses, it was
1340 not checking this. This is a hostage to fortune. Exim now panics and dies
1341 if the condition is not met. A case was found where this could be provoked
1342 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1343 components; fortuitously, this particular loophole had already been fixed
1344 by change 4.50/55 above.
1346 If there are any other similar loopholes, the new check in host_aton()
1347 itself should stop them being exploited. The report I received stated that
1348 data on the command line could provoke the exploit when Exim was running as
1349 exim, but did not say which command line option was involved. All I could
1350 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1351 running as the user.
1354 61. There was a buffer overflow vulnerability in the SPA authentication code
1355 (which came originally from the Samba project). I have added a test to the
1356 spa_base64_to_bits() function which I hope fixes it.
1359 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1360 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1362 63. The daemon start-up calls getloadavg() while still root for those OS that
1363 need the first call to be done as root, but it missed one case: when
1364 deliver_queue_load_max is set with deliver_drop_privilege. This is
1365 necessary for the benefit of the queue runner, because there is no re-exec
1366 when deliver_drop_privilege is set.
1368 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1369 This has been fixed.
1371 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1372 was in use, was not putting the data itself into the right store pool;
1373 consequently, it could be overwritten for a subsequent message in the same
1374 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1377 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1379 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1380 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1381 the function were treating the return as a boolean value, which happened to
1382 work because 0=false and not-0=true, but is not correct code.
1384 68. The host_aton() function was not handling scoped IPv6 addresses (those
1385 with, for example, "%eth0" on the end) correctly.
1387 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1388 negated items (that is, ~something) in unsigned ints. Some compilers
1389 apparently mutter when there is no cast.
1391 70. If an address verification called from an ACL failed, and did not produce a
1392 user-specific message (i.e. there was only a "system" message), nothing was
1393 put in $acl_verify_message. In this situation, it now puts the system
1396 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1397 session; change 4.31/43 added the unwanted input to the log line - except
1398 that it did not do this in the start of session case. It now does.
1400 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1401 This is wrong and can cause the other end to generate a synchronization
1402 error if it is another Exim or anything else that does the synchronization
1403 check. A QUIT command is no longer sent after a timeout.
1405 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1406 during host lookups.
1408 74. The defer_ok option of callout verification was not working if it was used
1409 when verifying addresses in header lines, that is, for this case:
1411 verify = header_sender/callout=defer_ok
1413 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1414 those file descriptors could be used for SMTP connections. If anything
1415 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1416 could be sent to the SMTP client, causing chaos. The daemon now opens
1417 stdin, stdout, and stderr to /dev/null when it puts itself into the
1420 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1421 log by default. The user can override this with suitable Perl magic.
1423 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1424 the log message when discard triggers, was not working for the DATA ACL or
1425 for the non-SMTP ACL.
1427 78. Error message wording change in sieve.c.
1429 79. If smtp_accept_max_per_host was set, the number of connections could be
1430 restricted to fewer than expected, because the daemon was trying to set up
1431 a new connection before checking whether the processes handling previous
1432 connections had finished. The check for completed processes is now done
1433 earlier. On busy systems, this bug wouldn't be noticed because something
1434 else would have woken the daemon, and it would have reaped the completed
1437 80. If a message was submitted locally by a user whose login name contained one
1438 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1439 It caused a spool format error. I have fixed the spool reading code. A
1440 related problem was that the "from" clause in the Received: line became
1441 illegal because of the space(s). It is now covered by ${quote_local_part.
1443 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1446 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1448 83. Patch from Sieve maintainer to fix unterminated string problem in
1449 "vacation" handling.
1451 84. Some minor changes to the Linux configuration files to help with other
1452 OS variants using glibc.
1454 85. One more patch for Sieve to update vacation handling to latest spec.
1457 ----------------------------------------------------
1458 See the note above about the 4.44 and 4.50 releases.
1459 ----------------------------------------------------
1465 1. Change 4.43/35 introduced a bug that caused file counts to be
1466 incorrectly computed when quota_filecount was set in an appendfile
1469 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1470 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1473 3. Give more explanation in the error message when the command for a transport
1474 filter fails to execute.
1476 4. There are several places where Exim runs a non-Exim command in a
1477 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1478 was being done only for the command run by the queryprogram router. It is
1479 now done for all such subprocesses. The other cases are: ${run, transport
1480 filters, and the commands run by the lmtp and pipe transports.
1482 5. Some older OS have a limit of 256 on the maximum number of file
1483 descriptors. Exim was using setrlimit() to set 1000 as a large value
1484 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1485 systems. I've change it so that if it can't get 1000, it tries for 256.
1487 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1488 was an oversight, and furthermore, ever since the addition of extra
1489 controls (e.g. 4.43/32), the checks on when to allow different forms of
1490 "control" were broken. There should now be diagnostics for all cases when a
1491 control that does not make sense is encountered.
1493 7. $recipients is now available in the predata ACL (oversight).
1495 8. Tidy the search cache before the fork to do a delivery from a message
1496 received from the command line. Otherwise the child will trigger a lookup
1497 failure and thereby defer the delivery if it tries to use (for example) a
1498 cached ldap connection that the parent has called unbind on.
1500 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1501 of $address_data from the recipient verification was clobbered by the
1502 sender verification.
1504 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1505 was its contents. (It was OK if the option was not defined at all.)
1507 11. A "Completed" log line is now written for messages that are removed from
1508 the spool by the -Mrm option.
1510 12. $host_address is now set to the target address during the checking of
1511 ignore_target_hosts.
1513 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1514 being passed; this would have caused $sender_host_name to have been used if
1515 matching the list had actually called for a host name (not very likely,
1516 since this list is usually IP addresses). A host name is now passed as
1519 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1520 code that specifies a non-restarting handler (typically sigaction() in
1521 modern systems) in an attempt to fix a rare and obscure crash bug.
1523 15. Narrowed the window for a race in the daemon that could cause it to ignore
1524 SIGCHLD signals. This is not a major problem, because they are used only to
1525 wake it up if nothing else does.
1527 16. A malformed maildirsize file could cause Exim to calculate negative values
1528 for the mailbox size or file count. Odd effects could occur as a result.
1529 The maildirsize information is now recalculated if the size or filecount
1532 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1533 support for a long time. Removed HAVE_SYS_VFS_H.
1535 18. Updated exipick to current release from John Jetmore.
1537 19. Allow an empty sender to be matched against a lookup in an address list.
1538 Previously the only cases considered were a regular expression, or an
1541 20. Exim went into a mad DNS lookup loop when doing a callout where the
1542 host was specified on the transport, if the DNS lookup yielded more than
1545 21. The RFC2047 encoding function was originally intended for short strings
1546 such as real names; it was not keeping to the 75-character limit for
1547 encoded words that the RFC imposes. It now respects the limit, and
1548 generates multiple encoded words if necessary. To be on the safe side, I
1549 have increased the buffer size for the ${rfc2047: expansion operator from
1552 22. Failure to deliver a bounce message always caused it to be frozen, even if
1553 there was an errors_to setting on the router. The errors_to setting is now
1556 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1557 canonical form (fully expanded) before being placed in
1558 $sender_host_address.
1560 24. Updated eximstats to version 1.33
1562 25. Include certificate and key file names in error message when GnuTLS fails
1563 to set them up, because the GnuTLS error message doesn't include the name
1564 of the failing file when there is a problem reading it.
1566 26. Expand error message when OpenSSL has problems setting up cert/key files.
1569 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1570 (this can affect the format of dates).
1572 28. exim_tidydb, when checking for the continued existence of a message for
1573 which it has found a message-specific retry record, was not finding
1574 messages that were in split spool directories. Consequently, it was
1575 deleting retry records that should have stayed in existence.
1577 29. eximstats updated to version 1.35
1578 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1579 1.35 - bugfix such that pie charts by volume are generated correctly
1581 30. The SPA authentication driver was not abandoning authentication and moving
1582 on to the next authenticator when an expansion was forced to fail,
1583 contradicting the general specification for all authenticators. Instead it
1584 was generating a temporary error. It now behaves as specified.
1586 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1587 (the order specifies the preference for clients). The order is now AES256,
1588 AES128, 3DES, ARCFOUR128.
1590 31. Small patch to Sieve code - explicitly set From: when generating an
1593 32. Exim crashed if a remote delivery caused a very long error message to be
1594 recorded - for instance if somebody sent an entire SpamAssassin report back
1595 as a large number of 550 error lines. This bug was coincidentally fixed by
1596 increasing the size of one of Exim's internal buffers (big_buffer) that
1597 happened as part of the Exiscan merge. However, to be on the safe side, I
1598 have made the code more robust (and fixed the comments that describe what
1601 33. Some experimental protocols are using DNS PTR records for new purposes. The
1602 keys for these records are domain names, not reversed IP addresses. The
1603 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1604 leaves it alone. Component reversal etc. now happens only for IP addresses.
1607 34. The host_aton() function is supposed to be passed a string that is known
1608 to be a valid IP address. However, in the case of IPv6 addresses, it was
1609 not checking this. This is a hostage to fortune. Exim now panics and dies
1610 if the condition is not met. A case was found where this could be provoked
1611 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1612 components; fortuitously, this particular loophole had already been fixed
1613 by change 4.50/55 or 4.44/33 above.
1615 If there are any other similar loopholes, the new check in host_aton()
1616 itself should stop them being exploited. The report I received stated that
1617 data on the command line could provoke the exploit when Exim was running as
1618 exim, but did not say which command line option was involved. All I could
1619 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1620 running as the user.
1623 35. There was a buffer overflow vulnerability in the SPA authentication code
1624 (which came originally from the Samba project). I have added a test to the
1625 spa_base64_to_bits() function which I hope fixes it.
1628 36. The daemon start-up calls getloadavg() while still root for those OS that
1629 need the first call to be done as root, but it missed one case: when
1630 deliver_queue_load_max is set with deliver_drop_privilege. This is
1631 necessary for the benefit of the queue runner, because there is no re-exec
1632 when deliver_drop_privilege is set.
1634 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1635 was in use, was not putting the data itself into the right store pool;
1636 consequently, it could be overwritten for a subsequent message in the same
1637 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1640 38. Sometimes the final signoff response after QUIT could fail to get
1641 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1642 < 0 before doing a fflush(). This bug looks as though it goes back to the
1643 introduction of TLS in release 3.20, but "sometimes" must have been rare
1644 because the tests only now provoked it.
1650 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1651 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1652 Now it returns an integer. A number of calls were still expecting a T/F
1653 return. Fortuitously, in all cases, the tests worked in OK situations,
1654 which is the norm. However, things would have gone wrong on any write
1655 failures on the smtp file descriptor. This function is used when sending
1656 messages over SMTP and also when doing verify callouts.
1658 2. When Exim is called to do synchronous delivery of a locally submitted
1659 message (the -odf or -odi options), it no longer closes stderr before doing
1662 3. Implemented the mua_wrapper option.
1664 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1666 5. Implemented the functions header_remove(), header_testname(),
1667 header_add_at_position(), and receive_remove_recipient(), and exported them
1670 6. If an ACL "warn" statement specified the addition of headers, Exim already
1671 inserted X-ACL-Warn: at the start if there was no header name. However, it
1672 was not making this test for the second and subsequent header lines if
1673 there were newlines in the string. This meant that an invalid header could
1674 be inserted if Exim was badly configured.
1676 7. Allow an ACL "warn" statement to add header lines at the start or after all
1677 the Received: headers, as well as at the end.
1679 8. Added the rcpt_4xx retry error code.
1681 9. Added postmaster_mailfrom=xxx to callout verification option.
1683 10. Added mailfrom=xxxx to the callout verification option, for verify=
1686 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1687 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1689 12. Inserted some casts to stop certain compilers warning when using pointer
1690 differences as field lengths or precisions in printf-type calls (mostly
1691 affecting debugging statements).
1693 13. Added optional readline() support for -be (dynamically loaded).
1695 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1696 same clock tick as a message's arrival, so that its received time was the
1697 same as the "first fail" time on the retry record, and that message
1698 remained on the queue past the ultimate address timeout, every queue runner
1699 would try a delivery (because it was past the ultimate address timeout) but
1700 after another failure, the ultimate address timeout, which should have then
1701 bounced the address, did not kick in. This was a "< instead of <=" error;
1702 in most cases the first failure would have been in the next clock tick
1703 after the received time, and all would be well.
1705 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1706 being recognized when the domain list was tested by the match_domain
1707 condition in an expansion string.
1709 16. Added the ${str2b64: operator.
1711 17. Exim was always calling setrlimit() to set a large limit for the number of
1712 processes, without checking whether the existing limit was already
1713 adequate. (It did check for the limit on file descriptors.) Furthermore,
1714 errors from getrlimit() and setrlimit() were being ignored. Now they are
1715 logged to the main and panic logs, but Exim does carry on, to try to do its
1716 job under whatever limits there are.
1718 18. Imported PCRE 5.0.
1720 19. Trivial typo in log message " temporarily refused connection" (the leading
1723 20. If the log selector return_path_on_delivery was set and an address was
1724 redirected to /dev/null, the delivery process crashed because it assumed
1725 that a return path would always be set for a "successful" delivery. In this
1726 case, the whole delivery is bypassed as an optimization, and therefore no
1729 21. Internal re-arrangement: the function for sending a challenge and reading
1730 a response while authentication was assuming a zero-terminated challenge
1731 string. It's now changed to take a pointer and a length, to allow for
1732 binary data in such strings.
1734 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1736 23. Exim was not respecting finduser_retries when seeking the login of the
1737 uid under which it was called; it was always trying 10 times. (The default
1738 setting of finduser_retries is zero.) Also, it was sleeping after the final
1739 failure, which is pointless.
1741 24. Implemented tls_on_connect_ports.
1743 25. Implemented acl_smtp_predata.
1745 26. If the domain in control=submission is set empty, Exim assumes that the
1746 authenticated id is a complete email address when it generates From: or
1747 Sender: header lines.
1749 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1750 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1751 chown and chgrp in /bin and hostname in /usr/bin.
1753 28. Exim was keeping the "process log" file open after each use, just as it
1754 does for the main log. This opens the possibility of it remaining open for
1755 long periods when the USR1 signal hits a daemon. Occasional processlog
1756 errors were reported, that could have been caused by this. Anyway, it seems
1757 much more sensible not to leave this file open at all, so that is what now
1760 29. The long-running daemon process does not normally write to the log once it
1761 has entered its main loop, and it closes the log before doing so. This is
1762 so that log files can straightforwardly be renamed and moved. However,
1763 there are a couple of unusual error situations where the daemon does write
1764 log entries, and I had neglected to close the log afterwards.
1766 30. The text of an SMTP error response that was received during a remote
1767 delivery was being truncated at 512 bytes. This is too short for some of
1768 the long messages that one sometimes sees. I've increased the limit to
1771 31. It is now possible to make retry rules that apply only when a message has a
1772 specific sender, in particular, an empty sender.
1774 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1775 it possible to be selective about when SMTP synchronization is enforced.
1777 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1779 32. Implemented hosts_connection_nolog.
1781 33. Added an ACL for QUIT.
1783 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1786 35. Added mailbox_size and mailbox_filecount to appendfile.
1788 36. Added control = no_multiline_responses to ACLs.
1790 37. There was a bug in the logic of the code that waits for the clock to tick
1791 in the case where the clock went backwards by a substantial amount such
1792 that the microsecond fraction of "now" was more than the microsecond
1793 fraction of "then" (but the whole seconds number was less).
1795 38. Added support for the libradius Radius client library this is found on
1796 FreeBSD (previously only the radiusclient library was supported).
1802 1. When certain lookups returned multiple values in the form name=value, the
1803 quoting of the values was not always being done properly. Specifically:
1804 (a) If the value started with a double quote, but contained no whitespace,
1806 (b) If the value contained whitespace other than a space character (i.e.
1807 tabs or newlines or carriage returns) it was not quoted.
1808 This fix has been applied to the mysql and pgsql lookups by writing a
1809 separate quoting function and calling it from the lookup code. The fix
1810 should probably also be applied to nisplus, ibase and oracle lookups, but
1811 since I cannot test any of those, I have not disturbed their existing code.
1813 2. A hit in the callout cache for a specific address caused a log line with no
1814 reason for rejecting RCPT. Now it says "Previous (cached) callout
1815 verification failure".
1817 3. There was an off-by-one bug in the queryprogram router. An over-long
1818 return line was truncated at 256 instead of 255 characters, thereby
1819 overflowing its buffer with the terminating zero. As well as fixing this, I
1820 have increased the buffer size to 1024 (and made a note to document this).
1822 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1823 when Exim is waiting for an SMTP response from a remote server, Exim
1824 restarts its select() call on the socket, thereby resetting its timeout.
1825 This is not a problem when such interrupts are rare. Somebody set up a cron
1826 job to run exiwhat every 2 minutes, which is less than the normal select()
1827 timeout (5 or 10 minutes). This meant that the select() timeout never
1828 kicked in because it was always reset. I have fixed this by comparing the
1829 time when an interrupt arrives with the time at the start of the first call
1830 to select(). If more time than the timeout has elapsed, the interrupt is
1831 treated as a timeout.
1833 5. Some internal re-factoring in preparation for the addition of Sieve
1834 extensions (by MH). In particular, the "personal" test is moved to a
1835 separate function, and given an option for scanning Cc: and Bcc: (which is
1836 not set for Exim filters).
1838 6. When Exim created an email address using the login of the caller as the
1839 local part (e.g. when creating a From: or Sender: header line), it was not
1840 quoting the local part when it contained special characters such as @.
1842 7. Installed new OpenBSD configuration files.
1844 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1845 try to make them clearer.
1847 9. Callout options, other than the timeout value, were being ignored when
1848 verifying sender addresses in header lines. For example, when using
1850 verify = header_sender/callout=no_cache
1852 the cache was (incorrectly) being used.
1854 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1855 only the Cygwin environment.
1857 11. When return_path_on_delivery was set as a log selector, if different remote
1858 addresses in the same message used different return paths and parallel
1859 remote delivery occurred, the wrong values would sometimes be logged.
1860 (Whenever a remote delivery process finished, the return path value from
1861 the most recently started remote delivery process was logged.)
1863 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1864 lines when AUTH and/or TLS are in use. This is the "received protocol"
1865 field. Exim used to use "asmtp" for authenticated SMTP, without any
1866 indication (in the protocol name) for TLS use. Now it follows the RFC and
1867 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1868 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1869 names appear in log lines as well as in Received: header lines.
1871 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1872 extensions, and comparison tests, and to fix some bugs.
1874 14. Changes to the "personal" filter test:
1876 (1) The test was buggy in that it was just doing the equivalent of
1877 "contains" tests on header lines. For example, if a user's address was
1878 anne@some.where, the "personal" test would incorrectly be true for
1880 To: susanne@some.where
1882 This test is now done by extracting each address from the header in turn,
1883 and checking the entire address. Other tests that are part of "personal"
1884 are now done using regular expressions (for example, to check local parts
1885 of addresses in From: header lines).
1887 (2) The list of non-personal local parts in From: addresses has been
1888 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1889 taken from the Sieve specification recommendations.
1891 (3) If the message contains any header line starting with "List-" it is
1892 treated as non-personal.
1894 (4) The test for "circular" in the Subject: header line has been removed
1895 because it now seems ill-conceived.
1897 15. Minor typos in src/EDITME comments corrected.
1899 16. Installed latest exipick from John Jetmore.
1901 17. If headers_add on a router specified a text string that was too long for
1902 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1903 of string_sprintf() is now avoided.
1905 18. $message_body_size was not set (it was always zero) when running the DATA
1906 ACL and the local_scan() function.
1908 19. For the "mail" command in an Exim filter, no default was being set for
1909 the once_repeat time, causing a random time value to be used if "once" was
1910 specified. (If the value happened to be <= 0, no repeat happened.) The
1911 default is now 0s, meaning "never repeat". The "vacation" command was OK
1912 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1913 (I found it when inspecting the code).
1915 20. There is now an overall timeout for performing a callout verification. It
1916 defaults to 4 times the callout timeout, which applies to individual SMTP
1917 commands during the callout. The overall timeout applies when there is more
1918 than one host that can be tried. The timeout is checked before trying the
1919 next host. This prevents very long delays if there are a large number of
1920 hosts and all are timing out (e.g. when the network connections are timing
1921 out). The value of the overall timeout can be changed by specifying an
1922 additional sub-option for "callout", called "maxwait". For example:
1924 verify = sender/callout=5s,maxwait=20s
1926 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1927 to the end before writing, but this should make it even safer).
1929 22. Exim was forgetting that it had advertised PIPELINING for the second and
1930 subsequent messages on an SMTP connection. It was also not resetting its
1931 memory on STARTTLS and an internal HELO.
1933 23. When Exim logs an SMTP synchronization error within a session, it now
1934 records whether PIPELINING has been advertised or not.
1936 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1937 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1938 rather than long int.
1940 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1942 26. Added the never_mail option to autoreply.
1948 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1949 crash if the getsockname() call failed; this can happen if a connection is
1950 closed very soon after it is established. The problem was simply in the
1951 order in which certain operations were done, causing Exim to try to write
1952 to the SMTP stream before it had set up the file descriptor. The bug has
1953 been fixed by making things happen in the correct order.
1959 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1960 before the connection was closed, thus losing the rejection response.
1962 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1963 some early Solaris releases, but causes trouble in current releases where
1964 socklen_t is defined.
1966 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1969 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1972 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1973 that releasing the top of it at the end releases what was used for sub-
1974 expansions (unless the block got too big). However, discard this block if
1975 the first thing is a variable or header, so that we can use its block when
1976 it is dynamic (useful for very large $message_headers, for example).
1978 6. Lookups now cache *every* query, not just the most recent. A new, separate
1979 store pool is used for this. It can be recovered when all lookup caches are
1980 flushed. Lookups now release memory at the end of their result strings.
1981 This has involved some general refactoring of the lookup sources.
1983 7. Some code has been added to the store_xxx() functions to reduce the amount
1984 of flapping under certain conditions.
1986 8. log_incoming_interface used to affect only the <= reception log lines. Now
1987 it causes the local interface and port to be added to several more SMTP log
1988 lines, for example "SMTP connection from", and rejection lines.
1990 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1992 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1994 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1995 could be overwritten at the end of the current message (or the start of a
1996 new message if it was set in a HELO ACL). The value is now preserved for
1997 the duration of the SMTP connection.
1999 12. If a transport had a headers_rewrite setting, and a matching header line
2000 contained an unqualified address, that address was qualified, even if it
2001 did not match any rewriting rules. The underlying bug was that the values
2002 of the flags that permit the existence of unqualified sender and recipient
2003 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2004 non-local messages, and by -bnq for local messages) were not being
2005 preserved with the message after it was received.
2007 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2008 "next input=" as part of the text comprising the host identity instead of
2009 the correct text. The code was using the same buffer for two different
2010 strings. However, depending on which order the printing function evaluated
2011 its arguments, the bug did not always show up. Under Linux, for example, my
2012 test suite worked just fine.
2014 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2015 doesn't work with some older versions of Perl. It has been changed to "my",
2016 which in any case is probably the better facility to use.
2018 15. A really picky compiler found some instances of statements for creating
2019 error messages that either had too many or two few arguments for the format
2022 16. The size of the buffer for calls to the DNS resolver has been increased
2023 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2024 for addresses that have a lot of PTR records. This alleviates a problem; it
2025 does not fully solve it.
2027 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2028 buffer now truncates the list and logs the incident, which is the same
2029 action as happens when Exim is looking up a host name and its aliases.
2030 Previously in this situation something unpredictable would happen;
2031 sometimes it was "internal error: store_reset failed".
2033 18. If a server dropped the connection unexpectedly when an Exim client was
2034 using GnuTLS and trying to read a response, the client delivery process
2035 crashed while trying to generate an error log message.
2037 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2038 string, for example:
2040 warn message = H1: something\nH2: something
2042 the text was added as a single header line from Exim's point of view
2043 though it ended up OK in the delivered message. However, searching for the
2044 second and subsequent header lines using $h_h2: did not work. This has been
2045 fixed. Similarly, if a system filter added multiple headers in this way,
2046 the routers could not see them.
2048 20. Expanded the error message when iplsearch is called with an invalid key to
2049 suggest using net-iplsearch in a host list.
2051 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2052 ACLs are no longer actually imposed (and a message to that effect is
2055 22. If a "gecos" field in a passwd entry contained escaped characters, in
2056 particular, if it contained a \" sequence, Exim got it wrong when building
2057 a From: or a Sender: header from that name. A second bug also caused
2058 incorrect handling when an unquoted " was present following a character
2059 that needed quoting.
2061 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2062 was not being matched caselessly.
2064 24. Arranged for all hyphens in the exim.8 source to be escaped with
2067 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2068 properly. Recipient callout cache records were still being keyed to include
2069 the sender, even when use_sender was set false. This led to far more
2070 callouts that were necessary. The sender is no longer included in the key
2071 when use_sender is false.
2073 26. Added "control = submission" modifier to ACLs.
2075 27. Added the ${base62d: operator to decode base 62 numbers.
2077 28. dnsdb lookups can now access SRV records.
2079 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2080 the configuration file.
2082 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2083 (-v) mode. This makes the output for a verbose queue run more intelligible.
2085 31. Added a use_postmaster feature to recipient callouts.
2087 32. Added the $body_zerocount variable, containing the number of binary zero
2088 bytes in the message body.
2090 33. The time of last modification of the "new" subdirectory is now used as the
2091 "mailbox time last read" when there is a quota error for a maildir
2094 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2096 35. Added +ignore_unknown as a special item in host lists.
2098 36. Code for decoding IPv6 addresses in host lists is now included, even if
2099 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2100 address was recognized as an IP address, but was then not correctly decoded
2101 into binary, causing unexpected and incorrect effects when compared with
2108 1. Very minor rewording of debugging text in manualroute to say "list of
2109 hosts" instead of "hostlist".
2111 2. If verify=header_syntax was set, and a header line with an unqualified
2112 address (no domain) and a large number of spaces between the end of the
2113 name and the colon was received, the reception process suffered a buffer
2114 overflow, and (when I tested it) crashed. This was caused by some obsolete
2115 code that should have been removed. The fix is to remove it!
2117 3. When running in the test harness, delay a bit after writing a bounce
2118 message to get a bit more predictability in the log output.
2120 4. Added a call to search_tidyup() just before forking a reception process. In
2121 theory, someone could use a lookup in the expansion of smtp_accept_max_
2122 per_host which, without the tidyup, could leave open a database connection.
2124 5. Added the variables $recipient_data and $sender_data which get set from a
2125 lookup success in an ACL "recipients" or "senders" condition, or a router
2126 "senders" option, similar to $domain_data and $local_part_data.
2128 6. Moved the writing of debug_print from before to after the "senders" test
2131 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2132 problems for message scanning, either using a data ACL, or using
2133 local_scan() because the Received: header was not generated till after they
2134 were called (in order to set the time as the time of reception completion).
2135 I have revised the way this works. The header is now generated after the
2136 body is received, but before the ACL or local_scan() are called. After they
2137 are run, the timestamp in the header is updated.
2143 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2144 before starting a queue runner without re-exec. This happened only when
2145 deliver_drop_privilege was set or when the Exim user was set to root. The
2146 effect of the bug was that timeouts during subsequent deliveries caused
2147 crashes instead of being properly handled. The handler is now left at its
2148 default (and expected) setting.
2150 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2151 message, again when deliver_drop_privilege is set or Exim is run as root.
2152 The bug described in (1) was not present in this case, but the tidying up
2153 of the other signals was missing. I have made the two cases consistent.
2155 3. The ignore_target_hosts setting on a manualroute router was being ignored
2156 for hosts that were looked up using the /MX notation.
2158 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2161 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2162 operated on the sender address. After changing the $sender_address to <>
2163 for the sender address verify, Exim was re-instated it as the original
2164 (before rewriting) address, but remembering that it had rewritten it, so it
2165 wasn't rewriting it again. This bug also had the effect of breaking the
2166 sender address verification caching when the sender address was rewritten.
2168 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2169 This has been changed so that if the ip literal address matches
2170 ignore_target_hosts, the router declines.
2172 7. Added expansion conditions match_domain, match_address, and match_local_
2173 part (NOT match_host).
2175 8. The placeholder for the Received: header didn't have a length field set.
2177 9. Added code to Exim itself and to exim_lock to test for a specific race
2178 condition that could lead to file corruption when using MBX delivery. The
2179 issue is with the lockfile that is created in /tmp. If this file is removed
2180 after a process has opened it but before that process has acquired a lock,
2181 there is the potential for a second process to recreate the file and also
2182 acquire a lock. This could lead to two Exim processes writing to the file
2183 at the same time. The added code performs the same test as UW imapd; it
2184 checks after acquiring the lock that its file descriptor still refers to
2185 the same named file.
2187 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2188 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2191 11. Added the smtp_active_hostname option. If used, this will typically be made
2192 to depend on the incoming interface address. Because $interface_address is
2193 not set up until the daemon has forked a reception process, error responses
2194 that can happen earlier (such as "too many connections") no longer contain
2197 12. If an expansion in a condition on a "warn" statement fails because a lookup
2198 defers, the "warn" statement is abandoned, and the next ACL statement is
2199 processed. Previously this caused the whole ACL to be aborted.
2201 13. Added the iplsearch lookup type.
2203 14. Added ident_timeout as a log selector.
2205 15. Added tls_certificate_verified as a log selector.
2207 16. Added a global option tls_require_ciphers (compare the smtp transport
2208 option of the same name). This controls incoming TLS connections.
2210 17. I finally figured out how to make tls_require_ciphers do a similar thing
2211 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2212 before starting the TLS session.
2214 18. Tabs are now shown as \t in -bP output.
2216 19. If the log selector return_path_on_delivery was set, Exim crashed when
2217 bouncing a message because it had too many Received: header lines.
2219 20. If two routers both had headers_remove settings, and the first one included
2220 a superfluous trailing colon, the final name in the first list and the
2221 first name in the second list were incorrectly joined into one item (with a
2222 colon in the middle).
2228 1. Added -C and -D options to the exinext utility, mainly to make it easier
2229 to include in the automated testing, but these could be helpful when
2230 multiple configurations are in use.
2232 2. The exinext utility was not formatting the output nicely when there was
2233 an alternate port involved in the retry record key, nor when there was a
2234 message id as well (for retries that were specific to a specific message
2235 and a specific host). It was also confused by IPv6 addresses, because of
2236 the additional colons they contain. I have fixed the IPv4 problem, and
2237 patched it up to do a reasonable job for IPv6.
2239 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2240 delivery, the log line now contains "pipelined" if PIPELINING was used.
2242 4. An SMTP transport process used to panic and die if the bind() call to set
2243 an explicit outgoing interface failed. This has been changed; it is now
2244 treated in the same way as a connect() failure.
2246 5. A reference to $sender_host_name in the part of a conditional expansion
2247 that was being skipped was still causing a DNS lookup. This no longer
2250 6. The def: expansion condition was not recognizing references to header lines
2251 that used bh_ and bheader_.
2253 7. Added the _cache feature to named lists.
2255 8. The code for checking quota_filecount in the appendfile transport was
2256 allowing one more file than it should have been.
2258 9. For compatibility with Sendmail, the command line option
2266 and sets the incoming protocol and host name (for trusted callers). The
2267 host name and its colon can be omitted when only the protocol is to be set.
2268 Note the Exim already has two private options, -pd and -ps, that refer to
2269 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2270 "s", but I don't think that's a major issue.
2272 10. A number of refactoring changes to the code, none of which should affect
2275 (a) The number of logging options was getting close to filling up the
2276 32-bit word that was used as a bit map. I have split them into two classes:
2277 those that are passed in the argument to log_write(), and those that are
2278 only ever tested independently outside of that function. These are now in
2279 separate 32-bit words, so there is plenty of room for expansion again.
2280 There is no change in the user interface or the logging behaviour.
2282 (b) When building, for example, log lines, the code previously used a
2283 macro that called string_cat() twice, in order to add two strings. This is
2284 not really sufficiently general. Furthermore, there was one instance where
2285 it was actually wrong because one of the argument was used twice, and in
2286 one call a function was used. (As it happened, calling the function twice
2287 did not affect the overall behaviour.) The macro has been replaced by a
2288 function that can join an arbitrary number of extra strings onto a growing
2291 (c) The code for expansion conditions now uses a table and a binary chop
2292 instead of a serial search (which was left over from when there were very
2293 few conditions). Also, it now recognizes conditions like "pam" even when
2294 the relevant support is not compiled in: a suitably worded error message is
2295 given if an attempt is made to use such a condition.
2297 11. Added ${time_interval:xxxxx}.
2299 12. A bug was causing one of the ddress fields not to be passed back correctly
2300 from remote delivery subprocesses. The field in question was not being
2301 subsequently used, so this caused to problems in practice.
2303 13. Added new log selectors queue_time and deliver_time.
2305 14. Might have fixed a bug in maildirsizefile handling that threw up
2306 "unexpected character" debug warnings, and recalculated the data
2307 unnecessarily. In any case, I expanded the warning message to give more
2310 15. Added the message "Restricted characters in address" to the statements in
2311 the default ACL that block characters like @ and % in local parts.
2313 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2314 Three changes have been made:
2316 (a) There was a serious bug; a negative response to MAIL caused the whole
2317 recipient domain to be cached as invalid, thereby blocking all messages
2318 to all local parts at the same domain, from all senders. This bug has
2319 been fixed. The domain is no longer cached after a negative response to
2320 MAIL if the sender used is not empty.
2322 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2325 (c) A new callout option, "use_sender" has been added for people who want
2326 the modified behaviour.
2332 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2335 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2336 indeed breaks things for older releases.
2338 3. Added additional logging to the case where there is a problem reading data
2339 from a filter that is running in a subprocess using a pipe, in order to
2340 try to track down a specific problem.
2342 4. Testing facility fudge: when running in the test harness and attempting
2343 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2344 getting "No route to host". Convert this to a timeout.
2346 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2349 6. Some OS don't have socklen_t but use size_t instead. This affects the
2350 fifth argument of getsockopt() amongst other things. This is now
2351 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2352 can be set for individual OS. I have set it for SunOS5, OSF1, and
2353 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2354 some earlier ones do not.
2356 7. Change 4.30/15 was not doing the test caselessly.
2358 8. The standard form for an IPv6 address literal was being rejected by address
2359 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2360 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2361 this, as well as the form without the "IPv6" on the front (but only when
2362 address literals are enabled, of course).
2364 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2366 10. Exim crashed if a message with an empty sender address specified by -f
2367 encountered a router with an errors_to setting. This could be provoked only
2368 by a command such as
2372 where an empty string was supplied; "<>" did not hit this bug.
2374 11. Installed PCRE release 4.5.
2376 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2377 remained set. It is now erased.
2379 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2380 times from message ids (which are base 36 rather than the normal 62).
2382 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2383 were being counted as actual protocol errors, and logged if the log
2384 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2385 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2386 and DATA following a set of rejected RCPTs do not count as protocol errors.
2387 In other words, Exim assumes they were pipelined, though this may not
2388 actually be the case. Of course, in all cases the client gets an
2389 appropriate error code.
2391 15. If a lookup fails in an ACL condition, a message about the failure may
2392 be available; it is used if testing the ACL cannot continue, because most
2393 such messages specify what the cause of the deferral is. However, some
2394 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2395 that caused an old message to be retained and used if a later statement
2396 caused a defer, replacing the real cause of the deferral.
2398 16. If an IP address had so many PTR records that the DNS lookup buffer
2399 was not large enough to hold them, Exim could crash while trying to process
2400 the truncated data. It now detects and logs this case.
2402 17. Further to 4.21/58, another change has been made: if (and only if) the
2403 first line of a message (the first header line) ends with CRLF, a bare LF
2404 in a subsequent header line has a space inserted after it, so as not to
2405 terminate the header.
2407 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2408 unnecessarily, used atoi() instead of strtol(), and didn't check the
2409 termination when getting file sizes from file names by regex.
2411 19. Completely re-implemented the support for maildirsize files, in the light
2412 of a number of problems with the previous contributed implementation
2413 (4.30/29). In particular:
2415 . If the quota is zero, the maildirsize file is maintained, but no quota is
2418 . If the maildir directory does not exist, it is created before any attempt
2419 to write a maildirsize file.
2421 . The quota value in the file is just a cache; if the quota is changed in
2422 the transport, the new value overrides.
2424 . A regular expression is available for excluding directories from the
2427 20. The autoreply transport checks the characters in options that define the
2428 message's headers; it allows continued headers, but it was checking with
2429 isspace() after an embedded newline instead of explicitly looking for a
2432 21. If all the "regular" hosts to which an address was routed had passed their
2433 expiry times, and had not reached their retry times, the address was
2434 bounced, even if fallback hosts were defined. Now Exim should go on to try
2437 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2438 equivalent code in the SMTP transport. Some hosts send humungous responses
2439 to HELO/EHLO, more than 1024 it seems.
2441 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2442 gives compiler warnings in some environments. I've now done it "properly",
2445 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2446 (because of problems with the built-in one) was declared to return uschar *
2447 instead of char *, causing compiler failure.
2449 25. Fixed a file descriptor leak when processing alias/forward files.
2451 26. Fixed a minor format string issue in dbfn.c.
2453 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2455 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2456 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2459 29. When a sender address is verified, it is cached, to save repeating the test
2460 when there is more than one recipient in a message. However, when the
2461 verification involves a callout, it is possible for different callout
2462 options to be set for different recipients. It is too complicated to keep
2463 track of this in the cache, so now Exim always runs a verification when a
2464 callout is required, relying on the callout cache for the optimization.
2465 The overhead is duplication of the address routing, but this should not be
2468 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2469 to be verified with callout=postmaster, and the main callout worked but the
2470 postmaster check failed, the verification correctly failed. However, if a
2471 subsequent RCPT command asked for sender verification *without* the
2472 postmaster check, incorrect caching caused this verification also to fail,
2475 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2476 it was not caching the DNS options (qualify_single, search_parents) that
2477 were used when the lookup failed. A subsequent lookup with different
2478 options therefore always gave the same answer, though there were cases
2479 where it should not have. (Example: a "domains = !$mx_any" option on a
2480 dnslookup router: the "domains" option is always processed without any
2481 widening, but the router might have qualify_single set.) Now Exim uses the
2482 cached value only when the same options are set.
2484 32. Added John Jetmore's "exipick" utility to the distribution.
2486 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2487 than a timeout (e.g. a certificate is required, and is not provided), an
2488 Exim server now closes the connection immediately. Previously it waited for
2489 the client to close - but if the client is SSL, it seems that they each
2490 wait for each other, leading to a delay before one of them times out.
2492 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2493 maintained 0.8.x compatibility because I don't think many are using it, and
2494 it is clearly obsolete.
2496 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2499 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2500 client certificate was expired. A simple patch fixes this, though I don't
2501 understand the full logic of why the verify callback is called multiple
2504 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2505 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2506 which causes problems with some clients (such as the Certicom SSL Plus
2507 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2508 disables the coutermeasure allowing Eudora to connect."
2510 38. Exim was not checking that a write() to a log file succeeded. This could
2511 lead to Bad Things if a log got too big, in particular if it hit a file
2512 size limit. Exim now panics and dies if it cannot write to a log file, just
2513 as it does if it cannot open a log file.
2515 39. Modified OS/Makefile-Linux so that it now contains
2517 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2519 The two -D definitions ensure that Exim is compiled with large file
2520 support, which makes it possible to handle log files that are bigger than
2523 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2524 instance) a domain was checked against a named list that involved a lookup,
2525 causing $domain_data to be set, then another domain was checked against the
2526 same list, then the first domain was re-checked, the value of $domain_data
2527 after the final check could be wrong. In particular, if the second check
2528 failed, it could be set empty. This bug probably also applied to
2531 41. The strip_trailing_dot option was not being applied to the address given
2532 with the -f command-line option.
2534 42. The code for reading a message's header from the spool was incrementing
2535 $received_count, but never initializing it. This meant that the value was
2536 incorrect (doubled) while delivering a message in the same process in which
2537 it was received. In the most common configuration of Exim, this never
2538 happens - a fresh exec is done - but it can happen when
2539 deliver_drop_privilege is set.
2541 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2542 it now includes up to 150 characters of the unexpected data in the log
2545 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2546 and building data strings. The size of both of these buffers was 10 000
2547 bytes - far larger than anybody would *ever* want, thought I. Needless to
2548 say, somebody hit the limit. I have increased the maximum line length to
2549 20 000 and the maximum data length of concatenated lines to 100 000. I have
2550 also fixed two bugs, because there was no checking on these buffers. Tsk,
2551 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2552 buffer is too small.
2554 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2555 lsearch lookups. Now it does.
2557 46. When parsing a route_list item in a manualroute router, a fixed-length
2558 buffer was used for the list of hosts. I made this 1024 bytes long,
2559 thinking that nobody would ever have a list of hosts that long. Wrong.
2560 Somebody had a whole pile of complicated expansion conditions, and the
2561 string was silently truncated, leading to an expansion error. It turns out
2562 that it is easier to change to an unlimited length (owing to other changes
2563 that have happened since this code was originally written) than to build
2564 structure for giving a limitation error. The length of the item that
2565 expands into the list of hosts is now unlimited.
2567 47. The lsearch lookup could not handle data where the length of text line was
2568 more than 4095 characters. Such lines were truncated, leading to shortened
2569 data being returned. It should now handle lines of any length.
2571 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2572 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2575 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2576 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2577 no longer happens in comment lines. A list of replacements is now placed
2578 at the head of all of the source files, except those whose only change is
2579 to replace PERL_COMMAND in the very first #! line.
2581 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2582 messages on the queue, with a bottom-up merge sort, using code contributed
2583 by Michael Haardt. This should make operations like -bp somewhat faster on
2584 large queues. It won't affect queue runners, except when queue_run_in_order
2587 51. Installed eximstats 1.31 in the distribution.
2589 52. Added support for SRV lookups to the dnslookup router.
2591 53. If an ACL referred to $message_body or $message_body_end, the value was not
2592 reset for any messages that followed in the same SMTP session.
2594 54. The store-handling optimization for building very long strings was not
2595 differentiating between the different store pools. I don't think this
2596 actually made any difference in practice, but I've tidied it.
2598 55. While running the routers to verify a sender address, $sender_address
2599 was still set to the sender address. This is wrong, because when routing to
2600 send a bounce to the sender, it would be empty. Therefore, I have changed
2601 it so that, while verifying a sender address, $sender_address is set to <>.
2602 (There is no change to what happens when verifying a recipient address.)
2604 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2605 target A or AAAA records (if not already returned) without resetting the
2606 qualify_single or search_parents options of the DNS resolver. These are
2607 inappropriate in this case because the targets of MX and SRV records must
2608 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2609 target that, when qualified, matched something in the local domain. These
2610 two options are now turned off when doing these lookups.
2612 57. It seems that at least some releases of Reiserfs (which does not have the
2613 concept of a fixed number of inodes) returns zero and not -1 for the
2614 number of available inodes. This interacted badly with check_spool_inodes,
2615 which assumed that -1 was the "no such thing" setting. What I have done is
2616 to check that the total number of inodes is greater than zero before doing
2617 the test of how many are available.
2619 58. When a "warn" ACL statement has a log_message modifier, the message is
2620 remembered, and not repeated. This is to avoid a lot of repetition when a
2621 message has many recipients that cause the same warning to be written.
2622 Howewer, Exim was preserving the list of already written lines for an
2623 entire SMTP session, which doesn't seem right. The memory is now reset if a
2624 new message is started.
2626 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2627 debugging output unless log_rewrite was also set.
2629 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2630 of (int)(handle) when we know that handle contains (void *)(-1).
2632 61. The Exim daemon panic-logs an error return when it closes the incoming
2633 connection. However "connection reset by peer" seems to be common, and
2634 isn't really an error worthy of noting specially, so that particular error
2637 62. When Exim is trying to find all the local interfaces, it used to panic and
2638 die if the ioctl to get the interface flags failed. However, it seems that
2639 on at least one OS (Solaris 9) it is possible to have an interface that is
2640 included in the list of interfaces, but for which you get a failure error
2641 for this call. This happens when the interface is not "plumbed" into a
2642 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2643 failure of the "get flags" call assumes that the interface is down.
2645 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2646 makes life easier for people who are doing arithmetic on fields extracted
2647 from dates, where you often get leading zeros that should not be
2648 interpreted as octal.
2650 64. Added qualify_domain to the redirect router, to override the global
2653 65. If a pathologically long header line contained very many addresses (the
2654 report of this problem mentioned 10 000) and each of them was rewritten,
2655 Exim could use up a very large amount of memory. (It kept on making new
2656 copies of the header line as it rewrote, and never released the old ones.)
2657 At the expense of a bit more processing, the header rewriting function has
2658 been changed so that it no longer eats memory in this way.
2660 66. The generation of the Received: header has been moved from the time that a
2661 message starts to be received, to the time that it finishes. The timestamp
2662 in the Received: header should now be very close to that of the <= log
2663 line. There are two side-effects of this change:
2665 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2666 logged header lines no longer include the local Received: line, because
2667 it has not yet been created. The same applies to a copy of the message
2668 that is returned to a non-SMTP sender when a message is rejected.
2670 (b) When a filter file is tested using -bf, no additional Received: header
2671 is added to the test message. After some thought, I decided that this
2674 This change does not affect the value of $received_for. It is still set
2675 after address rewriting, but before local_scan() is called.
2677 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2679 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2680 gave an unhelpful panic error message, and a defer error. I have managed to
2681 change this behaviour so that it now rejects any supplied certificate,
2682 which seems right, as the list of acceptable certificates is empty.
2684 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2685 gave an unhelpful defer error. I have not managed to make this reject any
2686 supplied certificates, but the error message it gives is "no certificate
2687 supplied", which is not helpful.
2689 70. exigrep's output now also includes lines that are not associated with any
2690 message, but which match the given pattern. Implemented by a patch from
2691 Martin Sluka, which also tidied up the Perl a bit.
2693 71. Recipient callout verification, like sender verification, was using <> in
2694 the MAIL FROM command. This isn't really the right thing, since the actual
2695 sender may affect whether the remote host accepts the recipient or not. I
2696 have changed it to use the actual sender in the callout; this means that
2697 the cache record is now keyed on a recipient/sender pair, not just the
2698 recipient address. There doesn't seem to be a real danger of callout loops,
2699 since a callout by the remote host to check the sender would use <>.
2700 [SEE ABOVE: changed after hitting problems.]
2702 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2703 temporary errors. However, in the case of such a code being given after
2704 the end of a data transmission (i.e. after ".") Exim was failing to write
2705 a retry record for the message. (Yes, there was some broken host that was
2706 actually sending 8xx at this point.)
2708 73. An unknown lookup type in a host list could cause Exim to panic-die when
2709 the list was checked. (An example that provoked this was putting <; in the
2710 middle of a list instead of at the start.) If this happened during a DATA
2711 ACL check, a -D file could be left lying around. This kind of configuration
2712 error no longer causes Exim to die; instead it causes a defer errror. The
2713 incident is still logged to the main and panic logs.
2715 74. Buglet left over from Exim 3 conversion. The message "too many messages
2716 in one connection" was written to the rejectlog but not the mainlog, except
2717 when address rewriting (yes!) was being logged.
2719 75. Added write_rejectlog option.
2721 76. When a system filter was run not as root (that is, when system_filter_user
2722 was set), the values of the $n variables were not being returned to the
2723 main process; thus, they were not subsequently available in the $sn
2726 77. Added +return_path_on_delivery log selector.
2728 78. A connection timeout was being treated differently from recipients deferred
2729 when testing hosts_max_try with a message that was older than the host's
2730 retry timeout. (The host should not be counted, thus allowing all hosts to
2731 be tried at least once before bouncing.) This may have been the cause of an
2732 occasionally reported bug whereby a message would remain on the queue
2733 longer than the retry timeout, but would be bounced if a delivery was
2734 forced. I say "may" because I never totally pinned down the problem;
2735 setting up timeout/retry tests is difficult. See also the next item.
2737 79. The ultimate address timeout was not being applied to errors that involved
2738 a combination of host plus message (for example, a timeout on a MAIL
2739 command). When an address resolved to a number of possible hosts, and they
2740 were not all tried for each delivery (e.g. because of hosts_max_try), a
2741 message could remain on the queue longer than the retry timeout.
2743 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2746 81. Fixed an obscure SMTP outgoing bug which required at least the following
2747 conditions: (a) there was another message waiting for the same server;
2748 (b) the server returned 5xx to all RCPT commands in the first message so
2749 that the message was not completed; (c) the server dropped the connection
2750 or gave a negative response to the RSET that Exim sends to abort the
2751 transaction. The observed case was a dropped connection after DATA that had
2752 been sent in pipelining mode. That is, the server had advertised PIPELINING
2753 but was not implementing it correctly. The effect of the bug was incorrect
2754 behaviour, such as trying another host, and this could lead to a crash.
2760 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2761 and daemon.c were passed as pointers to ints; they should have been
2762 pointers to socklen_t variables (which are typically unsigned ints).
2764 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2767 3. Fixed a really odd bug that affected only the testing scheme; patching a
2768 certain fixed string in the binary changed the value of another string that
2769 happened to be identical to the end of the original first string.
2771 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2772 name", it returns that address as the IP address. On some operating
2773 systems (e.g. Solaris), it also passes back the IP address string as the
2774 "host name". However, on others (e.g. Linux), it passes back an empty
2775 string. Exim wasn't checking for this, and was changing the host name to an
2776 empty string, assuming it had been canonicized.
2778 5. Although rare, it is permitted to have more than one PTR record for a given
2779 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2780 all the names associated with an address, because they do in Solaris.
2781 However, it seems that they do not in Linux for data that comes from the
2782 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2783 I found this out when I moved to a new Linux workstation and tried to run
2784 the Exim test suite.
2786 To get round this problem I have changed the code so that it now does its
2787 own call to the DNS to look up PTR records when searching for a host name.
2788 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2789 addresses that are only in /etc/hosts are still found.
2791 This behaviour is, however, controlled by an option called host_lookup_
2792 order, which defaults to "bydns:byaddr". If people want to use the other
2793 order, or indeed, just use one or the other means of lookup, they can
2794 specify it in this variable.
2796 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2797 some operating systems, this comes back from gethostbyaddr() as an empty
2798 string, and this is what Exim used to test for. However, it seems that in
2799 other systems, "." is yielded. Exim now tests for this case too.
2801 7. The values of check_spool_space and check_log_space are now held internally
2802 as a number of kilobytes instead of an absolute number of bytes. If a
2803 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2804 kilobyte. This means that much larger values can be stored.
2806 8. Exim monitor: an attempt to get the action menu when not actually pointing
2807 at a message produces an empty menu entitled "No message selected". This
2808 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2809 no entries in it ("Shell widget menu has zero width and/or height"). So I
2810 have added a single, blank menu entry in this case.
2812 9. Added ${quote_local_part.
2814 10. MIME decoding is now applied to the contents of Subject: header lines when
2817 11. Now that a reference to $sender_host_address automatically causes a reverse
2818 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2819 host lookup before query-style lookups in lists that might use this
2820 variable. This has therefore been abolished, and the "net-" prefix is no
2821 longer necessary for query-style lookups.
2823 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2824 have been a typo for LFLAGS, so it has been changed.
2826 13. The install script calls Exim with "-C /dev/null" in order to find the
2827 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2828 to be output. Howeve, since Exim outputs its version number before the
2829 error, it didn't break the script. It just looked ugly. I fixed this by
2830 always allowing "-C /dev/null" if the caller is root.
2832 14. Ignore overlarge ACL variable number when reading spool file - insurance
2833 against a later release with more variables having written the file.
2835 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2836 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2837 this, as well as the form without the "IPv6" on the front.
2839 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2840 OS/Makefile-Darwin file.
2842 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2843 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2845 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2846 variable. However, some OS use ints for inodes. I've added cast to long int
2847 to get rid of the compiler warning.
2849 19. I had forgotten to lock out "/../" in configuration file names when
2850 ALT_CONFIG_PREFIX was set.
2852 20. Routers used for verification do not need to specify transports. However,
2853 if such a router generated a host list, and callout was configured, Exim
2854 crashed, because it could not find a port number from the (non-existent)
2855 transport. It now assumes port 25 in this circumstance.
2857 21. Added the -t option to exigrep.
2859 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2860 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2861 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2862 them is set, LOOKUP_LSEARCH is forced.
2864 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2866 24. Added sender and host information to the "rejected by local_scan()" log
2867 line; previously there was no indication of these.
2869 25. Added .include_if_exists.
2871 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2872 Linux. It turns out that not all file systems support this. Apparently some
2873 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2874 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2875 not supported on the file descriptor, is now ignored when Exim is trying to
2876 sync a directory. This applies only to Linux.
2878 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2880 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2881 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2884 29. Added support for maildirsize files from supplied patch (modified a bit).
2886 30. The use of :fail: followed by an empty string could lead Exim to respond to
2887 sender verification failures with (e.g.):
2889 550 Verification failed for <xxx>
2890 550 Sender verify failed
2892 where the first response line was missing the '-' that indicates it is not
2893 the final line of the response.
2895 31. The loop for finding the name of the user that called Exim had a hardwired
2896 limit of 10; it now uses the value of finduser_retries, which is used for
2897 all other user lookups.
2899 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2902 33. Exim was neglecting to zero errno before one call of strtol() when
2903 expanding a string and expecting an integer value. On some systems this
2904 resulted in spurious "integer overflow" errors. Also, it was casting the
2905 result into an int without checking.
2907 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2908 did not work. The code looks as if it has *never* worked, though it appears
2909 to have been documented since at least releast 1.62. I have made it work.
2911 35. The "timeout_DNS" error in retry rules, also documented since at least
2912 1.62, also never worked. As it isn't clear exactly what this means, and
2913 clearly it isn't a major issue, I have abolished the feature by treating it
2914 as "timeout", and writing a warning to the main and panic logs.
2916 36. The display of retry rules for -brt wasn't always showing the error code
2919 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2920 timeout_connect_A, timeout_connect_MX.
2922 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2923 to the empty sender.
2925 39. The daemon was not analysing the content of -oX till after it had closed
2926 stderr and disconnected from the controlling terminal. This meant that any
2927 syntax errors were only noted on the panic log, and the return code from
2928 the command was 0. By re-arranging the code a little, I've made the
2929 decoding happen first, so such errors now appear on stderr, and the return
2930 code is 1. However, the actual setting up of the sockets still happens in
2931 the disconnected process, so errors there are still only recorded on the
2934 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2935 connections (as happens on some IP stacks) was logged at start up time as
2936 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2937 it from "IPv6 and IPv4", which means that two separate sockets are being
2940 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2941 says whether AF_INET or AF_INET6 was passed as an argument.
2943 42. Exiwhat output was messed up when time zones were included in log
2946 43. Exiwhat now gives more information about the daemon's listening ports,
2947 and whether -tls-on-connect was used.
2949 44. The "port" option of the smtp transport is now expanded.
2951 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2952 silently ignored. Now an error message is written to the main and panic
2955 46. There's a new ACL modifier called "logwrite" which writes to a log file
2956 as soon as it is encountered.
2958 47. Added $local_user_uid and $local_user_gid at routing time.
2960 48. Exim crashed when trying to verify a sender address that was being
2963 49. Exim was recognizing only a space character after ".include". It now also
2964 recognizes a tab character.
2966 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2967 extracting the relevant information from the specification. The man page no
2968 longer contains scrambled data for the -d option, and I've added a section
2969 at the front about calling Exim under different names.
2971 51. Added "extra_headers" argument to the "mail" command in filter files.
2973 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2976 53. Installed eximstats 1.29.
2978 54. Added transport_filter_timeout as a generic transport option.
2980 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2981 Cc: header lines. This was required by RFC 822, but it not required by RFC
2984 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2985 incoming messages that did not have them. Now it does so only if the
2986 message originates locally, that is, if there is no associated remote host
2987 address. When Resent- header lines are present, this applies to the Resent-
2988 lines rather than the non-Resent- lines.
2990 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2991 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2993 58. Messages for configuration errors now include the name of the main
2994 configuration file - useful now that there may be more than one file in a
2995 list (.included file names were always shown).
2997 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2998 for those rare installations that do not start the daemon as root or run it
2999 setuid root. I've cut out the call to initgroups() if the daemon is not
3002 60. The Exim user and group can now be bound into the binary as text strings
3003 that are looked up at the start of Exim's processing.
3005 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3007 62. Added $mailstore_basename variable.
3009 63. Installed patch to sieve.c from Michael Haardt.
3011 64. When Exim failed to open the panic log after failing to open the main log,
3012 the original message it was trying to log was written to stderr and debug
3013 output, but if they were not available (the usual case in production), it
3014 was lost. Now it is written to syslog before the two lines that record the
3015 failures to open the logs.
3017 65. Users' Exim filters run in subprocesses under the user's uid. It is
3018 possible for a "deliver" command or an alias in a "personal" command to
3019 provoke an address rewrite. If logging of address rewriting is configured,
3020 this fails because the process is not running as root or exim. There may be
3021 a better way of dealing with this, but for the moment (because 4.30 needs
3022 to be released), I have disabled address rewrite logging when running a
3023 filter in a non-root, non-exim process.
3029 1. The buildconfig auxiliary program wasn't quoting the value set for
3030 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3031 not defined. This bug was masked in 4.22 by the effect that was fixed in
3034 2. Some messages that were rejected after a message id was allocated were
3035 shown as "incomplete" by exigrep. It no longer does this for messages that
3036 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3038 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3039 have allow_domain_literals set, the ID did not get logged in the <= line.
3040 Domain literals are now always recognized in Message-ID: header lines.
3042 4. The first argument for a ${extract expansion item is the key name or field
3043 number. Leading and trailing spaces in this item were not being ignored,
3044 causing some misleading effects.
3046 5. When deliver_drop_privilege was set, single queue runner processes started
3047 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3048 same command in the process it spins off) were not dropping privilege.
3050 6. When the daemon running as "exim" started a queue runner, it always
3051 re-executed Exim in the spun-off process. This is a waste of effort when
3052 deliver_drop_privilege is set. The new process now just calls the
3053 queue-runner function directly.
3059 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3062 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3063 ignored. Though the use of -oP was forcing the writing of a pid file, it
3064 was always written to the default place.
3066 3. If the message "no IP address found for host xxxx" is generated during
3067 incoming verification, it is now followed by identification of the incoming
3068 connection (so you can more easily find what provoked it).
3070 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3072 5. Added some features to "harden" Exim a bit more against certain attacks:
3074 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3075 be put in Local/Makefile. This is like the never_users runtime option,
3076 but it cannot be overridden. The default setting is "root".
3078 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3079 prefix string with which any file named in a -C command line option
3082 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3083 is retained for -C and -D only if the caller of Exim is root. Without
3084 it, the exim user may also use -C and -D and retain privilege.
3086 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3087 command line option is disabled.
3089 6. Macro names set by the -D option must start with an upper case letter, just
3090 like macro names defined in the configuration file.
3092 7. Added "dereference=" facility to LDAP.
3094 8. Two instances of the typo "uknown" in the source files are fixed.
3096 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3097 the Configure-Makefile script screwed up while processing it.
3099 10. Incorporated PCRE 4.4.
3101 11. The SMTP synchronization check was not operating right at the start of an
3102 SMTP session. For example, it could not catch a HELO sent before the client
3103 waited for the greeting. There is now a check for outstanding input at the
3104 point when the greeting is written. Because of the duplex, asynchronous
3105 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3106 way, but not yet received, when the check is performed.
3108 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3109 on TCP/IP sockets, because this apparently causes some broken clients to
3112 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3113 unchanged) from the Cygwin maintainer.
3115 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3116 was supported instead of testing for mbx. Effectively a typo.
3118 15. The spa authenticator server code was not checking that the input it
3119 received was valid base64.
3121 16. The debug output line for the "set" modifier in ACLs was not showing the
3122 name of the variable that was being set.
3124 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3126 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3127 lookup on its own. Something else was needed to trigger the lookup. For
3128 example, a match in host_lookup or the need for a host name in a host list.
3129 Now, if $sender_host_name is referenced and the host name has not yet been
3130 looked up, a lookup is performed. If the lookup fails, the variable remains
3131 empty, and $host_lookup_failed is set to "1".
3133 19. Added "eqi" as a case-independent comparison operator.
3135 20. The saslauthd authentication condition could segfault if neither service
3136 nor realm was specified.
3138 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3139 error message that was logged was misleading, and incoming SMTP
3140 connections were dropped. The message is now more accurate, and temporary
3141 errors are given to SMTP connections.
3143 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3144 (except RSET) with a 421 error, until QUIT is received. However, it was
3145 failing to send a response to QUIT.
3147 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3148 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3149 ACL is now run after the helo_try_verify_hosts code.
3151 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3152 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3153 case-independent, so other case variants are also recognized). Apparently
3154 some systems use these upper case variants.
3156 25. If more than two messages were waiting for the same host, and a transport
3157 filter was specified for the transport, Exim sent two messages over the
3158 same TCP/IP connection, and then failed with "socket operation on non-
3159 socket" when it tried to send the third.
3161 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3163 27. The extern definition of crypt16() in expand.c was not being excluded when
3164 the OS had its own crypt16() function.
3166 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3167 as a preferred synonym for return_size_limit, both as an option and as an
3170 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3172 30. Changed the default configuration ACL to relax the local part checking rule
3173 for addresses that are not in any local domains. For these addresses,
3174 slashes and pipe symbols are allowed within local parts, but the sequence
3175 /../ is explicitly forbidden.
3177 31. SPA server authentication was not clearing the challenge buffer before
3180 32. log_message in a "warn" ACL statement was writing to the reject log as
3181 well as to the main log, which contradicts the documentation and doesn't
3182 seem right (because no rejection is happening). So I have stopped it.
3184 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3185 However, I am unable to do any testing of this.
3187 34. Fixed an infelicity in the appendfile transport. When checking directories
3188 for a mailbox, to see if any needed to be created, it was accidentally
3189 using path names with one or more superfluous leading slashes; tracing
3190 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3192 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3193 contents are added to the log line that is written for every discarded
3194 recipient. (Previously a log_message setting was ignored.)
3196 36. The ${quote: operator now quotes the string if it is empty.
3198 37. The install script runs exim in order to find its version number. If for
3199 some reason other than non-existence or emptiness, which it checks, it
3200 could not run './exim', it was installing it with an empty version number,
3201 i.e. as "exim-". This error state is now caught, and the installation is
3204 38. An argument was missing from the function that creates an error message
3205 when Exim fails to connect to the socket for saslauthd authentication.
3206 This could cause Exim to crash, or give a corrupted message.
3208 39. Added isip, isip4, and isip6 to ${if conditions.
3210 40. The ACL variables $acl_xx are now saved with the message, and can be
3211 accessed later in routers, transports, and filters.
3213 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3214 strings in the file are not string-expanded.
3216 42. If a MAIL command specified a SIZE value that was too large to fit into an
3217 int variable, the check against message_size_limit failed. Such values are
3218 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3219 day this will have to be increased, but I don't think I want to be around
3220 when emails are that large.
3227 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3228 iconv() is not standard in FreeBSD.
3230 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3231 IPv6 enabled. The observed symptom was a segmentation fault on return from
3232 the function os_common_find_running_interfaces() in src/os.c.
3234 3. In the check_special_case() function in daemon.c I had used "errno" as an
3235 argument name, which causes warnings on some systems. This was basically a
3236 typo, since it was named "eno" in the comments!
3238 4. The code that waits for the clock to tick (at a resolution of some fraction
3239 of a second) so as to ensure message-id uniqueness was always waiting for
3240 at least one whole tick, when it could have waited for less. [This is
3241 almost certainly not relevant at current processor speeds, where it is
3242 unlikely to ever wait at all. But we try to future-proof.]
3244 5. The function that sleeps for a time interval that includes fractions of a
3245 second contained a race. It did not block SIGALRM between setting the
3246 timer, and suspending (a couple of lines later). If the interval was short
3247 and the sigsuspend() was delayed until after it had expired, the suspension
3248 never ended. On busy systems this could lead to processes getting stuck for
3251 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3252 process, before it forks any delivery processes. The open lookup caching
3253 mechanism meant that the open file or database connection was passed into
3254 the delivery process. The problem was that delivery processes always tidy
3255 up cached lookup data. This could cause a problem for the next delivery
3256 process started by the queue runner, because the external queue runner
3257 process does not know about the closure. So the next delivery process
3258 still has data in the lookup cache. In the case of a file lookup, there was
3259 no problem because closing a file descriptor in a subprocess doesn't affect
3260 the parent. However, if the lookup was caching a connection to a database,
3261 the connection was closed, and the second delivery process was likely to
3262 see errors such as "PGSQL: query failed: server closed the connection
3263 unexpectedly". The problem has been fixed by closing all cached lookups
3264 in a queue runner before running a delivery process.
3266 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3267 seem to have the "const" qualifier which it has on other OS. I've
3270 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3271 *of the same type* (client or server) with the same public name that an
3272 error should be diagnosed.
3274 9. When Exim looked up a host name for an IP address, but failed to find the
3275 original IP address when looking up the host name (a safety check), it
3276 output the message "<ip address> does not match any IP for NULL", which was
3277 confusing, to say the least. The bug was that the host name should have
3278 appeared instead of "NULL".
3280 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3281 user that is built into the binary, and the -C or -D options is used, root
3282 privilege is dropped before the configuration file is read. In addition,
3283 logging is switched to stderr instead of the normal log files. If the
3284 configuration then re-defines the Exim user, the unprivileged environment
3285 is probably not what is expected, so Exim logs a panic warning message (but
3288 However, if deliver_drop_privilege is set, the unprivileged state may well
3289 be exactly what is intended, so the warning has been cut out in that case,
3290 and Exim is allowed to try to write to its normal log files.
3296 1. smtp_return_error_details was not giving details for temporary sender
3297 or receiver verification errors.
3299 2. Diagnose a configuration error if two authenticators have the same public
3302 3. Exim used not to create the message log file for a message until the first
3303 delivery attempt. This could be confusing when incoming messages were held
3304 for policy or load reasons. The message log file is now created at the time
3305 the message is received, and an initial "Received" line is written to it.
3307 4. The automatically generated man page for command line options had a minor
3308 bug that caused no ill effects; however, a more serious problem was that
3309 the procedure for building the man page automatically didn't always
3310 operate. Consequently, release 4.20 contains an out-of-date version. This
3311 shouldn't happen again.
3313 5. When building Exim with embedded Perl support, the script that builds the
3314 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3315 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3317 6. The freeze_tell option was not being used for messages that were frozen on
3318 arrival, either by an ACL or by local_scan().
3320 7. Added the smtp_incomplete_transaction log selector.
3322 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3323 was accepting AUTH without a new EHLO.
3325 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3326 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3329 10. Logging of TCP/IP connections (when configured) now happens in the main
3330 daemon process instead of the child process, so that the TCP/IP connection
3331 count is more accurate (but it can never be perfect).
3333 11. The use of "drop" in a nested ACL was not being handled correctly in the
3334 outer ACL. Now, if condition failure induced by the nested "drop" causes
3335 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3336 or "require"), the connection is dropped.
3338 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3339 that yield "discard" can now be used with an "accept" or a "discard" verb,
3340 but an error is generated for any others (because I can't see a useful way
3341 to define what should happen).
3343 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3344 are now processed in the same way as lines in the Exim configuration file.
3345 In particular, continuation lines are supported.
3347 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3349 15. Added -ti meaning -t -i.
3351 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3352 domains, and warn by logging if others are found.
3354 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3355 returned from GIFCONF when Exim is trying to find the list of interfaces on
3356 a host. The code in os.c has been modified to copy each ifreq to an aligned
3357 structure in all cases.
3359 Also, in some cases, the returned ifreq's were being copied to a 'struct
3360 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3361 means the last couple of bytes of an IPv6 address could be chopped if the
3362 ifreq contained only a normal sockaddr (14 bytes storage).
3364 18. Named domain lists were not supported in the hosts_treat_as_local option.
3365 An entry such as +xxxx was not recognized, and was treated as a literal
3368 19. Ensure that header lines added by a DATA ACL are included in the reject log
3369 if the ACL subsequently rejects the message.
3371 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3372 MD5 (which is deprecated).
3374 21. When testing a filter file using -bf, Exim was writing a message when it
3375 took the sender from a "From " line in the message, but it was not doing so
3376 when it took $return_path from a Return-Path: header line. It now does.
3378 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3379 with a valid header line field name (a series of printing characters
3380 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3382 23. Changed "disc" in the source to "disk" to conform to the documentation and
3383 the book and for uniformity.
3385 24. Ignore Sendmail's -Ooption=value command line item.
3387 25. When execve() failed while trying to run a command in a pipe transport,
3388 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3389 could be confused with a return value of 69 from the command itself. This
3390 has been changed to 127, the value the shell returns if it is asked to run
3391 a non-existent command. The wording for the related log line suggests a
3392 non-existent command as the problem.
3394 26. If received_header_text expands to an empty string, do not add a Received:
3395 header line to the message. (Well, it adds a token one on the spool, but
3396 marks it "old" so that it doesn't get used or transmitted.)
3398 27. Installed eximstats 1.28 (addition of -nt option).
3400 28. There was no check for failure on the call to getsockname() in the daemon
3401 code. This can fail if there is a shortage of resources on the system, with
3402 ENOMEM, for example. A temporary error is now given on failure.
3404 29. Contrary to the C standard, it seems that in some environments, the
3405 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3406 program. Exim now does this explicitly; it affects the formatting of
3407 timestamps using strftime().
3409 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3410 complaints. I've now initialized all the variables, to avoid this.
3412 32. Header lines added by a system filter were not being "seen" during
3413 transport-time rewrites.
3415 33. The info_callback() function passed to OpenSSL is set up with type void
3416 (*)(SSL *, int, int), as described somewhere. However, when calling the
3417 function (actually a macro) that sets it up, the type void(*)() is
3418 expected. I've put in a cast to prevent warnings from picky compilers.
3420 34. If a DNS black list lookup found a CNAME record, but there were no A
3421 records associated with the domain it pointed at, Exim crashed.
3423 35. If a DNS black list lookup returned more than one A record, Exim ignored
3424 all but the first. It now scans all returned addresses if a particular IP
3425 value is being sought. In this situation, the contents of the
3426 $dnslist_value variable are a list of all the addresses, separated by a
3429 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3430 to accept a host name and all its aliases if the forward lookup for any of
3431 them yielded the IP address of the incoming connection. Now it accepts only
3432 those names whose forward lookup yields the correct IP address. Any other
3433 names are discarded. This closes a loophole whereby a rogue DNS
3434 administrator could create reverse DNS records to break through a
3435 wildcarded host restriction in an ACL.
3437 37. If a user filter or a system filter that ran in a subprocess used any of
3438 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3439 the wrong values were passed to the pipe command ($thisaddress had the
3440 value of $0, $0 had the value of $1, etc). This bug was introduced by
3441 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3443 38. Improved the line breaking for long SMTP error messages from ACLs.
3444 Previously, if there was no break point between 40 and 75 characters, Exim
3445 left the rest of the message alone. Two changes have been made: (a) I've
3446 reduced the minimum length to 35 characters; (b) if it can't find a break
3447 point between 35 and 75 characters, it looks ahead and uses the first one
3448 that it finds. This may give the occasional overlong line, but at least the
3449 remaining text gets split now.
3451 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3452 file descriptors might be low, and that setting 1000 would always raise it.
3453 It turns out that in some environments, the limit is already over 1000 and
3454 that lowering it causes trouble. So now Exim takes care not to decrease it.
3456 40. When delivering a message, the value of $return_path is set to $sender_
3457 address at the start of routing (routers may change the value). By an
3458 oversight, this default was not being set up when an address was tested by
3459 -bt or -bv, which affected the outcome if any router or filter referred to
3462 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3463 log only when "message" or "log_message" are set. However, if one of the
3464 conditions was an address verification, or a call to a nested ACL, the
3465 messages generated by the underlying test were being passed through. This
3466 no longer happens. The underlying message is available in $acl_verify_
3467 message for both "message" and "log_message" expansions, so it can be
3468 passed through if needed.
3470 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3471 new expansion $bh_ to give the encoded byte string without charset
3472 translation. Translation happens only if iconv() is available; HAVE_ICONV
3473 indicates this at build time. HEADERS_CHARSET gives the charset to
3474 translate to; headers_charset can change it in the configuration, and
3475 "headers charset" can change it in an individual filter file.
3477 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3478 that creates RFC 2047 encoded "words" labels them as that charset instead
3479 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3480 expansion operator; (ii) when Exim creates a From: line for a local
3481 message; (iii) when a header line is rewritten to include a "phrase" part.
3483 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3484 buggy, causing it to skip the first lines of messages whose message ID
3485 ended in 'D'. This would not have bitten before Exim release 4.14, because
3486 message IDs were unlikely to end in 'D' before then. The effect was to have
3487 incorrect size information for certain domains.
3489 45. #include "config.h" was missing at the start of the crypt16.c module. This
3490 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3493 46. If there was a timeout during a "random" callout check, Exim treated it as
3494 a failure of the random address, and carried on sending RSET and the real
3495 address. If the delay was just some slowness somewhere, the response to the
3496 original RCPT would be taken as a response to RSET and so on, causing
3497 mayhem of various kinds.
3499 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3500 when I implemented it. It didn't allow for the fact that some option values
3501 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3502 the right test for positive values.
3504 48. Domain names in DNS records are case-independent. Exim always looks them up
3505 in lower case. Some resolvers return domain names in exactly the case they
3506 appear in the zone file, that is, they may contain uppercase letters. Not
3507 all resolvers do this - some return always lower case. Exim was treating a
3508 change of case by a resolver as a change of domain, similar to a widening
3509 of a domain abbreviation. This triggered its re-routing code and so it was
3510 trying to route what was effectively the same domain again. This normally
3511 caused routing to fail (because the router wouldn't handle the domain
3512 twice). Now Exim checks for this case specially, and just changes the
3513 casing of the domain that it ultimately uses when it transmits the message
3516 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3519 50. If a filter generated a file delivery with a non-absolute name (possible if
3520 no home directory exists for the router), the forbid_file option was not
3523 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3524 the existing equality matching.
3526 52. Exim was using ints instead of ino_t variables in some places where it was
3527 dealing with inode numbers.
3529 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3530 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3531 called TMPDIR, and if it finds it is different, it changes its value.
3533 54. The smtp_printf() function is now made available to local_scan() so
3534 additional output lines can be written before returning. There is also an
3535 smtp_fflush() function to enable the detection of a dropped connection.
3536 The variables smtp_input and smtp_batched_input are exported to
3539 55. Changed the default runtime configuration: the message "Unknown user"
3540 has been removed from the ACL, and instead placed on the localuser router,
3541 using the cannot_route_message feature. This means that any verification
3542 failures that generate their own messages won't get overridden. Similarly,
3543 the "Unrouteable address" message that was in the ACL for unverifiable
3544 relay addresses has also been removed.
3546 56. Added hosts_avoid_esmtp to the smtp transport.
3548 57. The exicyclog script was not checking for the esoteric option
3549 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3550 will work only if exicyclog is run under the appropriate euid.
3552 58. Following a discussion on the list, the rules by which Exim recognises line
3553 endings on incoming messages have been changed. The -dropcr and drop_cr
3554 options are now no-ops, retained only for backwards compatibility. The
3555 following line terminators are recognized: LF CRLF CR. However, special
3556 processing applies to CR:
3558 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3559 nor a local message in the state where . is a terminator.
3561 (ii) If a bare CR is encountered in a header line, an extra space is added
3562 after the line terminator so as not to end the header. The reasoning
3563 behind this is that bare CRs in header lines are most likely either
3564 to be mistakes, or people trying to play silly games.
3566 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3567 was being incorrectly given as 18 bytes larger than it should have been.
3568 This is a VOB (very old bug).
3570 60. This may never have affected anything current, but just in case it has:
3571 When the local host is found other than at the start of a list of hosts,
3572 the local host, those with the same MX, and any that follow, are discarded.
3573 When the list in question was part of a longer list of hosts, the following
3574 hosts (not currently being processed) were also being discarded. This no
3575 longer happens. I'm not sure if this situation could ever has previously
3578 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3581 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3582 header. This is something that is recommended in a new Internet Draft, and
3583 is something that is documented as being done by Sendmail. There are two
3584 possible values. For messages generated by the autoreply transport, Exim
3587 Auto-Submitted: auto-replied
3589 whereas for all other generated messages (e.g. bounces) it adds
3591 Auto-Submitted: auto-generated
3593 63. The "personal" condition in filters now includes a test for the
3594 Auto-Submitted: header. If it contains the string "auto-" the message it
3595 not considered personal.
3597 64. Added rcpt_include_affixes as a generic transport option.
3599 65. Added queue_only_override (default true).
3601 66. Added the syslog_duplication option.
3603 67. If what should have been the first header line of a message consisted of
3604 a space followed by a colon, Exim was mis-interpreting it as a header line.
3605 It isn't of course - it is syntactically invalid and should therefore be
3606 treated as the start of the message body. The misbehaviour could have
3607 caused a number of strange effects, including loss of data in subsequent
3608 header lines, and spool format errors.
3610 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3611 client host had authenticated. This control can now be exercised by an ACL
3612 for more flexibility.
3614 69. By default, callouts do not happen when testing with -bh. There is now a
3615 variant, -bhc, which does actually run the callout code, including
3616 consulting and updating the callout cache.
3618 70. Added support for saslauthd authentication, courtesy of Alexander
3621 71. If statvfs() failed on the spool or log directories while checking their
3622 size for availability, Exim confusingly gave the error "space shortage".
3623 Furthermore, in debugging mode it crashed with a floating point exception.
3624 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3625 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3626 serious problem, Exim now writes to the main and panic logs when this
3627 happens, with details of the failure. It then refuses to accept the
3628 incoming message, giving the message "spool directory problem" or "log
3629 directory problem" with a 421 code for SMTP messages.
3631 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3632 0, 1, and 2 exist, because some OS complain for execs without them (see
3633 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3634 descriptors. However, the code omitted to check that the open succeeded,
3635 causing mysterious errors if for some reason the permissions on /dev/null
3636 got screwed. Now Exim writes a message to the main and panic logs, and
3637 bombs out if it can't open /dev/null.
3639 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3640 interact so that it is all more flexible. It is supposed to remain
3641 backwards compatible. Also added extra_local_interfaces.
3643 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3644 to bomb out with an assertion failure - to the client this appears as a
3645 connection drop. This problem occurs in the part of the code that was taken
3646 from the Samba project. Fortunately, the assertion is in a very simple
3647 function, so I have fixed this by reproducing the function inline in the
3648 one place where it is called, and arranging for authentication to fail
3649 instead of killing the process with assert().
3651 75. The SPA client code was not working when the server requested OEM rather
3652 than Unicode encoding.
3654 76. Added code to make require_files with a specific uid setting more usable in
3655 the case where statting the file as root fails - usually a non-root-mounted
3656 NFS file system. When this happens and the failure is EACCES, Exim now
3657 forks a subprocess and does the per-uid checking as the relevant uid.
3659 77. Added process_log_path.
3661 78. If log_file_path was not explicitly set, a setting of check_log_space or
3662 check_log_inodes was ignored.
3664 79. If a space check for the spool or log partitions fails, the incident is now
3665 logged. Of course, in the latter case the data may get lost...
3667 80. Added the %p formatting code to string_format() so that it can be used to
3668 print addresses in debug_print(). Adjusted all the address printing in the
3669 debugging in store.c to use %p rather than %d.
3671 81. There was a concern that a line of code in smtp_in.c could overflow a
3672 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3673 initially expressed, the concern was not well-founded, because trailing
3674 spaces are removed early. However, if the trailing spaces were followed by
3675 a NULL, they did not get removed, so the overflow was possible. Two fixes
3678 (a) I re-wrote the offending code in a cleaner fashion.
3679 (b) If an incoming SMTP command contains a NULL character, it is rejected
3682 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3683 runs initgroups(), so that if the Exim user is in any additional groups,
3684 they will be used during message reception.
3690 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob