1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.294 2006/02/13 12:02:59 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.
119 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
120 $1, $2, $3) because the numerical variables can be reset during some
121 expansion items (e.g. "match"), thereby losing the authentication data.
123 PH/21 Make -bV show the size of off_t variables so that the test suite can
124 decide whether to run tests for quotas > 2G.
126 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
127 mailbox_size, and mailbox_filecount in the appendfile transport. If a
128 filecount value is greater than 2G or if a quota value is greater than 2G
129 on a system where the size of off_t is not greater than 4, a panic error
132 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
133 never match. The debug and -bh output now contains an explicit error
134 message indicating a malformed IPv4 address or mask.
136 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
137 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
140 PH/25 Do not write to syslog when running in the test harness. The only
141 occasion when this arises is a failure to open the main or panic logs
142 (for which there is an explicit test).
144 PH/26 Added the /no_tell option to "control=freeze".
151 PH/01 Two changes to the default runtime configuration:
153 (1) Move the checks for relay_from_hosts and authenticated clients from
154 after to before the (commented out) DNS black list checks.
156 (2) Add control=submission to the relay_from_hosts and authenticated
157 clients checks, on the grounds that messages accepted by these
158 statements are most likely to be submissions.
160 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
162 (1) Generate an error if the third argument for the ${prvs expansion is
165 (2) Treat a missing third argument of ${prvscheck as if it were an empty
168 (3) Reset the variables that are obtained from the first argument of
169 ${prvscheck and used in the second argument before leaving the code,
170 because their memory is reclaimed, so using them afterwards may do
173 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
174 one (it's much easier than Tom thought :-).
176 (5) Because of (4), we can now allow for the use of $prvscheck_result
177 inside the third argument.
179 PH/03 For some reason, the default setting of PATH when running a command from
180 a pipe transport was just "/usr/bin". I have changed it to
183 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
184 anything to be listed in the output from -bV.
186 PH/05 When a filter generated an autoreply, the entire To: header line was
187 quoted in the delivery log line, like this:
189 => >A.N.Other <ano@some.domain> <original@ddress> ...
191 This has been changed so that it extracts the operative address. There
192 may be more than one such address. If so, they are comma-separated, like
195 => >ano@some.domain,ona@other.domain <original@ddress> ...
197 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
198 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
199 not being looked up in the rDNS to get a host name, Exim was showing the
200 IP address twice in Received: lines, even though the IP addresses were
201 identical. For example:
203 Received: from [1.2.3.4] (helo=[1.2.3.4])
205 However, if the real host name was known, it was omitting the HELO data
206 if it matched the actual IP address. This has been tidied up so that it
207 doesn't show the same IP address twice.
209 PH/07 When both +timestamp and +memory debugging was on, the value given by
210 $tod_xxx expansions could be wrong, because the tod_stamp() function was
211 called by the debug printing, thereby overwriting the timestamp buffer.
212 Debugging no longer uses the tod_stamp() function when +timestamp is set.
214 PH/08 When the original message was included in an autoreply transport, it
215 always said "this is a copy of the message, including all the headers",
216 even if body_only or headers_only was set. It now gives an appropriate
219 PH/09 Applied a patch from the Sieve maintainer which:
221 o fixes some comments
222 o adds the (disabled) notify extension core
223 o adds some debug output for the result of if/elsif tests
224 o points to the current vacation draft in the documentation
225 and documents the missing references header update
229 o fixes a bug in processing the envelope test (when testing
230 multiple envelope elements, the last element determinted the
233 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
234 Electronic Mail") by including:
236 Auto-submitted: auto-generated
238 in the messages that it generates (bounce messages and others, such as
239 warnings). In the case of bounce messages for non-SMTP mesages, there was
240 also a typo: it was using "Auto_submitted" (underscore instead of
241 hyphen). Since every message generated by Exim is necessarily in response
242 to another message, thes have all been changed to:
244 Auto-Submitted: auto-replied
246 in accordance with these statements in the RFC:
248 The auto-replied keyword:
250 - SHOULD be used on messages sent in direct response to another
251 message by an automatic process,
253 - MUST NOT be used on manually-generated messages,
255 - MAY be used on Delivery Status Notifications (DSNs) and Message
256 Disposition Notifications (MDNs),
258 - MUST NOT be used on messages generated by automatic or periodic
259 processes, except for messages which are automatic responses to
262 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
263 to the default Received: header definition.
265 PH/12 Added log selector acl_warn_skipped (default on).
267 PH/13 After a successful wildlsearch lookup, discard the values of numeric
268 variables because (a) they are in the wrong storage pool and (b) even if
269 they were copied, it wouldn't work properly because of the caching.
271 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
272 checking when decoding. Apparently there are clients that generate
273 overlong encoded strings. Why am I not surprised?
275 PH/15 If the first argument of "${if match_address" was not empty, but did not
276 contain an "@" character, Exim crashed. Now it writes a panic log message
277 and treats the condition as false.
279 PH/16 In autoreply, treat an empty string for "once" the same as unset.
281 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
282 extension "envelope-auth". The code is finished and in agreement with
283 other implementations, but there is no documentation so far and in fact,
284 nobody wrote the draft yet. This extension is currently #undef'ed, thus
285 not changing the active code.
287 Print executed "if" and "elsif" statements when debugging is used. This
288 helps a great deal to understand what a filter does.
290 Document more things not specified clearly in RFC3028. I had all this
291 sorted out, when out of a sudden new issues came to my mind. Oops."
293 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
296 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
297 canonical form (as documented). However, after a host name lookup from
298 the IP address, check_host() was doing a simple string comparison with
299 addresses acquired from the DNS when checking that the found name did
300 have the original IP as one of its addresses. Since any found IPv6
301 addresses are likely to be in abbreviated form, the comparison could
302 fail. Luckily, there already exists a function for doing the comparison
303 by converting both addresses to binary, so now that is used instead of
306 PH/20 There was another similar case to PH/19, when a complete host name was
307 given in a host list; looking up its IP address could give an abbreviated
308 form, whereas the current host's name might or might not be abbreviated.
309 The same fix has been applied.
315 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
316 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
319 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
320 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
322 PH/03 Typo: missing ".o" in src/pcre/Makefile.
324 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
325 header line, restrict the check to what is listed in RFCs 2369 and 2929.
326 Also, for "Auto-Submitted", treat anything other than "no" as
327 non-personal, in accordance with RFC 3834. (Previously it treated
328 anything starting "auto-" as non-personal.)
330 TF/01 The control=submission/name=... option had a problem with syntax
331 errors if the name included a slash character. The /name= option
332 now slurps the rest of the string, so it can include any characters
333 but it must come last in the list of options (after /sender_retain
336 PH/05 Some modifications to the interface to the fake nameserver for the new
344 TK/01 Added the "success_on_redirect" address verification option. See
345 NewStuff for rationale and an example.
347 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
349 PH/02 Patch to exigrep to allow it to work on syslog lines.
351 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
352 fread() to skip over the body file's header line, because in Cygwin the
353 header line is locked and is inaccessible.
355 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
356 co-exist for some time) to make it clear that it is the Exim ID that is
357 referenced, not the Message-ID: header line.
359 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
360 string_format() function, because snprintf() does not exist on all
363 PH/06 The use of forbid_filter_existstest now also locks out the use of the
364 ${stat: expansion item.
366 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
367 protocol synchronization error", to keep the pedants happy.
369 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
370 well as for IRIX systems, when gcc is being used. See the host.c source
373 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
375 PH/10 Named domain lists were not working if used in a queue_smtp_domains
378 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
379 transport and to the smtp transport in LMTP mode.
381 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
383 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
384 run a filter in a subprocess. This could lead to confusion in subsequent
385 lookups in the parent process. There should also be a search_tidyup() at
386 the end of the subprocess.
388 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
389 only if the host matched helo_try_verify_hosts, which caused the
390 verification to occur when the EHLO/HELO command was issued. The ACL just
391 tested the remembered result. Now, if a previous verification attempt has
392 not happened, "verify = helo" does it there and then.
394 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
396 TK/03 Fix log output including CR from clamd.
398 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
399 exist provoked a memory error which could cause a segfault.
401 PH/15 Installed PCRE 6.2
403 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
405 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
406 of the problem. Specifically, suggested +O2 rather than +O1 for the
409 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
411 PH/20 If a delivery was routed to a non-standard port by means of an SRV
412 record, the port was not correctly logged when the outgoing_port log
413 selector was set (it logged the transort's default port).
415 PH/21 Added support for host-specific ports to manualroute, queryprogram,
416 fallback_hosts, and "hosts" in the smtp transport.
418 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
419 host errors such as "Connection refused".
421 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
422 authentication with radiusclient 0.4.9:
424 - Error returned from rc_read_config was caught wrongly
425 - Username/password not passed on to radius server due to wrong length.
427 The presumption is that some radiusclient API changes for 4.51/PH/17
428 were not taken care of correctly. The code is still untested by me (my
429 Linux distribution still has 0.3.2 of radiusclient), but it was
430 contributed by a Radius user.
432 PH/24 When doing a callout, the value of $domain wasn't set correctly when
433 expanding the "port" option of the smtp transport.
435 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
436 while reading a MIME header. Thanks to Tom Hughes for a patch.
438 PH/24 Include config.h inside local_scan.h so that configuration settings are
441 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
442 that in an ACL for RCPT (for example), you can examine exactly what was
445 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
446 commands, but it was not correctly comparing the address with the actual
447 client host address. Thus, it would show the EHLO address in Received:
448 header lines when this was not necessary.
450 PH/27 Added the % operator to ${eval:}.
452 PH/28 Exim tries to create and chdir to its spool directory when it starts;
453 it should be ignoring failures (because with -C, for example, it has lost
454 privilege). It wasn't ignoring creation failures other than "already
457 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
460 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
461 because an input line was too long, either on its own, or by virtue of
462 too many continuations, the temporary file was not being removed, and the
463 return code was incorrect.
465 PH/31 Missing "BOOL" in function definition in filtertest.c.
467 PH/32 Applied Sieve patches from the maintainer.
469 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
471 PH/33 Added "verify = not_blind".
473 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
474 Local/Makefile (with some defaults set). These are used in built scripts
475 such as exicyclog, but they have never been used in the exim_install
476 script (though there are many overriding facilities there). I have
477 arranged that the exim_install script now takes note of these two
480 PH/35 Installed configuration files for Dragonfly.
482 PH/36 When a locally submitted message by a trusted user did not contain a
483 From: header, and the sender address was obtained from -f or from an SMTP
484 MAIL command, and the trusted user did not use -F to supply a sender
485 name, $originator_name was incorrectly used when constructing a From:
486 header. Furthermore, $originator_name was used for submission mode
487 messages from external hosts without From: headers in a similar way,
488 which is clearly wrong.
490 PH/37 Added control=suppress_local_fixups.
492 PH/38 When log_selector = +received_sender was set, and the addition of the
493 sender made the log line's construction buffer exactly full, or one byte
494 less than full, an overflow happened when the terminating "\n" was
497 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
498 when the result of a list match is failure because a DNS lookup failed.
500 PH/40 RM_COMMAND is now used in the building process.
502 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
503 the "build-* directories that it finds.
505 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
506 address in a domain literal was a prefix of an interface address.
508 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
509 when verifying a sender address, unless rewrite_headers is false.
511 PH/44 Wrote a long comment about why errors_to addresses are verified as
512 recipients, not senders.
514 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
515 the ratelimit ACL was added.
517 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
519 PH/46 Added extra information about PostgreSQL errors to the error string.
521 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
522 avoiding the need to install special zones in a real server. This is
523 backwards compatible; if it can't find the fake resolver, it drops back.
524 Thus, both old and new test suites can be run.
526 TF/02 Added util/ratelimit.pl
528 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
529 clock is set back in time.
531 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
532 Candler <B.Candler@pobox.com>.
534 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
535 OK for addresses that are the result of redirections.
537 PH/48 A number of further additions for the benefit of the new test suite,
538 including a fake gethostbyname() that interfaces to the fake DNS resolver
541 TF/06 The fix for widen_domains has also been applied to qualify_single and
542 search_parents which are the other dnslookup options that can cause
545 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
548 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
550 TF/07 Exim produced the error message "an SRV record indicated no SMTP
551 service" if it encountered an MX record with an empty target hostname.
552 The message is now "an MX or SRV record indicated no SMTP service".
554 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
555 if the DNS of the sending site is misconfigured. This is quite a
556 common situation. This change restores the behaviour of treating a
557 helo verification defer as a failure.
559 PH/51 If self=fail was set on a router, the bounce message did not include the
560 actual error message.
566 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
568 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
569 command itself ended in error, the underlying message about the transport
570 filter timeout was being overwritten with the pipe command error. Now the
571 underlying error message should be appended to the second error message.
573 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
575 PH/02 Reduce the amount of output that "make" produces by default. Full output
576 can still be requested.
578 PH/03 The warning log line about a condition test deferring for a "warn" verb
579 was being output only once per connection, rather than after each
580 occurrence (because it was using the same function as for successful
581 "warn" verbs). This seems wrong, so I have changed it.
583 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
584 it should not have, which might have caused a crash in the right
585 circumstances, but probably never did.
587 PH/04 Installed a modified version of Tony Finch's patch to make submission
588 mode fix the return path as well as the Sender: header line, and to
589 add a /name= option so that you can make the user's friendly name appear
592 TF/03 Added the control = fakedefer ACL modifier.
594 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
595 Mark Lowes for thorough testing.
597 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
599 TK/03 Merged latest SRS patch from Miles Wilton.
601 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
602 with the definition in sysexits.h (which is #included earlier).
603 Fortunately, Exim does not actually use EX_OK. The code used to try to
604 preserve the sysexits.h value, by assumimg that macro definitions were
605 scanned for macro replacements. I have been disabused of this notion,
606 so now the code just undefines EX_OK before #including unistd.h.
608 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
609 in the smtp transport. When a block could not be written in a single
610 write() function, the timeout was being re-applied to each part-write.
611 This seems wrong - if the receiver was accepting one byte at a time it
612 would take for ever. The timeout is now adjusted when this happens. It
613 doesn't have to be particularly precise.
615 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
616 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
618 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
619 without a domain if the check to <postmaster@domain> fails.
621 SC/01 Eximstats: added -xls and the ability to specify output files
622 (patch written by Frank Heydlauf).
624 SC/02 Eximstats: use FileHandles for outputing results.
626 SC/03 Eximstats: allow any combination of xls, txt, and html output.
628 SC/04 Eximstats: fixed display of large numbers with -nvr option
630 SC/05 Eximstats: fixed merging of reports with empty tables.
632 SC/06 Eximstats: added the -include_original_destination flag
634 SC/07 Eximstats: removed tabs and trailing whitespace.
636 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
638 TK/06 MBOX spool code: Add real "From " MBOX separator line
639 so the .eml file is really in mbox format (even though
640 most programs do not really care). Patch from Alex Miller.
642 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
643 The latter is generated from $received_to and is only set if the
644 message has one envelope recipient. SA can use these headers,
645 obviously out-of-the-box. Patch from Alex Miller.
647 PH/08 The ${def test on a variable was returning false if the variable's
648 value was "0", contrary to what the specification has always said!
649 The result should be true unless the variable is empty.
651 PH/09 The syntax error of a character other than { following "${if
652 def:variable_name" (after optional whitespace) was not being diagnosed.
653 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
654 accidental colon was present, for example, could give incorrect results.
656 PH/10 Tidied the code in a number of places where the st_size field of a stat()
657 result is used (not including appendfile, where other changes are about
660 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
661 This involved changing a lot of size variables from int to off_t. It
662 should work with maildirs and everything.
664 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
665 spamd dying while we are connected to it.
667 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
670 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
671 function for BDB 4.3.
673 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
675 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
676 the sources for PCRE have been re-arranged and more files are now
679 PH/15 The code I had for printing potentially long long variables in PH/11
680 above was not the best (it lost precision). The length of off_t variables
681 is now inspected at build time, and an appropriate printing format (%ld
682 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
683 to be "long long int" or "long int". This is needed for the internal
684 formatting function string_vformat().
686 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
687 the configuration file to be ":syslog", then the script "guesses" where
688 the logs files are, rather than using the compiled in default. In our
689 case the guess is not the same as the compiled default, so the script
690 suddenly stopped working when I started to use syslog. The patch checks
691 to see if log_file_path is "". If so, it attempts to read it from exim
692 with no configuration file to get the compiled in version, before it
693 falls back to the previous guessing code."
695 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
696 implementing BATV in an Exim configuration. See NewStuff for the gory
699 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
700 Makefile that are specific to HP-UX.
702 PH/18 If the "use_postmaster" option was set for a recipient callout together
703 with the "random" option, the postmaster address was used as the MAIL
704 FROM address for the random test, but not for the subsequent recipient
705 test. It is now used for both.
707 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
708 patch removes a few documentation additions to RFC 3028, because the
709 latest draft now contains them. It adds the new en;ascii-case comparator
710 and a new error check for 8bit text in MIME parts. Comparator and
711 require names are now matched exactly. I enabled the subaddress
712 extension, but it is not well tested yet (read: it works for me)."
714 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
715 rework some of the code of TK/09 above to avoid the hardwired use of
716 "%lld" and "long long". Replaced the call to snprintf() with a call to
719 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
720 records point to non-existent hosts", "retry timeout exceeded", and
721 "retry time not reached for any host after a long failure period".
723 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
724 experimental DomainKeys support:
726 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
727 (2) On an error such as an illegally used "control", the wrong name for
728 the control was given.
730 These problems did NOT occur unless DomainKeys support was compiled.
732 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
734 PH/24 Added ${if match_ip condition.
736 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
737 databases so that it will be absolutely obvious if a crash occurs in the
738 DB library. This is a regular occurrence (often caused by mis-matched
741 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
742 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
743 picked up on a user's system that detects such things. There doesn't seem
744 to be a gcc warning option for this - only an attribute that has to be
745 put on the function's prototype. It seems that in Fedora Core 4 they have
746 set this on a number of new functions. No doubt there will be more in due
749 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
750 specify a transport. However, if an address that was verified by such a
751 router was the subject of a callout, Exim crashed because it tried to
752 read the rcpt_include_affixes from the non-existent transport. Now it
753 just assumes that the setting of that option is false. This bug was
754 introduced by 4.51/PH/31.
756 PH/28 Changed -d+all to exclude +memory, because that information is very
757 rarely of interest, but it makes the output a lot bigger. People tend to
758 do -d+all out of habit.
760 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
761 code in os-type was giving problems when libc.so lives in lib64, like on
764 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
765 aren't the modern standard, and it seems that some systems' include files
766 don't always have them. Exim was already checking for some of the newer
767 ones like T_AAAA, and defining it itself. I've added checks for all the
768 record types that Exim uses.
770 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
771 not automatically generating a new one, as it is supposed to. This
772 prevented TLS from working. If the file did exist, but contained invalid
773 data, a new version was generated, as expected. It was only the case of a
774 non-existent file that was broken.
776 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
777 with a change in libdomainkeys > 0.64.
779 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
780 from DNS. If the selector record carries the flag, it now has
781 precedence over the domain-wide flag.
783 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
785 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
786 the use of a port name that isn't defined in /etc/services) occurred, the
787 message was deferred as in a normal delivery, and thus remained on the
788 spool, instead of being failed because of the mua_wrapper setting. This
789 is now fixed, and I tidied up some of the mua_wrapper messages at the
792 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
793 the messages in a hash of arrays rather than using individual hashes.
794 This is a bit cleaner and results in dramatic memory savings, albeit
795 at a slight CPU cost.
797 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
798 as requested by Marc Sherman.
800 SC/10 Eximstats: added histograms for user specified patterns as requested
803 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
805 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
806 fopen() in the content-scanning modules that did not already have it.
812 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
813 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
815 TK/02 Fix ACL "control" statement not being available in MIME ACL.
817 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
819 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
820 to test Sieve filters that use "vacation".
822 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
823 that changes the way the GnuTLS parameters are stored in the cache file.
824 The new format can be generated externally. For backward compatibility,
825 if the data in the cache doesn't make sense, Exim assumes it has read an
826 old-format file, and it generates new data and writes a new file. This
827 means that you can't go back to an older release without removing the
830 PH/03 A redirect router that has both "unseen" and "one_time" set does not
831 work if there are any delivery delays because "one_time" forces the
832 parent to be marked "delivered", so its unseen clone is never tried
833 again. For this reason, Exim now forbids the simultaneous setting of
836 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
837 redirected to themselves ("homonym" addresses). Read the long ChangeLog
838 entry if you want to know the details. The fix, however, neglected to
839 consider the case when local delivery batching is involved. The test for
840 "previously delivered" was not happening when checking to see if an
841 address could be batched with a previous (undelivered) one; under
842 certain circumstances this could lead to multiple deliveries to the same
845 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
846 in its include files, and this causes problems building Exim.
848 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
849 header_syntax) but Exim was just ignoring anything given after a slash.
850 In particular, this caused confusion with an attempt to use "verify =
851 reverse_host_lookup/defer_ok". An error is now given when options are
852 supplied for verify items that do not have them. (Maybe reverse_host_
853 lookup should have a defer_ok option, but that's a different point.)
855 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
856 defined by RFC 821) to 2048, because there were problems with some AUTH
857 commands, and RFC 1869 says the size should be increased for extended
858 SMTP commands that take arguments.
860 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
863 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
864 "unknown" error; now it says that the functionality isn't in the binary.
866 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
867 an address' error message when a string expansion fails (syntax or
868 whatever). Otherwise the password may appear in the log. Following change
869 PH/42 below, there is no longer a chance of it appearing in a bounce
872 PH/11 Installed exipick version 20050225.0 from John Jetmore.
874 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
875 of its addresses was ever tried. (Bugzilla bug #2.)
877 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
878 the result incorrectly in the debug output. (It correctly added a newline
879 to what was transported.)
881 TF/01 Added $received_time.
883 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
884 commented out examples of how to interface to a virus scanner and to
885 SpamAssassin. Also added commented examples of av_scanner and
886 spamd_address settings.
888 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
889 and controls are allowed in which ACLs. There were a couple of minor
890 errors. Some of the entries in the conditions table (which is a table of
891 where they are NOT allowed) were getting very unwieldy; rewrote them as a
892 negation of where the condition IS allowed.
894 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
896 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
897 header file does not have a version number, so I've had to invent a new
898 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
899 API. The code is untested by me (my Linux distribution still has 0.3.2 of
900 radiusclient), but it was contributed by a Radius user.
902 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
903 files or directories, for OpenSSL.
905 PH/19 When an Exim process that is running as root has to create an Exim log
906 file, it does so in a subprocess that runs as exim:exim so as to get the
907 ownership right at creation (otherwise, other Exim processes might see
908 the file with the wrong ownership). There was no test for failure of this
909 fork() call, which would lead to the process getting stuck as it waited
910 for a non-existent subprocess. Forks do occasionally fail when resources
911 run out. I reviewed all the other calls to fork(); they all seem to check
914 PH/20 When checking for unexpected SMTP input at connect time (before writing
915 the banner), Exim was not dealing correctly with a non-positive return
916 from the read() function. If the client had disconnected by this time,
917 the result was a log entry for a synchronization error with an empty
918 string after "input=" when read() returned zero. If read() returned -1
919 (an event I could not check), uninitialized data bytes were printed.
920 There were reports of junk text (parts of files, etc) appearing after
923 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
925 PH/22 Added support for macro redefinition, and (re)definition in between
926 driver and ACL definitions.
928 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
929 forgetting to use the resulting value; it was using the unexpanded value.
931 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
932 hadn't been configured. The fix is from Juergen Kreileder, who
933 understands it better than I do:
935 "Here's what I see happening with three configured cyrus_sasl
936 authenticators configured (plain, login, cram-md5):
938 On startup auth_cyrus_sasl_init() gets called for each of these.
939 This means three calls to sasl_listmech() without a specified mech_list.
940 => SASL tests which mechs of all available mechs actually work
941 => three warnings about OTP not working
942 => the returned list contains: plain, login, cram-md5, digest-md5, ...
944 With the patch, sasl_listmech() also gets called three times. But now
945 SASL's mech_list option is set to the server_mech specified in the the
946 authenticator. Or in other words, the answer from sasl_listmech()
947 gets limited to just the mech you're testing for (which is different
949 => the return list contains just 'plain' or 'login', 'cram-md5' or
950 nothing depending on the value of ob->server_mech.
952 I've just tested the patch: Authentication still works fine,
953 unavailable mechs specified in the exim configuration are still
954 caught, and the auth.log warnings about OTP are gone."
956 PH/25 When debugging is enabled, the contents of the command line are added
957 to the debugging output, even when log_selector=+arguments is not
960 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
961 answer is "GNU", and only if the return is "GNU/something" is the answer
964 PH/27 $acl_verify_message is now set immediately after the failure of a
965 verification in an ACL, and so is available in subsequent modifiers. In
966 particular, the message can be preserved by coding like this:
968 warn !verify = sender
969 set acl_m0 = $acl_verify_message
971 Previously, $acl_verify_message was set only while expanding "message"
972 and "log_message" when a very denied access.
974 PH/28 Modified OS/os.c-Linux with
976 -#ifndef OS_LOAD_AVERAGE
977 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
979 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
980 nomenclature these days.)
982 PH/29 Installed patch from the Sieve maintainer that adds the options
983 sieve_useraddress and sieve_subaddress to the redirect router.
985 PH/30 In these circumstances:
986 . Two addresses routed to the same list of hosts;
987 . First host does not offer TLS;
988 . First host accepts first address;
989 . First host gives temporary error to second address;
990 . Second host offers TLS and a TLS session is established;
991 . Second host accepts second address.
992 Exim incorrectly logged both deliveries with the TLS parameters (cipher
993 and peerdn, if requested) that were in fact used only for the second
996 PH/31 When doing a callout as part of verifying an address, Exim was not paying
997 attention to any local part prefix or suffix that was matched by the
998 router that accepted the address. It now behaves in the same way as it
999 does for delivery: the affixes are removed from the local part unless
1000 rcpt_include_affixes is set on the transport.
1002 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1003 timeout during the DATA phase of an incoming message.
1005 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1006 applied a patch sanctioned by the Sieve maintainer.
1008 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1009 the uid or gid is negative. A case of a negative gid caused this to be
1010 noticed. The fix allows for either to be negative.
1012 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1013 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1016 PH/36 The change PH/12 above was broken. Fixed it.
1018 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1019 the grounds that routing the same address twice would always produce the
1020 same answer. This might have been true once, but it is certainly no
1021 longer true now. Routing a child address may depend on the previous
1022 routing that produced that child. Some complicated redirection strategies
1023 went wrong when messages had multiple recipients, and made Exim's
1024 behaviour dependent on the order in which the addresses were given.
1026 I have moved the duplicate checking until after the routing is complete.
1027 Exim scans the addresses that are assigned to local and remote
1028 transports, and removes any duplicates. This means that more work will be
1029 done, as duplicates will always all be routed, but duplicates are
1030 presumably rare, so I don't expect this is of any significance.
1032 For deliveries to pipes, files, and autoreplies, the duplicate checking
1033 still happens during the routing process, since they are not going to be
1036 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1037 It corrects a timeout issue with spamd. This is Ian's comment: "The
1038 background is that sometimes spamd either never reads data from a
1039 connection it has accepted, or it never writes response data. The exiscan
1040 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1041 blindly assumes that writes won't block so it may never time out."
1043 PH/39 Allow G after quota size as well as K and M.
1045 PH/40 The value set for $authenticated_id in an authenticator may not contain
1046 binary zeroes or newlines because the value is written to log lines and
1047 to spool files. There was no check on this. Now the value is run through
1048 the string_printing() function so that such characters are converted to
1049 printable escape sequences.
1051 PH/41 $message_linecount is a new variable that contains the total number of
1052 lines in the message. Compare $body_linecount, which is the count for the
1055 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1056 bounce and delay warning messages, except in certain special cases, which
1059 (a) An SMTP error message from a remote host;
1060 (b) A message specified in a :fail: redirection;
1061 (c) A message specified in a "fail" command in a system filter;
1062 (d) A message specified in a FAIL return from the queryprogram router;
1063 (e) A message specified by the cannot_route_message router option.
1065 In these cases only, Exim does include the error details in bounce and
1066 warning messages. There are also a few cases where bland messages such
1067 as "unrouteable address" or "local delivery error" are given.
1069 PH/43 $value is now also set for the "else" part of a ${run expansion.
1071 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1072 being worked on, but at least Exim now implements the latest version to
1075 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1076 process to complete was treated as a delivery failure, a timeout while
1077 writing the message to the pipe was logged, but erroneously treated as a
1078 successful delivery. Such timeouts include transport filter timeouts. For
1079 consistency with the overall process timeout, these timeouts are now
1080 treated as errors, giving rise to delivery failures by default. However,
1081 there is now a new Boolean option for the pipe transport called
1082 timeout_defer, which, if set TRUE, converts the failures into defers for
1083 both kinds of timeout. A transport filter timeout is now identified in
1086 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1087 systems where "make" and "gmake" are different, calling "gmake" at top
1088 level broke things. I've arranged for the value of $(MAKE) to be passed
1089 from the Makefile to this script so that it can call the same version of
1093 A note about Exim versions 4.44 and 4.50
1094 ----------------------------------------
1096 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1097 changes of various kinds. As a consequence, a big documentation update was
1098 needed. This delayed the release for rather longer than seemed good, especially
1099 in the light of a couple of (minor) security issues. Therefore, the changes
1100 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1101 release. So 4.44 and 4.50 are in effect two different branches that both start
1104 I have left the 4.50 change log unchanged; it contains all the changes since
1105 4.43. The change log for 4.44 is below; many of its items are identical to
1106 those for 4.50. This seems to be the most sensible way to preserve the
1107 historical information.
1113 1. Minor wording change to the doc/README.SIEVE file.
1115 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1116 computation of the current number of files was incorrect.
1118 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1119 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1122 4. Give more explanation in the error message when the command for a transport
1123 filter fails to execute.
1125 5. There are several places where Exim runs a non-Exim command in a
1126 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1127 was being done only for the command run by the queryprogram router. It is
1128 now done for all such subprocesses. The other cases are: ${run, transport
1129 filters, and the commands run by the lmtp and pipe transports.
1131 6. Added CONFIGURE_GROUP build-time option.
1133 7. Some older OS have a limit of 256 on the maximum number of file
1134 descriptors. Exim was using setrlimit() to set 1000 as a large value
1135 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1136 systems. I've change it so that if it can't get 1000, it tries for 256.
1138 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1139 was an oversight, and furthermore, ever since the addition of extra
1140 controls (e.g. 4.43/32), the checks on when to allow different forms of
1141 "control" were broken. There should now be diagnostics for all cases when a
1142 control that does not make sense is encountered.
1144 9. Added the /retain_sender option to "control=submission".
1146 10. $recipients is now available in the predata ACL (oversight).
1148 11. Tidy the search cache before the fork to do a delivery from a message
1149 received from the command line. Otherwise the child will trigger a lookup
1150 failure and thereby defer the delivery if it tries to use (for example) a
1151 cached ldap connection that the parent has called unbind on.
1153 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1154 of $address_data from the recipient verification was clobbered by the
1155 sender verification.
1157 13. The value of address_data from a sender verification is now available in
1158 $sender_address_data in subsequent conditions in the ACL statement.
1160 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1162 15. Added a new option "connect=<time>" to callout options, to set a different
1165 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1166 was its contents. (It was OK if the option was not defined at all.)
1168 17. A "Completed" log line is now written for messages that are removed from
1169 the spool by the -Mrm option.
1171 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1172 information about exactly what failed.
1174 19. Added -dd to debug only the daemon process.
1176 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1177 handles timeouts, both on the server side and network timeouts. Renamed the
1178 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1180 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1181 It is now set to "smtps".
1183 22. $host_address is now set to the target address during the checking of
1184 ignore_target_hosts.
1186 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1187 being passed; this would have caused $sender_host_name to have been used if
1188 matching the list had actually called for a host name (not very likely,
1189 since this list is usually IP addresses). A host name is now passed as
1192 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1193 code that specifies a non-restarting handler (typically sigaction() in
1194 modern systems) in an attempt to fix a rare and obscure crash bug.
1196 25. Narrowed the window for a race in the daemon that could cause it to ignore
1197 SIGCHLD signals. This is not a major problem, because they are used only to
1198 wake it up if nothing else does.
1200 26. A malformed maildirsize file could cause Exim to calculate negative values
1201 for the mailbox size or file count. Odd effects could occur as a result.
1202 The maildirsize information is now recalculated if the size or filecount
1205 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1206 support for a long time. Removed HAVE_SYS_VFS_H.
1208 28. Installed the latest version of exipick from John Jetmore.
1210 29. In an address list, if the pattern was not a regular expression, an empty
1211 subject address (from a bounce message) matched only if the pattern was an
1212 empty string. Non-empty patterns were not even tested. This was the wrong
1213 because it is perfectly reasonable to use an empty address as part of a
1214 database query. An empty address is now tested by patterns that are
1215 lookups. However, all the other forms of pattern expect the subject to
1216 contain a local part and a domain, and therefore, for them, an empty
1217 address still always fails if the pattern is not itself empty.
1219 30. Exim went into a mad DNS loop when attempting to do a callout where the
1220 host was specified on an smtp transport, and looking it up yielded more
1221 than one IP address.
1223 31. Re-factored the code for checking spool and log partition space into a
1224 function that finds that data and another that does the check. The former
1225 is then used to implement four new variables: $spool_space, $log_space,
1226 $spool_inodes, and $log_inodes.
1228 32. The RFC2047 encoding function was originally intended for short strings
1229 such as real names; it was not keeping to the 75-character limit for
1230 encoded words that the RFC imposes. It now respects the limit, and
1231 generates multiple encoded words if necessary. To be on the safe side, I
1232 have increased the buffer size for the ${rfc2047: expansion operator from
1235 33. It is now permitted to omit both strings after an "if" condition; if the
1236 condition is true, the result is "true". As before, when the second string
1237 is omitted, a false condition yields an empty string. This makes it less
1238 cumbersome to write custom ACL and router conditions.
1240 34. Failure to deliver a bounce message always caused it to be frozen, even if
1241 there was an errors_to setting on the router. The errors_to setting is now
1244 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1245 canonical form (fully expanded) before being placed in
1246 $sender_host_address.
1248 36. The table in the code that translates DNS record types into text (T_A to
1249 "A" for instance) was missing entries for NS and CNAME. It is just possible
1250 that this could have caused confusion if both these types were looked up
1251 for the same domain, because the text type is used as part of Exim's
1252 per-process caching. But the chance of anyone hitting this buglet seems
1255 37. The dnsdb lookup has been extended in a number of ways.
1257 (1) There is a new type, "zns", which walks up the domain tree until it
1258 finds some nameserver records. It should be used with care.
1260 (2) There is a new type, "mxh", which is like "mx" except that it returns
1261 just the host names, not the priorities.
1263 (3) It is now possible to give a list of domains (or IP addresses) to be
1264 looked up. The behaviour when one of the lookups defers can be
1265 controlled by a keyword.
1267 (4) It is now possible to specify the separator character for use when
1268 multiple records are returned.
1270 38. The dnslists ACL condition has been extended: it is now possible to supply
1271 a list of IP addresses and/or domains to be looked up in a particular DNS
1274 39. Added log_selector=+queue_time_overall.
1276 40. When running the queue in the test harness, wait just a tad after forking a
1277 delivery process, to get repeatability of debugging output.
1279 41. Include certificate and key file names in error message when GnuTLS fails
1280 to set them up, because the GnuTLS error message doesn't include the name
1281 of the failing file when there is a problem reading it.
1283 42. Allow both -bf and -bF in the same test run.
1285 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1287 44. The "Exiscan patch" is now merged into the mainline Exim source.
1289 45. Sometimes the final signoff response after QUIT could fail to get
1290 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1291 < 0 before doing a fflush(). This bug looks as though it goes back to the
1292 introduction of TLS in release 3.20, but "sometimes" must have been rare
1293 because the tests only now provoked it.
1295 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1296 (this can affect the format of dates).
1298 47. exim_tidydb, when checking for the continued existence of a message for
1299 which it has found a message-specific retry record, was not finding
1300 messages that were in split spool directories. Consequently, it was
1301 deleting retry records that should have stayed in existence.
1303 48. Steve fixed some bugs in eximstats.
1305 49. The SPA authentication driver was not abandoning authentication and moving
1306 on to the next authenticator when an expansion was forced to fail,
1307 contradicting the general specification for all authenticators. Instead it
1308 was generating a temporary error. It now behaves as specified.
1310 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1311 (the order specifies the preference for clients). The order is now AES256,
1312 AES128, 3DES, ARCFOUR128.
1314 51. Small patch to Sieve code - explicitly set From: when generating an
1317 52. Exim crashed if a remote delivery caused a very long error message to be
1318 recorded - for instance if somebody sent an entire SpamAssassin report back
1319 as a large number of 550 error lines. This bug was coincidentally fixed by
1320 increasing the size of one of Exim's internal buffers (big_buffer) that
1321 happened as part of the Exiscan merge. However, to be on the safe side, I
1322 have made the code more robust (and fixed the comments that describe what
1325 53. Now that there can be additional text after "Completed" in log lines (if
1326 the queue_time_overall log selector is set), a one-byte patch to exigrep
1327 was needed to allow it to recognize "Completed" as not the last thing in
1330 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1331 patch that reportedly fixes this has been added. I am not expert enough to
1332 create a test for it. This is what the patch creator wrote:
1334 "I found a little strange behaviour of ldap code when working with
1335 Windows 2003 AD Domain, where users was placed in more than one
1336 Organization Units. When I tried to give exim partial DN, the exit code
1337 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1338 But simultaneously result of request was absolutely normal ldap result,
1339 so I produce this patch..."
1341 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1342 REFERENCE, so I have modified the code to exclude the patch when that macro
1345 55. Some experimental protocols are using DNS PTR records for new purposes. The
1346 keys for these records are domain names, not reversed IP addresses. The
1347 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1348 leaves it alone. Component reversal etc. now happens only for IP addresses.
1351 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1353 57. Double the size of the debug message buffer (to 2048) so that more of very
1354 long debug lines gets shown.
1356 58. The exicyclog utility now does better if the number of log files to keep
1357 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1359 59. Two changes related to the smtp_active_hostname option:
1361 (1) $smtp_active_hostname is now available as a variable.
1362 (2) The default for smtp_banner uses $smtp_active_hostname instead
1363 of $primary_hostname.
1365 60. The host_aton() function is supposed to be passed a string that is known
1366 to be a valid IP address. However, in the case of IPv6 addresses, it was
1367 not checking this. This is a hostage to fortune. Exim now panics and dies
1368 if the condition is not met. A case was found where this could be provoked
1369 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1370 components; fortuitously, this particular loophole had already been fixed
1371 by change 4.50/55 above.
1373 If there are any other similar loopholes, the new check in host_aton()
1374 itself should stop them being exploited. The report I received stated that
1375 data on the command line could provoke the exploit when Exim was running as
1376 exim, but did not say which command line option was involved. All I could
1377 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1378 running as the user.
1381 61. There was a buffer overflow vulnerability in the SPA authentication code
1382 (which came originally from the Samba project). I have added a test to the
1383 spa_base64_to_bits() function which I hope fixes it.
1386 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1387 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1389 63. The daemon start-up calls getloadavg() while still root for those OS that
1390 need the first call to be done as root, but it missed one case: when
1391 deliver_queue_load_max is set with deliver_drop_privilege. This is
1392 necessary for the benefit of the queue runner, because there is no re-exec
1393 when deliver_drop_privilege is set.
1395 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1396 This has been fixed.
1398 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1399 was in use, was not putting the data itself into the right store pool;
1400 consequently, it could be overwritten for a subsequent message in the same
1401 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1404 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1406 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1407 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1408 the function were treating the return as a boolean value, which happened to
1409 work because 0=false and not-0=true, but is not correct code.
1411 68. The host_aton() function was not handling scoped IPv6 addresses (those
1412 with, for example, "%eth0" on the end) correctly.
1414 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1415 negated items (that is, ~something) in unsigned ints. Some compilers
1416 apparently mutter when there is no cast.
1418 70. If an address verification called from an ACL failed, and did not produce a
1419 user-specific message (i.e. there was only a "system" message), nothing was
1420 put in $acl_verify_message. In this situation, it now puts the system
1423 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1424 session; change 4.31/43 added the unwanted input to the log line - except
1425 that it did not do this in the start of session case. It now does.
1427 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1428 This is wrong and can cause the other end to generate a synchronization
1429 error if it is another Exim or anything else that does the synchronization
1430 check. A QUIT command is no longer sent after a timeout.
1432 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1433 during host lookups.
1435 74. The defer_ok option of callout verification was not working if it was used
1436 when verifying addresses in header lines, that is, for this case:
1438 verify = header_sender/callout=defer_ok
1440 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1441 those file descriptors could be used for SMTP connections. If anything
1442 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1443 could be sent to the SMTP client, causing chaos. The daemon now opens
1444 stdin, stdout, and stderr to /dev/null when it puts itself into the
1447 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1448 log by default. The user can override this with suitable Perl magic.
1450 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1451 the log message when discard triggers, was not working for the DATA ACL or
1452 for the non-SMTP ACL.
1454 78. Error message wording change in sieve.c.
1456 79. If smtp_accept_max_per_host was set, the number of connections could be
1457 restricted to fewer than expected, because the daemon was trying to set up
1458 a new connection before checking whether the processes handling previous
1459 connections had finished. The check for completed processes is now done
1460 earlier. On busy systems, this bug wouldn't be noticed because something
1461 else would have woken the daemon, and it would have reaped the completed
1464 80. If a message was submitted locally by a user whose login name contained one
1465 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1466 It caused a spool format error. I have fixed the spool reading code. A
1467 related problem was that the "from" clause in the Received: line became
1468 illegal because of the space(s). It is now covered by ${quote_local_part.
1470 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1473 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1475 83. Patch from Sieve maintainer to fix unterminated string problem in
1476 "vacation" handling.
1478 84. Some minor changes to the Linux configuration files to help with other
1479 OS variants using glibc.
1481 85. One more patch for Sieve to update vacation handling to latest spec.
1484 ----------------------------------------------------
1485 See the note above about the 4.44 and 4.50 releases.
1486 ----------------------------------------------------
1492 1. Change 4.43/35 introduced a bug that caused file counts to be
1493 incorrectly computed when quota_filecount was set in an appendfile
1496 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1497 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1500 3. Give more explanation in the error message when the command for a transport
1501 filter fails to execute.
1503 4. There are several places where Exim runs a non-Exim command in a
1504 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1505 was being done only for the command run by the queryprogram router. It is
1506 now done for all such subprocesses. The other cases are: ${run, transport
1507 filters, and the commands run by the lmtp and pipe transports.
1509 5. Some older OS have a limit of 256 on the maximum number of file
1510 descriptors. Exim was using setrlimit() to set 1000 as a large value
1511 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1512 systems. I've change it so that if it can't get 1000, it tries for 256.
1514 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1515 was an oversight, and furthermore, ever since the addition of extra
1516 controls (e.g. 4.43/32), the checks on when to allow different forms of
1517 "control" were broken. There should now be diagnostics for all cases when a
1518 control that does not make sense is encountered.
1520 7. $recipients is now available in the predata ACL (oversight).
1522 8. Tidy the search cache before the fork to do a delivery from a message
1523 received from the command line. Otherwise the child will trigger a lookup
1524 failure and thereby defer the delivery if it tries to use (for example) a
1525 cached ldap connection that the parent has called unbind on.
1527 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1528 of $address_data from the recipient verification was clobbered by the
1529 sender verification.
1531 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1532 was its contents. (It was OK if the option was not defined at all.)
1534 11. A "Completed" log line is now written for messages that are removed from
1535 the spool by the -Mrm option.
1537 12. $host_address is now set to the target address during the checking of
1538 ignore_target_hosts.
1540 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1541 being passed; this would have caused $sender_host_name to have been used if
1542 matching the list had actually called for a host name (not very likely,
1543 since this list is usually IP addresses). A host name is now passed as
1546 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1547 code that specifies a non-restarting handler (typically sigaction() in
1548 modern systems) in an attempt to fix a rare and obscure crash bug.
1550 15. Narrowed the window for a race in the daemon that could cause it to ignore
1551 SIGCHLD signals. This is not a major problem, because they are used only to
1552 wake it up if nothing else does.
1554 16. A malformed maildirsize file could cause Exim to calculate negative values
1555 for the mailbox size or file count. Odd effects could occur as a result.
1556 The maildirsize information is now recalculated if the size or filecount
1559 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1560 support for a long time. Removed HAVE_SYS_VFS_H.
1562 18. Updated exipick to current release from John Jetmore.
1564 19. Allow an empty sender to be matched against a lookup in an address list.
1565 Previously the only cases considered were a regular expression, or an
1568 20. Exim went into a mad DNS lookup loop when doing a callout where the
1569 host was specified on the transport, if the DNS lookup yielded more than
1572 21. The RFC2047 encoding function was originally intended for short strings
1573 such as real names; it was not keeping to the 75-character limit for
1574 encoded words that the RFC imposes. It now respects the limit, and
1575 generates multiple encoded words if necessary. To be on the safe side, I
1576 have increased the buffer size for the ${rfc2047: expansion operator from
1579 22. Failure to deliver a bounce message always caused it to be frozen, even if
1580 there was an errors_to setting on the router. The errors_to setting is now
1583 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1584 canonical form (fully expanded) before being placed in
1585 $sender_host_address.
1587 24. Updated eximstats to version 1.33
1589 25. Include certificate and key file names in error message when GnuTLS fails
1590 to set them up, because the GnuTLS error message doesn't include the name
1591 of the failing file when there is a problem reading it.
1593 26. Expand error message when OpenSSL has problems setting up cert/key files.
1596 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1597 (this can affect the format of dates).
1599 28. exim_tidydb, when checking for the continued existence of a message for
1600 which it has found a message-specific retry record, was not finding
1601 messages that were in split spool directories. Consequently, it was
1602 deleting retry records that should have stayed in existence.
1604 29. eximstats updated to version 1.35
1605 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1606 1.35 - bugfix such that pie charts by volume are generated correctly
1608 30. The SPA authentication driver was not abandoning authentication and moving
1609 on to the next authenticator when an expansion was forced to fail,
1610 contradicting the general specification for all authenticators. Instead it
1611 was generating a temporary error. It now behaves as specified.
1613 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1614 (the order specifies the preference for clients). The order is now AES256,
1615 AES128, 3DES, ARCFOUR128.
1617 31. Small patch to Sieve code - explicitly set From: when generating an
1620 32. Exim crashed if a remote delivery caused a very long error message to be
1621 recorded - for instance if somebody sent an entire SpamAssassin report back
1622 as a large number of 550 error lines. This bug was coincidentally fixed by
1623 increasing the size of one of Exim's internal buffers (big_buffer) that
1624 happened as part of the Exiscan merge. However, to be on the safe side, I
1625 have made the code more robust (and fixed the comments that describe what
1628 33. Some experimental protocols are using DNS PTR records for new purposes. The
1629 keys for these records are domain names, not reversed IP addresses. The
1630 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1631 leaves it alone. Component reversal etc. now happens only for IP addresses.
1634 34. The host_aton() function is supposed to be passed a string that is known
1635 to be a valid IP address. However, in the case of IPv6 addresses, it was
1636 not checking this. This is a hostage to fortune. Exim now panics and dies
1637 if the condition is not met. A case was found where this could be provoked
1638 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1639 components; fortuitously, this particular loophole had already been fixed
1640 by change 4.50/55 or 4.44/33 above.
1642 If there are any other similar loopholes, the new check in host_aton()
1643 itself should stop them being exploited. The report I received stated that
1644 data on the command line could provoke the exploit when Exim was running as
1645 exim, but did not say which command line option was involved. All I could
1646 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1647 running as the user.
1650 35. There was a buffer overflow vulnerability in the SPA authentication code
1651 (which came originally from the Samba project). I have added a test to the
1652 spa_base64_to_bits() function which I hope fixes it.
1655 36. The daemon start-up calls getloadavg() while still root for those OS that
1656 need the first call to be done as root, but it missed one case: when
1657 deliver_queue_load_max is set with deliver_drop_privilege. This is
1658 necessary for the benefit of the queue runner, because there is no re-exec
1659 when deliver_drop_privilege is set.
1661 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1662 was in use, was not putting the data itself into the right store pool;
1663 consequently, it could be overwritten for a subsequent message in the same
1664 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1667 38. Sometimes the final signoff response after QUIT could fail to get
1668 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1669 < 0 before doing a fflush(). This bug looks as though it goes back to the
1670 introduction of TLS in release 3.20, but "sometimes" must have been rare
1671 because the tests only now provoked it.
1677 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1678 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1679 Now it returns an integer. A number of calls were still expecting a T/F
1680 return. Fortuitously, in all cases, the tests worked in OK situations,
1681 which is the norm. However, things would have gone wrong on any write
1682 failures on the smtp file descriptor. This function is used when sending
1683 messages over SMTP and also when doing verify callouts.
1685 2. When Exim is called to do synchronous delivery of a locally submitted
1686 message (the -odf or -odi options), it no longer closes stderr before doing
1689 3. Implemented the mua_wrapper option.
1691 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1693 5. Implemented the functions header_remove(), header_testname(),
1694 header_add_at_position(), and receive_remove_recipient(), and exported them
1697 6. If an ACL "warn" statement specified the addition of headers, Exim already
1698 inserted X-ACL-Warn: at the start if there was no header name. However, it
1699 was not making this test for the second and subsequent header lines if
1700 there were newlines in the string. This meant that an invalid header could
1701 be inserted if Exim was badly configured.
1703 7. Allow an ACL "warn" statement to add header lines at the start or after all
1704 the Received: headers, as well as at the end.
1706 8. Added the rcpt_4xx retry error code.
1708 9. Added postmaster_mailfrom=xxx to callout verification option.
1710 10. Added mailfrom=xxxx to the callout verification option, for verify=
1713 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1714 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1716 12. Inserted some casts to stop certain compilers warning when using pointer
1717 differences as field lengths or precisions in printf-type calls (mostly
1718 affecting debugging statements).
1720 13. Added optional readline() support for -be (dynamically loaded).
1722 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1723 same clock tick as a message's arrival, so that its received time was the
1724 same as the "first fail" time on the retry record, and that message
1725 remained on the queue past the ultimate address timeout, every queue runner
1726 would try a delivery (because it was past the ultimate address timeout) but
1727 after another failure, the ultimate address timeout, which should have then
1728 bounced the address, did not kick in. This was a "< instead of <=" error;
1729 in most cases the first failure would have been in the next clock tick
1730 after the received time, and all would be well.
1732 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1733 being recognized when the domain list was tested by the match_domain
1734 condition in an expansion string.
1736 16. Added the ${str2b64: operator.
1738 17. Exim was always calling setrlimit() to set a large limit for the number of
1739 processes, without checking whether the existing limit was already
1740 adequate. (It did check for the limit on file descriptors.) Furthermore,
1741 errors from getrlimit() and setrlimit() were being ignored. Now they are
1742 logged to the main and panic logs, but Exim does carry on, to try to do its
1743 job under whatever limits there are.
1745 18. Imported PCRE 5.0.
1747 19. Trivial typo in log message " temporarily refused connection" (the leading
1750 20. If the log selector return_path_on_delivery was set and an address was
1751 redirected to /dev/null, the delivery process crashed because it assumed
1752 that a return path would always be set for a "successful" delivery. In this
1753 case, the whole delivery is bypassed as an optimization, and therefore no
1756 21. Internal re-arrangement: the function for sending a challenge and reading
1757 a response while authentication was assuming a zero-terminated challenge
1758 string. It's now changed to take a pointer and a length, to allow for
1759 binary data in such strings.
1761 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1763 23. Exim was not respecting finduser_retries when seeking the login of the
1764 uid under which it was called; it was always trying 10 times. (The default
1765 setting of finduser_retries is zero.) Also, it was sleeping after the final
1766 failure, which is pointless.
1768 24. Implemented tls_on_connect_ports.
1770 25. Implemented acl_smtp_predata.
1772 26. If the domain in control=submission is set empty, Exim assumes that the
1773 authenticated id is a complete email address when it generates From: or
1774 Sender: header lines.
1776 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1777 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1778 chown and chgrp in /bin and hostname in /usr/bin.
1780 28. Exim was keeping the "process log" file open after each use, just as it
1781 does for the main log. This opens the possibility of it remaining open for
1782 long periods when the USR1 signal hits a daemon. Occasional processlog
1783 errors were reported, that could have been caused by this. Anyway, it seems
1784 much more sensible not to leave this file open at all, so that is what now
1787 29. The long-running daemon process does not normally write to the log once it
1788 has entered its main loop, and it closes the log before doing so. This is
1789 so that log files can straightforwardly be renamed and moved. However,
1790 there are a couple of unusual error situations where the daemon does write
1791 log entries, and I had neglected to close the log afterwards.
1793 30. The text of an SMTP error response that was received during a remote
1794 delivery was being truncated at 512 bytes. This is too short for some of
1795 the long messages that one sometimes sees. I've increased the limit to
1798 31. It is now possible to make retry rules that apply only when a message has a
1799 specific sender, in particular, an empty sender.
1801 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1802 it possible to be selective about when SMTP synchronization is enforced.
1804 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1806 32. Implemented hosts_connection_nolog.
1808 33. Added an ACL for QUIT.
1810 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1813 35. Added mailbox_size and mailbox_filecount to appendfile.
1815 36. Added control = no_multiline_responses to ACLs.
1817 37. There was a bug in the logic of the code that waits for the clock to tick
1818 in the case where the clock went backwards by a substantial amount such
1819 that the microsecond fraction of "now" was more than the microsecond
1820 fraction of "then" (but the whole seconds number was less).
1822 38. Added support for the libradius Radius client library this is found on
1823 FreeBSD (previously only the radiusclient library was supported).
1829 1. When certain lookups returned multiple values in the form name=value, the
1830 quoting of the values was not always being done properly. Specifically:
1831 (a) If the value started with a double quote, but contained no whitespace,
1833 (b) If the value contained whitespace other than a space character (i.e.
1834 tabs or newlines or carriage returns) it was not quoted.
1835 This fix has been applied to the mysql and pgsql lookups by writing a
1836 separate quoting function and calling it from the lookup code. The fix
1837 should probably also be applied to nisplus, ibase and oracle lookups, but
1838 since I cannot test any of those, I have not disturbed their existing code.
1840 2. A hit in the callout cache for a specific address caused a log line with no
1841 reason for rejecting RCPT. Now it says "Previous (cached) callout
1842 verification failure".
1844 3. There was an off-by-one bug in the queryprogram router. An over-long
1845 return line was truncated at 256 instead of 255 characters, thereby
1846 overflowing its buffer with the terminating zero. As well as fixing this, I
1847 have increased the buffer size to 1024 (and made a note to document this).
1849 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1850 when Exim is waiting for an SMTP response from a remote server, Exim
1851 restarts its select() call on the socket, thereby resetting its timeout.
1852 This is not a problem when such interrupts are rare. Somebody set up a cron
1853 job to run exiwhat every 2 minutes, which is less than the normal select()
1854 timeout (5 or 10 minutes). This meant that the select() timeout never
1855 kicked in because it was always reset. I have fixed this by comparing the
1856 time when an interrupt arrives with the time at the start of the first call
1857 to select(). If more time than the timeout has elapsed, the interrupt is
1858 treated as a timeout.
1860 5. Some internal re-factoring in preparation for the addition of Sieve
1861 extensions (by MH). In particular, the "personal" test is moved to a
1862 separate function, and given an option for scanning Cc: and Bcc: (which is
1863 not set for Exim filters).
1865 6. When Exim created an email address using the login of the caller as the
1866 local part (e.g. when creating a From: or Sender: header line), it was not
1867 quoting the local part when it contained special characters such as @.
1869 7. Installed new OpenBSD configuration files.
1871 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1872 try to make them clearer.
1874 9. Callout options, other than the timeout value, were being ignored when
1875 verifying sender addresses in header lines. For example, when using
1877 verify = header_sender/callout=no_cache
1879 the cache was (incorrectly) being used.
1881 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1882 only the Cygwin environment.
1884 11. When return_path_on_delivery was set as a log selector, if different remote
1885 addresses in the same message used different return paths and parallel
1886 remote delivery occurred, the wrong values would sometimes be logged.
1887 (Whenever a remote delivery process finished, the return path value from
1888 the most recently started remote delivery process was logged.)
1890 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1891 lines when AUTH and/or TLS are in use. This is the "received protocol"
1892 field. Exim used to use "asmtp" for authenticated SMTP, without any
1893 indication (in the protocol name) for TLS use. Now it follows the RFC and
1894 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1895 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1896 names appear in log lines as well as in Received: header lines.
1898 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1899 extensions, and comparison tests, and to fix some bugs.
1901 14. Changes to the "personal" filter test:
1903 (1) The test was buggy in that it was just doing the equivalent of
1904 "contains" tests on header lines. For example, if a user's address was
1905 anne@some.where, the "personal" test would incorrectly be true for
1907 To: susanne@some.where
1909 This test is now done by extracting each address from the header in turn,
1910 and checking the entire address. Other tests that are part of "personal"
1911 are now done using regular expressions (for example, to check local parts
1912 of addresses in From: header lines).
1914 (2) The list of non-personal local parts in From: addresses has been
1915 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1916 taken from the Sieve specification recommendations.
1918 (3) If the message contains any header line starting with "List-" it is
1919 treated as non-personal.
1921 (4) The test for "circular" in the Subject: header line has been removed
1922 because it now seems ill-conceived.
1924 15. Minor typos in src/EDITME comments corrected.
1926 16. Installed latest exipick from John Jetmore.
1928 17. If headers_add on a router specified a text string that was too long for
1929 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1930 of string_sprintf() is now avoided.
1932 18. $message_body_size was not set (it was always zero) when running the DATA
1933 ACL and the local_scan() function.
1935 19. For the "mail" command in an Exim filter, no default was being set for
1936 the once_repeat time, causing a random time value to be used if "once" was
1937 specified. (If the value happened to be <= 0, no repeat happened.) The
1938 default is now 0s, meaning "never repeat". The "vacation" command was OK
1939 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1940 (I found it when inspecting the code).
1942 20. There is now an overall timeout for performing a callout verification. It
1943 defaults to 4 times the callout timeout, which applies to individual SMTP
1944 commands during the callout. The overall timeout applies when there is more
1945 than one host that can be tried. The timeout is checked before trying the
1946 next host. This prevents very long delays if there are a large number of
1947 hosts and all are timing out (e.g. when the network connections are timing
1948 out). The value of the overall timeout can be changed by specifying an
1949 additional sub-option for "callout", called "maxwait". For example:
1951 verify = sender/callout=5s,maxwait=20s
1953 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1954 to the end before writing, but this should make it even safer).
1956 22. Exim was forgetting that it had advertised PIPELINING for the second and
1957 subsequent messages on an SMTP connection. It was also not resetting its
1958 memory on STARTTLS and an internal HELO.
1960 23. When Exim logs an SMTP synchronization error within a session, it now
1961 records whether PIPELINING has been advertised or not.
1963 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1964 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1965 rather than long int.
1967 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1969 26. Added the never_mail option to autoreply.
1975 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1976 crash if the getsockname() call failed; this can happen if a connection is
1977 closed very soon after it is established. The problem was simply in the
1978 order in which certain operations were done, causing Exim to try to write
1979 to the SMTP stream before it had set up the file descriptor. The bug has
1980 been fixed by making things happen in the correct order.
1986 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1987 before the connection was closed, thus losing the rejection response.
1989 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1990 some early Solaris releases, but causes trouble in current releases where
1991 socklen_t is defined.
1993 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1996 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1999 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2000 that releasing the top of it at the end releases what was used for sub-
2001 expansions (unless the block got too big). However, discard this block if
2002 the first thing is a variable or header, so that we can use its block when
2003 it is dynamic (useful for very large $message_headers, for example).
2005 6. Lookups now cache *every* query, not just the most recent. A new, separate
2006 store pool is used for this. It can be recovered when all lookup caches are
2007 flushed. Lookups now release memory at the end of their result strings.
2008 This has involved some general refactoring of the lookup sources.
2010 7. Some code has been added to the store_xxx() functions to reduce the amount
2011 of flapping under certain conditions.
2013 8. log_incoming_interface used to affect only the <= reception log lines. Now
2014 it causes the local interface and port to be added to several more SMTP log
2015 lines, for example "SMTP connection from", and rejection lines.
2017 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2019 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2021 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2022 could be overwritten at the end of the current message (or the start of a
2023 new message if it was set in a HELO ACL). The value is now preserved for
2024 the duration of the SMTP connection.
2026 12. If a transport had a headers_rewrite setting, and a matching header line
2027 contained an unqualified address, that address was qualified, even if it
2028 did not match any rewriting rules. The underlying bug was that the values
2029 of the flags that permit the existence of unqualified sender and recipient
2030 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2031 non-local messages, and by -bnq for local messages) were not being
2032 preserved with the message after it was received.
2034 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2035 "next input=" as part of the text comprising the host identity instead of
2036 the correct text. The code was using the same buffer for two different
2037 strings. However, depending on which order the printing function evaluated
2038 its arguments, the bug did not always show up. Under Linux, for example, my
2039 test suite worked just fine.
2041 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2042 doesn't work with some older versions of Perl. It has been changed to "my",
2043 which in any case is probably the better facility to use.
2045 15. A really picky compiler found some instances of statements for creating
2046 error messages that either had too many or two few arguments for the format
2049 16. The size of the buffer for calls to the DNS resolver has been increased
2050 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2051 for addresses that have a lot of PTR records. This alleviates a problem; it
2052 does not fully solve it.
2054 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2055 buffer now truncates the list and logs the incident, which is the same
2056 action as happens when Exim is looking up a host name and its aliases.
2057 Previously in this situation something unpredictable would happen;
2058 sometimes it was "internal error: store_reset failed".
2060 18. If a server dropped the connection unexpectedly when an Exim client was
2061 using GnuTLS and trying to read a response, the client delivery process
2062 crashed while trying to generate an error log message.
2064 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2065 string, for example:
2067 warn message = H1: something\nH2: something
2069 the text was added as a single header line from Exim's point of view
2070 though it ended up OK in the delivered message. However, searching for the
2071 second and subsequent header lines using $h_h2: did not work. This has been
2072 fixed. Similarly, if a system filter added multiple headers in this way,
2073 the routers could not see them.
2075 20. Expanded the error message when iplsearch is called with an invalid key to
2076 suggest using net-iplsearch in a host list.
2078 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2079 ACLs are no longer actually imposed (and a message to that effect is
2082 22. If a "gecos" field in a passwd entry contained escaped characters, in
2083 particular, if it contained a \" sequence, Exim got it wrong when building
2084 a From: or a Sender: header from that name. A second bug also caused
2085 incorrect handling when an unquoted " was present following a character
2086 that needed quoting.
2088 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2089 was not being matched caselessly.
2091 24. Arranged for all hyphens in the exim.8 source to be escaped with
2094 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2095 properly. Recipient callout cache records were still being keyed to include
2096 the sender, even when use_sender was set false. This led to far more
2097 callouts that were necessary. The sender is no longer included in the key
2098 when use_sender is false.
2100 26. Added "control = submission" modifier to ACLs.
2102 27. Added the ${base62d: operator to decode base 62 numbers.
2104 28. dnsdb lookups can now access SRV records.
2106 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2107 the configuration file.
2109 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2110 (-v) mode. This makes the output for a verbose queue run more intelligible.
2112 31. Added a use_postmaster feature to recipient callouts.
2114 32. Added the $body_zerocount variable, containing the number of binary zero
2115 bytes in the message body.
2117 33. The time of last modification of the "new" subdirectory is now used as the
2118 "mailbox time last read" when there is a quota error for a maildir
2121 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2123 35. Added +ignore_unknown as a special item in host lists.
2125 36. Code for decoding IPv6 addresses in host lists is now included, even if
2126 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2127 address was recognized as an IP address, but was then not correctly decoded
2128 into binary, causing unexpected and incorrect effects when compared with
2135 1. Very minor rewording of debugging text in manualroute to say "list of
2136 hosts" instead of "hostlist".
2138 2. If verify=header_syntax was set, and a header line with an unqualified
2139 address (no domain) and a large number of spaces between the end of the
2140 name and the colon was received, the reception process suffered a buffer
2141 overflow, and (when I tested it) crashed. This was caused by some obsolete
2142 code that should have been removed. The fix is to remove it!
2144 3. When running in the test harness, delay a bit after writing a bounce
2145 message to get a bit more predictability in the log output.
2147 4. Added a call to search_tidyup() just before forking a reception process. In
2148 theory, someone could use a lookup in the expansion of smtp_accept_max_
2149 per_host which, without the tidyup, could leave open a database connection.
2151 5. Added the variables $recipient_data and $sender_data which get set from a
2152 lookup success in an ACL "recipients" or "senders" condition, or a router
2153 "senders" option, similar to $domain_data and $local_part_data.
2155 6. Moved the writing of debug_print from before to after the "senders" test
2158 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2159 problems for message scanning, either using a data ACL, or using
2160 local_scan() because the Received: header was not generated till after they
2161 were called (in order to set the time as the time of reception completion).
2162 I have revised the way this works. The header is now generated after the
2163 body is received, but before the ACL or local_scan() are called. After they
2164 are run, the timestamp in the header is updated.
2170 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2171 before starting a queue runner without re-exec. This happened only when
2172 deliver_drop_privilege was set or when the Exim user was set to root. The
2173 effect of the bug was that timeouts during subsequent deliveries caused
2174 crashes instead of being properly handled. The handler is now left at its
2175 default (and expected) setting.
2177 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2178 message, again when deliver_drop_privilege is set or Exim is run as root.
2179 The bug described in (1) was not present in this case, but the tidying up
2180 of the other signals was missing. I have made the two cases consistent.
2182 3. The ignore_target_hosts setting on a manualroute router was being ignored
2183 for hosts that were looked up using the /MX notation.
2185 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2188 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2189 operated on the sender address. After changing the $sender_address to <>
2190 for the sender address verify, Exim was re-instated it as the original
2191 (before rewriting) address, but remembering that it had rewritten it, so it
2192 wasn't rewriting it again. This bug also had the effect of breaking the
2193 sender address verification caching when the sender address was rewritten.
2195 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2196 This has been changed so that if the ip literal address matches
2197 ignore_target_hosts, the router declines.
2199 7. Added expansion conditions match_domain, match_address, and match_local_
2200 part (NOT match_host).
2202 8. The placeholder for the Received: header didn't have a length field set.
2204 9. Added code to Exim itself and to exim_lock to test for a specific race
2205 condition that could lead to file corruption when using MBX delivery. The
2206 issue is with the lockfile that is created in /tmp. If this file is removed
2207 after a process has opened it but before that process has acquired a lock,
2208 there is the potential for a second process to recreate the file and also
2209 acquire a lock. This could lead to two Exim processes writing to the file
2210 at the same time. The added code performs the same test as UW imapd; it
2211 checks after acquiring the lock that its file descriptor still refers to
2212 the same named file.
2214 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2215 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2218 11. Added the smtp_active_hostname option. If used, this will typically be made
2219 to depend on the incoming interface address. Because $interface_address is
2220 not set up until the daemon has forked a reception process, error responses
2221 that can happen earlier (such as "too many connections") no longer contain
2224 12. If an expansion in a condition on a "warn" statement fails because a lookup
2225 defers, the "warn" statement is abandoned, and the next ACL statement is
2226 processed. Previously this caused the whole ACL to be aborted.
2228 13. Added the iplsearch lookup type.
2230 14. Added ident_timeout as a log selector.
2232 15. Added tls_certificate_verified as a log selector.
2234 16. Added a global option tls_require_ciphers (compare the smtp transport
2235 option of the same name). This controls incoming TLS connections.
2237 17. I finally figured out how to make tls_require_ciphers do a similar thing
2238 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2239 before starting the TLS session.
2241 18. Tabs are now shown as \t in -bP output.
2243 19. If the log selector return_path_on_delivery was set, Exim crashed when
2244 bouncing a message because it had too many Received: header lines.
2246 20. If two routers both had headers_remove settings, and the first one included
2247 a superfluous trailing colon, the final name in the first list and the
2248 first name in the second list were incorrectly joined into one item (with a
2249 colon in the middle).
2255 1. Added -C and -D options to the exinext utility, mainly to make it easier
2256 to include in the automated testing, but these could be helpful when
2257 multiple configurations are in use.
2259 2. The exinext utility was not formatting the output nicely when there was
2260 an alternate port involved in the retry record key, nor when there was a
2261 message id as well (for retries that were specific to a specific message
2262 and a specific host). It was also confused by IPv6 addresses, because of
2263 the additional colons they contain. I have fixed the IPv4 problem, and
2264 patched it up to do a reasonable job for IPv6.
2266 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2267 delivery, the log line now contains "pipelined" if PIPELINING was used.
2269 4. An SMTP transport process used to panic and die if the bind() call to set
2270 an explicit outgoing interface failed. This has been changed; it is now
2271 treated in the same way as a connect() failure.
2273 5. A reference to $sender_host_name in the part of a conditional expansion
2274 that was being skipped was still causing a DNS lookup. This no longer
2277 6. The def: expansion condition was not recognizing references to header lines
2278 that used bh_ and bheader_.
2280 7. Added the _cache feature to named lists.
2282 8. The code for checking quota_filecount in the appendfile transport was
2283 allowing one more file than it should have been.
2285 9. For compatibility with Sendmail, the command line option
2293 and sets the incoming protocol and host name (for trusted callers). The
2294 host name and its colon can be omitted when only the protocol is to be set.
2295 Note the Exim already has two private options, -pd and -ps, that refer to
2296 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2297 "s", but I don't think that's a major issue.
2299 10. A number of refactoring changes to the code, none of which should affect
2302 (a) The number of logging options was getting close to filling up the
2303 32-bit word that was used as a bit map. I have split them into two classes:
2304 those that are passed in the argument to log_write(), and those that are
2305 only ever tested independently outside of that function. These are now in
2306 separate 32-bit words, so there is plenty of room for expansion again.
2307 There is no change in the user interface or the logging behaviour.
2309 (b) When building, for example, log lines, the code previously used a
2310 macro that called string_cat() twice, in order to add two strings. This is
2311 not really sufficiently general. Furthermore, there was one instance where
2312 it was actually wrong because one of the argument was used twice, and in
2313 one call a function was used. (As it happened, calling the function twice
2314 did not affect the overall behaviour.) The macro has been replaced by a
2315 function that can join an arbitrary number of extra strings onto a growing
2318 (c) The code for expansion conditions now uses a table and a binary chop
2319 instead of a serial search (which was left over from when there were very
2320 few conditions). Also, it now recognizes conditions like "pam" even when
2321 the relevant support is not compiled in: a suitably worded error message is
2322 given if an attempt is made to use such a condition.
2324 11. Added ${time_interval:xxxxx}.
2326 12. A bug was causing one of the ddress fields not to be passed back correctly
2327 from remote delivery subprocesses. The field in question was not being
2328 subsequently used, so this caused to problems in practice.
2330 13. Added new log selectors queue_time and deliver_time.
2332 14. Might have fixed a bug in maildirsizefile handling that threw up
2333 "unexpected character" debug warnings, and recalculated the data
2334 unnecessarily. In any case, I expanded the warning message to give more
2337 15. Added the message "Restricted characters in address" to the statements in
2338 the default ACL that block characters like @ and % in local parts.
2340 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2341 Three changes have been made:
2343 (a) There was a serious bug; a negative response to MAIL caused the whole
2344 recipient domain to be cached as invalid, thereby blocking all messages
2345 to all local parts at the same domain, from all senders. This bug has
2346 been fixed. The domain is no longer cached after a negative response to
2347 MAIL if the sender used is not empty.
2349 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2352 (c) A new callout option, "use_sender" has been added for people who want
2353 the modified behaviour.
2359 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2362 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2363 indeed breaks things for older releases.
2365 3. Added additional logging to the case where there is a problem reading data
2366 from a filter that is running in a subprocess using a pipe, in order to
2367 try to track down a specific problem.
2369 4. Testing facility fudge: when running in the test harness and attempting
2370 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2371 getting "No route to host". Convert this to a timeout.
2373 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2376 6. Some OS don't have socklen_t but use size_t instead. This affects the
2377 fifth argument of getsockopt() amongst other things. This is now
2378 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2379 can be set for individual OS. I have set it for SunOS5, OSF1, and
2380 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2381 some earlier ones do not.
2383 7. Change 4.30/15 was not doing the test caselessly.
2385 8. The standard form for an IPv6 address literal was being rejected by address
2386 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2387 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2388 this, as well as the form without the "IPv6" on the front (but only when
2389 address literals are enabled, of course).
2391 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2393 10. Exim crashed if a message with an empty sender address specified by -f
2394 encountered a router with an errors_to setting. This could be provoked only
2395 by a command such as
2399 where an empty string was supplied; "<>" did not hit this bug.
2401 11. Installed PCRE release 4.5.
2403 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2404 remained set. It is now erased.
2406 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2407 times from message ids (which are base 36 rather than the normal 62).
2409 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2410 were being counted as actual protocol errors, and logged if the log
2411 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2412 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2413 and DATA following a set of rejected RCPTs do not count as protocol errors.
2414 In other words, Exim assumes they were pipelined, though this may not
2415 actually be the case. Of course, in all cases the client gets an
2416 appropriate error code.
2418 15. If a lookup fails in an ACL condition, a message about the failure may
2419 be available; it is used if testing the ACL cannot continue, because most
2420 such messages specify what the cause of the deferral is. However, some
2421 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2422 that caused an old message to be retained and used if a later statement
2423 caused a defer, replacing the real cause of the deferral.
2425 16. If an IP address had so many PTR records that the DNS lookup buffer
2426 was not large enough to hold them, Exim could crash while trying to process
2427 the truncated data. It now detects and logs this case.
2429 17. Further to 4.21/58, another change has been made: if (and only if) the
2430 first line of a message (the first header line) ends with CRLF, a bare LF
2431 in a subsequent header line has a space inserted after it, so as not to
2432 terminate the header.
2434 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2435 unnecessarily, used atoi() instead of strtol(), and didn't check the
2436 termination when getting file sizes from file names by regex.
2438 19. Completely re-implemented the support for maildirsize files, in the light
2439 of a number of problems with the previous contributed implementation
2440 (4.30/29). In particular:
2442 . If the quota is zero, the maildirsize file is maintained, but no quota is
2445 . If the maildir directory does not exist, it is created before any attempt
2446 to write a maildirsize file.
2448 . The quota value in the file is just a cache; if the quota is changed in
2449 the transport, the new value overrides.
2451 . A regular expression is available for excluding directories from the
2454 20. The autoreply transport checks the characters in options that define the
2455 message's headers; it allows continued headers, but it was checking with
2456 isspace() after an embedded newline instead of explicitly looking for a
2459 21. If all the "regular" hosts to which an address was routed had passed their
2460 expiry times, and had not reached their retry times, the address was
2461 bounced, even if fallback hosts were defined. Now Exim should go on to try
2464 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2465 equivalent code in the SMTP transport. Some hosts send humungous responses
2466 to HELO/EHLO, more than 1024 it seems.
2468 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2469 gives compiler warnings in some environments. I've now done it "properly",
2472 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2473 (because of problems with the built-in one) was declared to return uschar *
2474 instead of char *, causing compiler failure.
2476 25. Fixed a file descriptor leak when processing alias/forward files.
2478 26. Fixed a minor format string issue in dbfn.c.
2480 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2482 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2483 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2486 29. When a sender address is verified, it is cached, to save repeating the test
2487 when there is more than one recipient in a message. However, when the
2488 verification involves a callout, it is possible for different callout
2489 options to be set for different recipients. It is too complicated to keep
2490 track of this in the cache, so now Exim always runs a verification when a
2491 callout is required, relying on the callout cache for the optimization.
2492 The overhead is duplication of the address routing, but this should not be
2495 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2496 to be verified with callout=postmaster, and the main callout worked but the
2497 postmaster check failed, the verification correctly failed. However, if a
2498 subsequent RCPT command asked for sender verification *without* the
2499 postmaster check, incorrect caching caused this verification also to fail,
2502 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2503 it was not caching the DNS options (qualify_single, search_parents) that
2504 were used when the lookup failed. A subsequent lookup with different
2505 options therefore always gave the same answer, though there were cases
2506 where it should not have. (Example: a "domains = !$mx_any" option on a
2507 dnslookup router: the "domains" option is always processed without any
2508 widening, but the router might have qualify_single set.) Now Exim uses the
2509 cached value only when the same options are set.
2511 32. Added John Jetmore's "exipick" utility to the distribution.
2513 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2514 than a timeout (e.g. a certificate is required, and is not provided), an
2515 Exim server now closes the connection immediately. Previously it waited for
2516 the client to close - but if the client is SSL, it seems that they each
2517 wait for each other, leading to a delay before one of them times out.
2519 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2520 maintained 0.8.x compatibility because I don't think many are using it, and
2521 it is clearly obsolete.
2523 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2526 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2527 client certificate was expired. A simple patch fixes this, though I don't
2528 understand the full logic of why the verify callback is called multiple
2531 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2532 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2533 which causes problems with some clients (such as the Certicom SSL Plus
2534 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2535 disables the coutermeasure allowing Eudora to connect."
2537 38. Exim was not checking that a write() to a log file succeeded. This could
2538 lead to Bad Things if a log got too big, in particular if it hit a file
2539 size limit. Exim now panics and dies if it cannot write to a log file, just
2540 as it does if it cannot open a log file.
2542 39. Modified OS/Makefile-Linux so that it now contains
2544 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2546 The two -D definitions ensure that Exim is compiled with large file
2547 support, which makes it possible to handle log files that are bigger than
2550 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2551 instance) a domain was checked against a named list that involved a lookup,
2552 causing $domain_data to be set, then another domain was checked against the
2553 same list, then the first domain was re-checked, the value of $domain_data
2554 after the final check could be wrong. In particular, if the second check
2555 failed, it could be set empty. This bug probably also applied to
2558 41. The strip_trailing_dot option was not being applied to the address given
2559 with the -f command-line option.
2561 42. The code for reading a message's header from the spool was incrementing
2562 $received_count, but never initializing it. This meant that the value was
2563 incorrect (doubled) while delivering a message in the same process in which
2564 it was received. In the most common configuration of Exim, this never
2565 happens - a fresh exec is done - but it can happen when
2566 deliver_drop_privilege is set.
2568 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2569 it now includes up to 150 characters of the unexpected data in the log
2572 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2573 and building data strings. The size of both of these buffers was 10 000
2574 bytes - far larger than anybody would *ever* want, thought I. Needless to
2575 say, somebody hit the limit. I have increased the maximum line length to
2576 20 000 and the maximum data length of concatenated lines to 100 000. I have
2577 also fixed two bugs, because there was no checking on these buffers. Tsk,
2578 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2579 buffer is too small.
2581 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2582 lsearch lookups. Now it does.
2584 46. When parsing a route_list item in a manualroute router, a fixed-length
2585 buffer was used for the list of hosts. I made this 1024 bytes long,
2586 thinking that nobody would ever have a list of hosts that long. Wrong.
2587 Somebody had a whole pile of complicated expansion conditions, and the
2588 string was silently truncated, leading to an expansion error. It turns out
2589 that it is easier to change to an unlimited length (owing to other changes
2590 that have happened since this code was originally written) than to build
2591 structure for giving a limitation error. The length of the item that
2592 expands into the list of hosts is now unlimited.
2594 47. The lsearch lookup could not handle data where the length of text line was
2595 more than 4095 characters. Such lines were truncated, leading to shortened
2596 data being returned. It should now handle lines of any length.
2598 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2599 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2602 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2603 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2604 no longer happens in comment lines. A list of replacements is now placed
2605 at the head of all of the source files, except those whose only change is
2606 to replace PERL_COMMAND in the very first #! line.
2608 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2609 messages on the queue, with a bottom-up merge sort, using code contributed
2610 by Michael Haardt. This should make operations like -bp somewhat faster on
2611 large queues. It won't affect queue runners, except when queue_run_in_order
2614 51. Installed eximstats 1.31 in the distribution.
2616 52. Added support for SRV lookups to the dnslookup router.
2618 53. If an ACL referred to $message_body or $message_body_end, the value was not
2619 reset for any messages that followed in the same SMTP session.
2621 54. The store-handling optimization for building very long strings was not
2622 differentiating between the different store pools. I don't think this
2623 actually made any difference in practice, but I've tidied it.
2625 55. While running the routers to verify a sender address, $sender_address
2626 was still set to the sender address. This is wrong, because when routing to
2627 send a bounce to the sender, it would be empty. Therefore, I have changed
2628 it so that, while verifying a sender address, $sender_address is set to <>.
2629 (There is no change to what happens when verifying a recipient address.)
2631 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2632 target A or AAAA records (if not already returned) without resetting the
2633 qualify_single or search_parents options of the DNS resolver. These are
2634 inappropriate in this case because the targets of MX and SRV records must
2635 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2636 target that, when qualified, matched something in the local domain. These
2637 two options are now turned off when doing these lookups.
2639 57. It seems that at least some releases of Reiserfs (which does not have the
2640 concept of a fixed number of inodes) returns zero and not -1 for the
2641 number of available inodes. This interacted badly with check_spool_inodes,
2642 which assumed that -1 was the "no such thing" setting. What I have done is
2643 to check that the total number of inodes is greater than zero before doing
2644 the test of how many are available.
2646 58. When a "warn" ACL statement has a log_message modifier, the message is
2647 remembered, and not repeated. This is to avoid a lot of repetition when a
2648 message has many recipients that cause the same warning to be written.
2649 Howewer, Exim was preserving the list of already written lines for an
2650 entire SMTP session, which doesn't seem right. The memory is now reset if a
2651 new message is started.
2653 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2654 debugging output unless log_rewrite was also set.
2656 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2657 of (int)(handle) when we know that handle contains (void *)(-1).
2659 61. The Exim daemon panic-logs an error return when it closes the incoming
2660 connection. However "connection reset by peer" seems to be common, and
2661 isn't really an error worthy of noting specially, so that particular error
2664 62. When Exim is trying to find all the local interfaces, it used to panic and
2665 die if the ioctl to get the interface flags failed. However, it seems that
2666 on at least one OS (Solaris 9) it is possible to have an interface that is
2667 included in the list of interfaces, but for which you get a failure error
2668 for this call. This happens when the interface is not "plumbed" into a
2669 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2670 failure of the "get flags" call assumes that the interface is down.
2672 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2673 makes life easier for people who are doing arithmetic on fields extracted
2674 from dates, where you often get leading zeros that should not be
2675 interpreted as octal.
2677 64. Added qualify_domain to the redirect router, to override the global
2680 65. If a pathologically long header line contained very many addresses (the
2681 report of this problem mentioned 10 000) and each of them was rewritten,
2682 Exim could use up a very large amount of memory. (It kept on making new
2683 copies of the header line as it rewrote, and never released the old ones.)
2684 At the expense of a bit more processing, the header rewriting function has
2685 been changed so that it no longer eats memory in this way.
2687 66. The generation of the Received: header has been moved from the time that a
2688 message starts to be received, to the time that it finishes. The timestamp
2689 in the Received: header should now be very close to that of the <= log
2690 line. There are two side-effects of this change:
2692 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2693 logged header lines no longer include the local Received: line, because
2694 it has not yet been created. The same applies to a copy of the message
2695 that is returned to a non-SMTP sender when a message is rejected.
2697 (b) When a filter file is tested using -bf, no additional Received: header
2698 is added to the test message. After some thought, I decided that this
2701 This change does not affect the value of $received_for. It is still set
2702 after address rewriting, but before local_scan() is called.
2704 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2706 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2707 gave an unhelpful panic error message, and a defer error. I have managed to
2708 change this behaviour so that it now rejects any supplied certificate,
2709 which seems right, as the list of acceptable certificates is empty.
2711 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2712 gave an unhelpful defer error. I have not managed to make this reject any
2713 supplied certificates, but the error message it gives is "no certificate
2714 supplied", which is not helpful.
2716 70. exigrep's output now also includes lines that are not associated with any
2717 message, but which match the given pattern. Implemented by a patch from
2718 Martin Sluka, which also tidied up the Perl a bit.
2720 71. Recipient callout verification, like sender verification, was using <> in
2721 the MAIL FROM command. This isn't really the right thing, since the actual
2722 sender may affect whether the remote host accepts the recipient or not. I
2723 have changed it to use the actual sender in the callout; this means that
2724 the cache record is now keyed on a recipient/sender pair, not just the
2725 recipient address. There doesn't seem to be a real danger of callout loops,
2726 since a callout by the remote host to check the sender would use <>.
2727 [SEE ABOVE: changed after hitting problems.]
2729 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2730 temporary errors. However, in the case of such a code being given after
2731 the end of a data transmission (i.e. after ".") Exim was failing to write
2732 a retry record for the message. (Yes, there was some broken host that was
2733 actually sending 8xx at this point.)
2735 73. An unknown lookup type in a host list could cause Exim to panic-die when
2736 the list was checked. (An example that provoked this was putting <; in the
2737 middle of a list instead of at the start.) If this happened during a DATA
2738 ACL check, a -D file could be left lying around. This kind of configuration
2739 error no longer causes Exim to die; instead it causes a defer errror. The
2740 incident is still logged to the main and panic logs.
2742 74. Buglet left over from Exim 3 conversion. The message "too many messages
2743 in one connection" was written to the rejectlog but not the mainlog, except
2744 when address rewriting (yes!) was being logged.
2746 75. Added write_rejectlog option.
2748 76. When a system filter was run not as root (that is, when system_filter_user
2749 was set), the values of the $n variables were not being returned to the
2750 main process; thus, they were not subsequently available in the $sn
2753 77. Added +return_path_on_delivery log selector.
2755 78. A connection timeout was being treated differently from recipients deferred
2756 when testing hosts_max_try with a message that was older than the host's
2757 retry timeout. (The host should not be counted, thus allowing all hosts to
2758 be tried at least once before bouncing.) This may have been the cause of an
2759 occasionally reported bug whereby a message would remain on the queue
2760 longer than the retry timeout, but would be bounced if a delivery was
2761 forced. I say "may" because I never totally pinned down the problem;
2762 setting up timeout/retry tests is difficult. See also the next item.
2764 79. The ultimate address timeout was not being applied to errors that involved
2765 a combination of host plus message (for example, a timeout on a MAIL
2766 command). When an address resolved to a number of possible hosts, and they
2767 were not all tried for each delivery (e.g. because of hosts_max_try), a
2768 message could remain on the queue longer than the retry timeout.
2770 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2773 81. Fixed an obscure SMTP outgoing bug which required at least the following
2774 conditions: (a) there was another message waiting for the same server;
2775 (b) the server returned 5xx to all RCPT commands in the first message so
2776 that the message was not completed; (c) the server dropped the connection
2777 or gave a negative response to the RSET that Exim sends to abort the
2778 transaction. The observed case was a dropped connection after DATA that had
2779 been sent in pipelining mode. That is, the server had advertised PIPELINING
2780 but was not implementing it correctly. The effect of the bug was incorrect
2781 behaviour, such as trying another host, and this could lead to a crash.
2787 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2788 and daemon.c were passed as pointers to ints; they should have been
2789 pointers to socklen_t variables (which are typically unsigned ints).
2791 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2794 3. Fixed a really odd bug that affected only the testing scheme; patching a
2795 certain fixed string in the binary changed the value of another string that
2796 happened to be identical to the end of the original first string.
2798 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2799 name", it returns that address as the IP address. On some operating
2800 systems (e.g. Solaris), it also passes back the IP address string as the
2801 "host name". However, on others (e.g. Linux), it passes back an empty
2802 string. Exim wasn't checking for this, and was changing the host name to an
2803 empty string, assuming it had been canonicized.
2805 5. Although rare, it is permitted to have more than one PTR record for a given
2806 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2807 all the names associated with an address, because they do in Solaris.
2808 However, it seems that they do not in Linux for data that comes from the
2809 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2810 I found this out when I moved to a new Linux workstation and tried to run
2811 the Exim test suite.
2813 To get round this problem I have changed the code so that it now does its
2814 own call to the DNS to look up PTR records when searching for a host name.
2815 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2816 addresses that are only in /etc/hosts are still found.
2818 This behaviour is, however, controlled by an option called host_lookup_
2819 order, which defaults to "bydns:byaddr". If people want to use the other
2820 order, or indeed, just use one or the other means of lookup, they can
2821 specify it in this variable.
2823 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2824 some operating systems, this comes back from gethostbyaddr() as an empty
2825 string, and this is what Exim used to test for. However, it seems that in
2826 other systems, "." is yielded. Exim now tests for this case too.
2828 7. The values of check_spool_space and check_log_space are now held internally
2829 as a number of kilobytes instead of an absolute number of bytes. If a
2830 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2831 kilobyte. This means that much larger values can be stored.
2833 8. Exim monitor: an attempt to get the action menu when not actually pointing
2834 at a message produces an empty menu entitled "No message selected". This
2835 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2836 no entries in it ("Shell widget menu has zero width and/or height"). So I
2837 have added a single, blank menu entry in this case.
2839 9. Added ${quote_local_part.
2841 10. MIME decoding is now applied to the contents of Subject: header lines when
2844 11. Now that a reference to $sender_host_address automatically causes a reverse
2845 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2846 host lookup before query-style lookups in lists that might use this
2847 variable. This has therefore been abolished, and the "net-" prefix is no
2848 longer necessary for query-style lookups.
2850 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2851 have been a typo for LFLAGS, so it has been changed.
2853 13. The install script calls Exim with "-C /dev/null" in order to find the
2854 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2855 to be output. Howeve, since Exim outputs its version number before the
2856 error, it didn't break the script. It just looked ugly. I fixed this by
2857 always allowing "-C /dev/null" if the caller is root.
2859 14. Ignore overlarge ACL variable number when reading spool file - insurance
2860 against a later release with more variables having written the file.
2862 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2863 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2864 this, as well as the form without the "IPv6" on the front.
2866 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2867 OS/Makefile-Darwin file.
2869 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2870 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2872 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2873 variable. However, some OS use ints for inodes. I've added cast to long int
2874 to get rid of the compiler warning.
2876 19. I had forgotten to lock out "/../" in configuration file names when
2877 ALT_CONFIG_PREFIX was set.
2879 20. Routers used for verification do not need to specify transports. However,
2880 if such a router generated a host list, and callout was configured, Exim
2881 crashed, because it could not find a port number from the (non-existent)
2882 transport. It now assumes port 25 in this circumstance.
2884 21. Added the -t option to exigrep.
2886 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2887 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2888 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2889 them is set, LOOKUP_LSEARCH is forced.
2891 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2893 24. Added sender and host information to the "rejected by local_scan()" log
2894 line; previously there was no indication of these.
2896 25. Added .include_if_exists.
2898 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2899 Linux. It turns out that not all file systems support this. Apparently some
2900 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2901 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2902 not supported on the file descriptor, is now ignored when Exim is trying to
2903 sync a directory. This applies only to Linux.
2905 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2907 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2908 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2911 29. Added support for maildirsize files from supplied patch (modified a bit).
2913 30. The use of :fail: followed by an empty string could lead Exim to respond to
2914 sender verification failures with (e.g.):
2916 550 Verification failed for <xxx>
2917 550 Sender verify failed
2919 where the first response line was missing the '-' that indicates it is not
2920 the final line of the response.
2922 31. The loop for finding the name of the user that called Exim had a hardwired
2923 limit of 10; it now uses the value of finduser_retries, which is used for
2924 all other user lookups.
2926 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2929 33. Exim was neglecting to zero errno before one call of strtol() when
2930 expanding a string and expecting an integer value. On some systems this
2931 resulted in spurious "integer overflow" errors. Also, it was casting the
2932 result into an int without checking.
2934 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2935 did not work. The code looks as if it has *never* worked, though it appears
2936 to have been documented since at least releast 1.62. I have made it work.
2938 35. The "timeout_DNS" error in retry rules, also documented since at least
2939 1.62, also never worked. As it isn't clear exactly what this means, and
2940 clearly it isn't a major issue, I have abolished the feature by treating it
2941 as "timeout", and writing a warning to the main and panic logs.
2943 36. The display of retry rules for -brt wasn't always showing the error code
2946 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2947 timeout_connect_A, timeout_connect_MX.
2949 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2950 to the empty sender.
2952 39. The daemon was not analysing the content of -oX till after it had closed
2953 stderr and disconnected from the controlling terminal. This meant that any
2954 syntax errors were only noted on the panic log, and the return code from
2955 the command was 0. By re-arranging the code a little, I've made the
2956 decoding happen first, so such errors now appear on stderr, and the return
2957 code is 1. However, the actual setting up of the sockets still happens in
2958 the disconnected process, so errors there are still only recorded on the
2961 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2962 connections (as happens on some IP stacks) was logged at start up time as
2963 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2964 it from "IPv6 and IPv4", which means that two separate sockets are being
2967 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2968 says whether AF_INET or AF_INET6 was passed as an argument.
2970 42. Exiwhat output was messed up when time zones were included in log
2973 43. Exiwhat now gives more information about the daemon's listening ports,
2974 and whether -tls-on-connect was used.
2976 44. The "port" option of the smtp transport is now expanded.
2978 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2979 silently ignored. Now an error message is written to the main and panic
2982 46. There's a new ACL modifier called "logwrite" which writes to a log file
2983 as soon as it is encountered.
2985 47. Added $local_user_uid and $local_user_gid at routing time.
2987 48. Exim crashed when trying to verify a sender address that was being
2990 49. Exim was recognizing only a space character after ".include". It now also
2991 recognizes a tab character.
2993 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2994 extracting the relevant information from the specification. The man page no
2995 longer contains scrambled data for the -d option, and I've added a section
2996 at the front about calling Exim under different names.
2998 51. Added "extra_headers" argument to the "mail" command in filter files.
3000 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3003 53. Installed eximstats 1.29.
3005 54. Added transport_filter_timeout as a generic transport option.
3007 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3008 Cc: header lines. This was required by RFC 822, but it not required by RFC
3011 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3012 incoming messages that did not have them. Now it does so only if the
3013 message originates locally, that is, if there is no associated remote host
3014 address. When Resent- header lines are present, this applies to the Resent-
3015 lines rather than the non-Resent- lines.
3017 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3018 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3020 58. Messages for configuration errors now include the name of the main
3021 configuration file - useful now that there may be more than one file in a
3022 list (.included file names were always shown).
3024 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3025 for those rare installations that do not start the daemon as root or run it
3026 setuid root. I've cut out the call to initgroups() if the daemon is not
3029 60. The Exim user and group can now be bound into the binary as text strings
3030 that are looked up at the start of Exim's processing.
3032 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3034 62. Added $mailstore_basename variable.
3036 63. Installed patch to sieve.c from Michael Haardt.
3038 64. When Exim failed to open the panic log after failing to open the main log,
3039 the original message it was trying to log was written to stderr and debug
3040 output, but if they were not available (the usual case in production), it
3041 was lost. Now it is written to syslog before the two lines that record the
3042 failures to open the logs.
3044 65. Users' Exim filters run in subprocesses under the user's uid. It is
3045 possible for a "deliver" command or an alias in a "personal" command to
3046 provoke an address rewrite. If logging of address rewriting is configured,
3047 this fails because the process is not running as root or exim. There may be
3048 a better way of dealing with this, but for the moment (because 4.30 needs
3049 to be released), I have disabled address rewrite logging when running a
3050 filter in a non-root, non-exim process.
3056 1. The buildconfig auxiliary program wasn't quoting the value set for
3057 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3058 not defined. This bug was masked in 4.22 by the effect that was fixed in
3061 2. Some messages that were rejected after a message id was allocated were
3062 shown as "incomplete" by exigrep. It no longer does this for messages that
3063 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3065 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3066 have allow_domain_literals set, the ID did not get logged in the <= line.
3067 Domain literals are now always recognized in Message-ID: header lines.
3069 4. The first argument for a ${extract expansion item is the key name or field
3070 number. Leading and trailing spaces in this item were not being ignored,
3071 causing some misleading effects.
3073 5. When deliver_drop_privilege was set, single queue runner processes started
3074 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3075 same command in the process it spins off) were not dropping privilege.
3077 6. When the daemon running as "exim" started a queue runner, it always
3078 re-executed Exim in the spun-off process. This is a waste of effort when
3079 deliver_drop_privilege is set. The new process now just calls the
3080 queue-runner function directly.
3086 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3089 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3090 ignored. Though the use of -oP was forcing the writing of a pid file, it
3091 was always written to the default place.
3093 3. If the message "no IP address found for host xxxx" is generated during
3094 incoming verification, it is now followed by identification of the incoming
3095 connection (so you can more easily find what provoked it).
3097 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3099 5. Added some features to "harden" Exim a bit more against certain attacks:
3101 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3102 be put in Local/Makefile. This is like the never_users runtime option,
3103 but it cannot be overridden. The default setting is "root".
3105 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3106 prefix string with which any file named in a -C command line option
3109 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3110 is retained for -C and -D only if the caller of Exim is root. Without
3111 it, the exim user may also use -C and -D and retain privilege.
3113 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3114 command line option is disabled.
3116 6. Macro names set by the -D option must start with an upper case letter, just
3117 like macro names defined in the configuration file.
3119 7. Added "dereference=" facility to LDAP.
3121 8. Two instances of the typo "uknown" in the source files are fixed.
3123 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3124 the Configure-Makefile script screwed up while processing it.
3126 10. Incorporated PCRE 4.4.
3128 11. The SMTP synchronization check was not operating right at the start of an
3129 SMTP session. For example, it could not catch a HELO sent before the client
3130 waited for the greeting. There is now a check for outstanding input at the
3131 point when the greeting is written. Because of the duplex, asynchronous
3132 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3133 way, but not yet received, when the check is performed.
3135 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3136 on TCP/IP sockets, because this apparently causes some broken clients to
3139 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3140 unchanged) from the Cygwin maintainer.
3142 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3143 was supported instead of testing for mbx. Effectively a typo.
3145 15. The spa authenticator server code was not checking that the input it
3146 received was valid base64.
3148 16. The debug output line for the "set" modifier in ACLs was not showing the
3149 name of the variable that was being set.
3151 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3153 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3154 lookup on its own. Something else was needed to trigger the lookup. For
3155 example, a match in host_lookup or the need for a host name in a host list.
3156 Now, if $sender_host_name is referenced and the host name has not yet been
3157 looked up, a lookup is performed. If the lookup fails, the variable remains
3158 empty, and $host_lookup_failed is set to "1".
3160 19. Added "eqi" as a case-independent comparison operator.
3162 20. The saslauthd authentication condition could segfault if neither service
3163 nor realm was specified.
3165 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3166 error message that was logged was misleading, and incoming SMTP
3167 connections were dropped. The message is now more accurate, and temporary
3168 errors are given to SMTP connections.
3170 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3171 (except RSET) with a 421 error, until QUIT is received. However, it was
3172 failing to send a response to QUIT.
3174 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3175 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3176 ACL is now run after the helo_try_verify_hosts code.
3178 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3179 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3180 case-independent, so other case variants are also recognized). Apparently
3181 some systems use these upper case variants.
3183 25. If more than two messages were waiting for the same host, and a transport
3184 filter was specified for the transport, Exim sent two messages over the
3185 same TCP/IP connection, and then failed with "socket operation on non-
3186 socket" when it tried to send the third.
3188 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3190 27. The extern definition of crypt16() in expand.c was not being excluded when
3191 the OS had its own crypt16() function.
3193 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3194 as a preferred synonym for return_size_limit, both as an option and as an
3197 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3199 30. Changed the default configuration ACL to relax the local part checking rule
3200 for addresses that are not in any local domains. For these addresses,
3201 slashes and pipe symbols are allowed within local parts, but the sequence
3202 /../ is explicitly forbidden.
3204 31. SPA server authentication was not clearing the challenge buffer before
3207 32. log_message in a "warn" ACL statement was writing to the reject log as
3208 well as to the main log, which contradicts the documentation and doesn't
3209 seem right (because no rejection is happening). So I have stopped it.
3211 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3212 However, I am unable to do any testing of this.
3214 34. Fixed an infelicity in the appendfile transport. When checking directories
3215 for a mailbox, to see if any needed to be created, it was accidentally
3216 using path names with one or more superfluous leading slashes; tracing
3217 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3219 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3220 contents are added to the log line that is written for every discarded
3221 recipient. (Previously a log_message setting was ignored.)
3223 36. The ${quote: operator now quotes the string if it is empty.
3225 37. The install script runs exim in order to find its version number. If for
3226 some reason other than non-existence or emptiness, which it checks, it
3227 could not run './exim', it was installing it with an empty version number,
3228 i.e. as "exim-". This error state is now caught, and the installation is
3231 38. An argument was missing from the function that creates an error message
3232 when Exim fails to connect to the socket for saslauthd authentication.
3233 This could cause Exim to crash, or give a corrupted message.
3235 39. Added isip, isip4, and isip6 to ${if conditions.
3237 40. The ACL variables $acl_xx are now saved with the message, and can be
3238 accessed later in routers, transports, and filters.
3240 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3241 strings in the file are not string-expanded.
3243 42. If a MAIL command specified a SIZE value that was too large to fit into an
3244 int variable, the check against message_size_limit failed. Such values are
3245 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3246 day this will have to be increased, but I don't think I want to be around
3247 when emails are that large.
3254 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3255 iconv() is not standard in FreeBSD.
3257 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3258 IPv6 enabled. The observed symptom was a segmentation fault on return from
3259 the function os_common_find_running_interfaces() in src/os.c.
3261 3. In the check_special_case() function in daemon.c I had used "errno" as an
3262 argument name, which causes warnings on some systems. This was basically a
3263 typo, since it was named "eno" in the comments!
3265 4. The code that waits for the clock to tick (at a resolution of some fraction
3266 of a second) so as to ensure message-id uniqueness was always waiting for
3267 at least one whole tick, when it could have waited for less. [This is
3268 almost certainly not relevant at current processor speeds, where it is
3269 unlikely to ever wait at all. But we try to future-proof.]
3271 5. The function that sleeps for a time interval that includes fractions of a
3272 second contained a race. It did not block SIGALRM between setting the
3273 timer, and suspending (a couple of lines later). If the interval was short
3274 and the sigsuspend() was delayed until after it had expired, the suspension
3275 never ended. On busy systems this could lead to processes getting stuck for
3278 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3279 process, before it forks any delivery processes. The open lookup caching
3280 mechanism meant that the open file or database connection was passed into
3281 the delivery process. The problem was that delivery processes always tidy
3282 up cached lookup data. This could cause a problem for the next delivery
3283 process started by the queue runner, because the external queue runner
3284 process does not know about the closure. So the next delivery process
3285 still has data in the lookup cache. In the case of a file lookup, there was
3286 no problem because closing a file descriptor in a subprocess doesn't affect
3287 the parent. However, if the lookup was caching a connection to a database,
3288 the connection was closed, and the second delivery process was likely to
3289 see errors such as "PGSQL: query failed: server closed the connection
3290 unexpectedly". The problem has been fixed by closing all cached lookups
3291 in a queue runner before running a delivery process.
3293 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3294 seem to have the "const" qualifier which it has on other OS. I've
3297 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3298 *of the same type* (client or server) with the same public name that an
3299 error should be diagnosed.
3301 9. When Exim looked up a host name for an IP address, but failed to find the
3302 original IP address when looking up the host name (a safety check), it
3303 output the message "<ip address> does not match any IP for NULL", which was
3304 confusing, to say the least. The bug was that the host name should have
3305 appeared instead of "NULL".
3307 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3308 user that is built into the binary, and the -C or -D options is used, root
3309 privilege is dropped before the configuration file is read. In addition,
3310 logging is switched to stderr instead of the normal log files. If the
3311 configuration then re-defines the Exim user, the unprivileged environment
3312 is probably not what is expected, so Exim logs a panic warning message (but
3315 However, if deliver_drop_privilege is set, the unprivileged state may well
3316 be exactly what is intended, so the warning has been cut out in that case,
3317 and Exim is allowed to try to write to its normal log files.
3323 1. smtp_return_error_details was not giving details for temporary sender
3324 or receiver verification errors.
3326 2. Diagnose a configuration error if two authenticators have the same public
3329 3. Exim used not to create the message log file for a message until the first
3330 delivery attempt. This could be confusing when incoming messages were held
3331 for policy or load reasons. The message log file is now created at the time
3332 the message is received, and an initial "Received" line is written to it.
3334 4. The automatically generated man page for command line options had a minor
3335 bug that caused no ill effects; however, a more serious problem was that
3336 the procedure for building the man page automatically didn't always
3337 operate. Consequently, release 4.20 contains an out-of-date version. This
3338 shouldn't happen again.
3340 5. When building Exim with embedded Perl support, the script that builds the
3341 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3342 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3344 6. The freeze_tell option was not being used for messages that were frozen on
3345 arrival, either by an ACL or by local_scan().
3347 7. Added the smtp_incomplete_transaction log selector.
3349 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3350 was accepting AUTH without a new EHLO.
3352 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3353 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3356 10. Logging of TCP/IP connections (when configured) now happens in the main
3357 daemon process instead of the child process, so that the TCP/IP connection
3358 count is more accurate (but it can never be perfect).
3360 11. The use of "drop" in a nested ACL was not being handled correctly in the
3361 outer ACL. Now, if condition failure induced by the nested "drop" causes
3362 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3363 or "require"), the connection is dropped.
3365 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3366 that yield "discard" can now be used with an "accept" or a "discard" verb,
3367 but an error is generated for any others (because I can't see a useful way
3368 to define what should happen).
3370 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3371 are now processed in the same way as lines in the Exim configuration file.
3372 In particular, continuation lines are supported.
3374 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3376 15. Added -ti meaning -t -i.
3378 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3379 domains, and warn by logging if others are found.
3381 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3382 returned from GIFCONF when Exim is trying to find the list of interfaces on
3383 a host. The code in os.c has been modified to copy each ifreq to an aligned
3384 structure in all cases.
3386 Also, in some cases, the returned ifreq's were being copied to a 'struct
3387 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3388 means the last couple of bytes of an IPv6 address could be chopped if the
3389 ifreq contained only a normal sockaddr (14 bytes storage).
3391 18. Named domain lists were not supported in the hosts_treat_as_local option.
3392 An entry such as +xxxx was not recognized, and was treated as a literal
3395 19. Ensure that header lines added by a DATA ACL are included in the reject log
3396 if the ACL subsequently rejects the message.
3398 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3399 MD5 (which is deprecated).
3401 21. When testing a filter file using -bf, Exim was writing a message when it
3402 took the sender from a "From " line in the message, but it was not doing so
3403 when it took $return_path from a Return-Path: header line. It now does.
3405 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3406 with a valid header line field name (a series of printing characters
3407 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3409 23. Changed "disc" in the source to "disk" to conform to the documentation and
3410 the book and for uniformity.
3412 24. Ignore Sendmail's -Ooption=value command line item.
3414 25. When execve() failed while trying to run a command in a pipe transport,
3415 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3416 could be confused with a return value of 69 from the command itself. This
3417 has been changed to 127, the value the shell returns if it is asked to run
3418 a non-existent command. The wording for the related log line suggests a
3419 non-existent command as the problem.
3421 26. If received_header_text expands to an empty string, do not add a Received:
3422 header line to the message. (Well, it adds a token one on the spool, but
3423 marks it "old" so that it doesn't get used or transmitted.)
3425 27. Installed eximstats 1.28 (addition of -nt option).
3427 28. There was no check for failure on the call to getsockname() in the daemon
3428 code. This can fail if there is a shortage of resources on the system, with
3429 ENOMEM, for example. A temporary error is now given on failure.
3431 29. Contrary to the C standard, it seems that in some environments, the
3432 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3433 program. Exim now does this explicitly; it affects the formatting of
3434 timestamps using strftime().
3436 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3437 complaints. I've now initialized all the variables, to avoid this.
3439 32. Header lines added by a system filter were not being "seen" during
3440 transport-time rewrites.
3442 33. The info_callback() function passed to OpenSSL is set up with type void
3443 (*)(SSL *, int, int), as described somewhere. However, when calling the
3444 function (actually a macro) that sets it up, the type void(*)() is
3445 expected. I've put in a cast to prevent warnings from picky compilers.
3447 34. If a DNS black list lookup found a CNAME record, but there were no A
3448 records associated with the domain it pointed at, Exim crashed.
3450 35. If a DNS black list lookup returned more than one A record, Exim ignored
3451 all but the first. It now scans all returned addresses if a particular IP
3452 value is being sought. In this situation, the contents of the
3453 $dnslist_value variable are a list of all the addresses, separated by a
3456 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3457 to accept a host name and all its aliases if the forward lookup for any of
3458 them yielded the IP address of the incoming connection. Now it accepts only
3459 those names whose forward lookup yields the correct IP address. Any other
3460 names are discarded. This closes a loophole whereby a rogue DNS
3461 administrator could create reverse DNS records to break through a
3462 wildcarded host restriction in an ACL.
3464 37. If a user filter or a system filter that ran in a subprocess used any of
3465 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3466 the wrong values were passed to the pipe command ($thisaddress had the
3467 value of $0, $0 had the value of $1, etc). This bug was introduced by
3468 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3470 38. Improved the line breaking for long SMTP error messages from ACLs.
3471 Previously, if there was no break point between 40 and 75 characters, Exim
3472 left the rest of the message alone. Two changes have been made: (a) I've
3473 reduced the minimum length to 35 characters; (b) if it can't find a break
3474 point between 35 and 75 characters, it looks ahead and uses the first one
3475 that it finds. This may give the occasional overlong line, but at least the
3476 remaining text gets split now.
3478 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3479 file descriptors might be low, and that setting 1000 would always raise it.
3480 It turns out that in some environments, the limit is already over 1000 and
3481 that lowering it causes trouble. So now Exim takes care not to decrease it.
3483 40. When delivering a message, the value of $return_path is set to $sender_
3484 address at the start of routing (routers may change the value). By an
3485 oversight, this default was not being set up when an address was tested by
3486 -bt or -bv, which affected the outcome if any router or filter referred to
3489 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3490 log only when "message" or "log_message" are set. However, if one of the
3491 conditions was an address verification, or a call to a nested ACL, the
3492 messages generated by the underlying test were being passed through. This
3493 no longer happens. The underlying message is available in $acl_verify_
3494 message for both "message" and "log_message" expansions, so it can be
3495 passed through if needed.
3497 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3498 new expansion $bh_ to give the encoded byte string without charset
3499 translation. Translation happens only if iconv() is available; HAVE_ICONV
3500 indicates this at build time. HEADERS_CHARSET gives the charset to
3501 translate to; headers_charset can change it in the configuration, and
3502 "headers charset" can change it in an individual filter file.
3504 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3505 that creates RFC 2047 encoded "words" labels them as that charset instead
3506 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3507 expansion operator; (ii) when Exim creates a From: line for a local
3508 message; (iii) when a header line is rewritten to include a "phrase" part.
3510 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3511 buggy, causing it to skip the first lines of messages whose message ID
3512 ended in 'D'. This would not have bitten before Exim release 4.14, because
3513 message IDs were unlikely to end in 'D' before then. The effect was to have
3514 incorrect size information for certain domains.
3516 45. #include "config.h" was missing at the start of the crypt16.c module. This
3517 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3520 46. If there was a timeout during a "random" callout check, Exim treated it as
3521 a failure of the random address, and carried on sending RSET and the real
3522 address. If the delay was just some slowness somewhere, the response to the
3523 original RCPT would be taken as a response to RSET and so on, causing
3524 mayhem of various kinds.
3526 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3527 when I implemented it. It didn't allow for the fact that some option values
3528 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3529 the right test for positive values.
3531 48. Domain names in DNS records are case-independent. Exim always looks them up
3532 in lower case. Some resolvers return domain names in exactly the case they
3533 appear in the zone file, that is, they may contain uppercase letters. Not
3534 all resolvers do this - some return always lower case. Exim was treating a
3535 change of case by a resolver as a change of domain, similar to a widening
3536 of a domain abbreviation. This triggered its re-routing code and so it was
3537 trying to route what was effectively the same domain again. This normally
3538 caused routing to fail (because the router wouldn't handle the domain
3539 twice). Now Exim checks for this case specially, and just changes the
3540 casing of the domain that it ultimately uses when it transmits the message
3543 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3546 50. If a filter generated a file delivery with a non-absolute name (possible if
3547 no home directory exists for the router), the forbid_file option was not
3550 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3551 the existing equality matching.
3553 52. Exim was using ints instead of ino_t variables in some places where it was
3554 dealing with inode numbers.
3556 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3557 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3558 called TMPDIR, and if it finds it is different, it changes its value.
3560 54. The smtp_printf() function is now made available to local_scan() so
3561 additional output lines can be written before returning. There is also an
3562 smtp_fflush() function to enable the detection of a dropped connection.
3563 The variables smtp_input and smtp_batched_input are exported to
3566 55. Changed the default runtime configuration: the message "Unknown user"
3567 has been removed from the ACL, and instead placed on the localuser router,
3568 using the cannot_route_message feature. This means that any verification
3569 failures that generate their own messages won't get overridden. Similarly,
3570 the "Unrouteable address" message that was in the ACL for unverifiable
3571 relay addresses has also been removed.
3573 56. Added hosts_avoid_esmtp to the smtp transport.
3575 57. The exicyclog script was not checking for the esoteric option
3576 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3577 will work only if exicyclog is run under the appropriate euid.
3579 58. Following a discussion on the list, the rules by which Exim recognises line
3580 endings on incoming messages have been changed. The -dropcr and drop_cr
3581 options are now no-ops, retained only for backwards compatibility. The
3582 following line terminators are recognized: LF CRLF CR. However, special
3583 processing applies to CR:
3585 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3586 nor a local message in the state where . is a terminator.
3588 (ii) If a bare CR is encountered in a header line, an extra space is added
3589 after the line terminator so as not to end the header. The reasoning
3590 behind this is that bare CRs in header lines are most likely either
3591 to be mistakes, or people trying to play silly games.
3593 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3594 was being incorrectly given as 18 bytes larger than it should have been.
3595 This is a VOB (very old bug).
3597 60. This may never have affected anything current, but just in case it has:
3598 When the local host is found other than at the start of a list of hosts,
3599 the local host, those with the same MX, and any that follow, are discarded.
3600 When the list in question was part of a longer list of hosts, the following
3601 hosts (not currently being processed) were also being discarded. This no
3602 longer happens. I'm not sure if this situation could ever has previously
3605 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3608 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3609 header. This is something that is recommended in a new Internet Draft, and
3610 is something that is documented as being done by Sendmail. There are two
3611 possible values. For messages generated by the autoreply transport, Exim
3614 Auto-Submitted: auto-replied
3616 whereas for all other generated messages (e.g. bounces) it adds
3618 Auto-Submitted: auto-generated
3620 63. The "personal" condition in filters now includes a test for the
3621 Auto-Submitted: header. If it contains the string "auto-" the message it
3622 not considered personal.
3624 64. Added rcpt_include_affixes as a generic transport option.
3626 65. Added queue_only_override (default true).
3628 66. Added the syslog_duplication option.
3630 67. If what should have been the first header line of a message consisted of
3631 a space followed by a colon, Exim was mis-interpreting it as a header line.
3632 It isn't of course - it is syntactically invalid and should therefore be
3633 treated as the start of the message body. The misbehaviour could have
3634 caused a number of strange effects, including loss of data in subsequent
3635 header lines, and spool format errors.
3637 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3638 client host had authenticated. This control can now be exercised by an ACL
3639 for more flexibility.
3641 69. By default, callouts do not happen when testing with -bh. There is now a
3642 variant, -bhc, which does actually run the callout code, including
3643 consulting and updating the callout cache.
3645 70. Added support for saslauthd authentication, courtesy of Alexander
3648 71. If statvfs() failed on the spool or log directories while checking their
3649 size for availability, Exim confusingly gave the error "space shortage".
3650 Furthermore, in debugging mode it crashed with a floating point exception.
3651 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3652 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3653 serious problem, Exim now writes to the main and panic logs when this
3654 happens, with details of the failure. It then refuses to accept the
3655 incoming message, giving the message "spool directory problem" or "log
3656 directory problem" with a 421 code for SMTP messages.
3658 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3659 0, 1, and 2 exist, because some OS complain for execs without them (see
3660 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3661 descriptors. However, the code omitted to check that the open succeeded,
3662 causing mysterious errors if for some reason the permissions on /dev/null
3663 got screwed. Now Exim writes a message to the main and panic logs, and
3664 bombs out if it can't open /dev/null.
3666 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3667 interact so that it is all more flexible. It is supposed to remain
3668 backwards compatible. Also added extra_local_interfaces.
3670 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3671 to bomb out with an assertion failure - to the client this appears as a
3672 connection drop. This problem occurs in the part of the code that was taken
3673 from the Samba project. Fortunately, the assertion is in a very simple
3674 function, so I have fixed this by reproducing the function inline in the
3675 one place where it is called, and arranging for authentication to fail
3676 instead of killing the process with assert().
3678 75. The SPA client code was not working when the server requested OEM rather
3679 than Unicode encoding.
3681 76. Added code to make require_files with a specific uid setting more usable in
3682 the case where statting the file as root fails - usually a non-root-mounted
3683 NFS file system. When this happens and the failure is EACCES, Exim now
3684 forks a subprocess and does the per-uid checking as the relevant uid.
3686 77. Added process_log_path.
3688 78. If log_file_path was not explicitly set, a setting of check_log_space or
3689 check_log_inodes was ignored.
3691 79. If a space check for the spool or log partitions fails, the incident is now
3692 logged. Of course, in the latter case the data may get lost...
3694 80. Added the %p formatting code to string_format() so that it can be used to
3695 print addresses in debug_print(). Adjusted all the address printing in the
3696 debugging in store.c to use %p rather than %d.
3698 81. There was a concern that a line of code in smtp_in.c could overflow a
3699 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3700 initially expressed, the concern was not well-founded, because trailing
3701 spaces are removed early. However, if the trailing spaces were followed by
3702 a NULL, they did not get removed, so the overflow was possible. Two fixes
3705 (a) I re-wrote the offending code in a cleaner fashion.
3706 (b) If an incoming SMTP command contains a NULL character, it is rejected
3709 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3710 runs initgroups(), so that if the Exim user is in any additional groups,
3711 they will be used during message reception.
3717 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob