1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.288 2006/02/08 16:10:46 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.
114 PH/01 Two changes to the default runtime configuration:
116 (1) Move the checks for relay_from_hosts and authenticated clients from
117 after to before the (commented out) DNS black list checks.
119 (2) Add control=submission to the relay_from_hosts and authenticated
120 clients checks, on the grounds that messages accepted by these
121 statements are most likely to be submissions.
123 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
125 (1) Generate an error if the third argument for the ${prvs expansion is
128 (2) Treat a missing third argument of ${prvscheck as if it were an empty
131 (3) Reset the variables that are obtained from the first argument of
132 ${prvscheck and used in the second argument before leaving the code,
133 because their memory is reclaimed, so using them afterwards may do
136 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
137 one (it's much easier than Tom thought :-).
139 (5) Because of (4), we can now allow for the use of $prvscheck_result
140 inside the third argument.
142 PH/03 For some reason, the default setting of PATH when running a command from
143 a pipe transport was just "/usr/bin". I have changed it to
146 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
147 anything to be listed in the output from -bV.
149 PH/05 When a filter generated an autoreply, the entire To: header line was
150 quoted in the delivery log line, like this:
152 => >A.N.Other <ano@some.domain> <original@ddress> ...
154 This has been changed so that it extracts the operative address. There
155 may be more than one such address. If so, they are comma-separated, like
158 => >ano@some.domain,ona@other.domain <original@ddress> ...
160 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
161 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
162 not being looked up in the rDNS to get a host name, Exim was showing the
163 IP address twice in Received: lines, even though the IP addresses were
164 identical. For example:
166 Received: from [1.2.3.4] (helo=[1.2.3.4])
168 However, if the real host name was known, it was omitting the HELO data
169 if it matched the actual IP address. This has been tidied up so that it
170 doesn't show the same IP address twice.
172 PH/07 When both +timestamp and +memory debugging was on, the value given by
173 $tod_xxx expansions could be wrong, because the tod_stamp() function was
174 called by the debug printing, thereby overwriting the timestamp buffer.
175 Debugging no longer uses the tod_stamp() function when +timestamp is set.
177 PH/08 When the original message was included in an autoreply transport, it
178 always said "this is a copy of the message, including all the headers",
179 even if body_only or headers_only was set. It now gives an appropriate
182 PH/09 Applied a patch from the Sieve maintainer which:
184 o fixes some comments
185 o adds the (disabled) notify extension core
186 o adds some debug output for the result of if/elsif tests
187 o points to the current vacation draft in the documentation
188 and documents the missing references header update
192 o fixes a bug in processing the envelope test (when testing
193 multiple envelope elements, the last element determinted the
196 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
197 Electronic Mail") by including:
199 Auto-submitted: auto-generated
201 in the messages that it generates (bounce messages and others, such as
202 warnings). In the case of bounce messages for non-SMTP mesages, there was
203 also a typo: it was using "Auto_submitted" (underscore instead of
204 hyphen). Since every message generated by Exim is necessarily in response
205 to another message, thes have all been changed to:
207 Auto-Submitted: auto-replied
209 in accordance with these statements in the RFC:
211 The auto-replied keyword:
213 - SHOULD be used on messages sent in direct response to another
214 message by an automatic process,
216 - MUST NOT be used on manually-generated messages,
218 - MAY be used on Delivery Status Notifications (DSNs) and Message
219 Disposition Notifications (MDNs),
221 - MUST NOT be used on messages generated by automatic or periodic
222 processes, except for messages which are automatic responses to
225 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
226 to the default Received: header definition.
228 PH/12 Added log selector acl_warn_skipped (default on).
230 PH/13 After a successful wildlsearch lookup, discard the values of numeric
231 variables because (a) they are in the wrong storage pool and (b) even if
232 they were copied, it wouldn't work properly because of the caching.
234 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
235 checking when decoding. Apparently there are clients that generate
236 overlong encoded strings. Why am I not surprised?
238 PH/15 If the first argument of "${if match_address" was not empty, but did not
239 contain an "@" character, Exim crashed. Now it writes a panic log message
240 and treats the condition as false.
242 PH/16 In autoreply, treat an empty string for "once" the same as unset.
244 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
245 extension "envelope-auth". The code is finished and in agreement with
246 other implementations, but there is no documentation so far and in fact,
247 nobody wrote the draft yet. This extension is currently #undef'ed, thus
248 not changing the active code.
250 Print executed "if" and "elsif" statements when debugging is used. This
251 helps a great deal to understand what a filter does.
253 Document more things not specified clearly in RFC3028. I had all this
254 sorted out, when out of a sudden new issues came to my mind. Oops."
256 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
259 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
260 canonical form (as documented). However, after a host name lookup from
261 the IP address, check_host() was doing a simple string comparison with
262 addresses acquired from the DNS when checking that the found name did
263 have the original IP as one of its addresses. Since any found IPv6
264 addresses are likely to be in abbreviated form, the comparison could
265 fail. Luckily, there already exists a function for doing the comparison
266 by converting both addresses to binary, so now that is used instead of
269 PH/20 There was another similar case to PH/19, when a complete host name was
270 given in a host list; looking up its IP address could give an abbreviated
271 form, whereas the current host's name might or might not be abbreviated.
272 The same fix has been applied.
278 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
279 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
282 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
283 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
285 PH/03 Typo: missing ".o" in src/pcre/Makefile.
287 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
288 header line, restrict the check to what is listed in RFCs 2369 and 2929.
289 Also, for "Auto-Submitted", treat anything other than "no" as
290 non-personal, in accordance with RFC 3834. (Previously it treated
291 anything starting "auto-" as non-personal.)
293 TF/01 The control=submission/name=... option had a problem with syntax
294 errors if the name included a slash character. The /name= option
295 now slurps the rest of the string, so it can include any characters
296 but it must come last in the list of options (after /sender_retain
299 PH/05 Some modifications to the interface to the fake nameserver for the new
307 TK/01 Added the "success_on_redirect" address verification option. See
308 NewStuff for rationale and an example.
310 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
312 PH/02 Patch to exigrep to allow it to work on syslog lines.
314 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
315 fread() to skip over the body file's header line, because in Cygwin the
316 header line is locked and is inaccessible.
318 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
319 co-exist for some time) to make it clear that it is the Exim ID that is
320 referenced, not the Message-ID: header line.
322 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
323 string_format() function, because snprintf() does not exist on all
326 PH/06 The use of forbid_filter_existstest now also locks out the use of the
327 ${stat: expansion item.
329 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
330 protocol synchronization error", to keep the pedants happy.
332 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
333 well as for IRIX systems, when gcc is being used. See the host.c source
336 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
338 PH/10 Named domain lists were not working if used in a queue_smtp_domains
341 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
342 transport and to the smtp transport in LMTP mode.
344 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
346 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
347 run a filter in a subprocess. This could lead to confusion in subsequent
348 lookups in the parent process. There should also be a search_tidyup() at
349 the end of the subprocess.
351 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
352 only if the host matched helo_try_verify_hosts, which caused the
353 verification to occur when the EHLO/HELO command was issued. The ACL just
354 tested the remembered result. Now, if a previous verification attempt has
355 not happened, "verify = helo" does it there and then.
357 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
359 TK/03 Fix log output including CR from clamd.
361 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
362 exist provoked a memory error which could cause a segfault.
364 PH/15 Installed PCRE 6.2
366 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
368 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
369 of the problem. Specifically, suggested +O2 rather than +O1 for the
372 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
374 PH/20 If a delivery was routed to a non-standard port by means of an SRV
375 record, the port was not correctly logged when the outgoing_port log
376 selector was set (it logged the transort's default port).
378 PH/21 Added support for host-specific ports to manualroute, queryprogram,
379 fallback_hosts, and "hosts" in the smtp transport.
381 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
382 host errors such as "Connection refused".
384 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
385 authentication with radiusclient 0.4.9:
387 - Error returned from rc_read_config was caught wrongly
388 - Username/password not passed on to radius server due to wrong length.
390 The presumption is that some radiusclient API changes for 4.51/PH/17
391 were not taken care of correctly. The code is still untested by me (my
392 Linux distribution still has 0.3.2 of radiusclient), but it was
393 contributed by a Radius user.
395 PH/24 When doing a callout, the value of $domain wasn't set correctly when
396 expanding the "port" option of the smtp transport.
398 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
399 while reading a MIME header. Thanks to Tom Hughes for a patch.
401 PH/24 Include config.h inside local_scan.h so that configuration settings are
404 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
405 that in an ACL for RCPT (for example), you can examine exactly what was
408 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
409 commands, but it was not correctly comparing the address with the actual
410 client host address. Thus, it would show the EHLO address in Received:
411 header lines when this was not necessary.
413 PH/27 Added the % operator to ${eval:}.
415 PH/28 Exim tries to create and chdir to its spool directory when it starts;
416 it should be ignoring failures (because with -C, for example, it has lost
417 privilege). It wasn't ignoring creation failures other than "already
420 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
423 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
424 because an input line was too long, either on its own, or by virtue of
425 too many continuations, the temporary file was not being removed, and the
426 return code was incorrect.
428 PH/31 Missing "BOOL" in function definition in filtertest.c.
430 PH/32 Applied Sieve patches from the maintainer.
432 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
434 PH/33 Added "verify = not_blind".
436 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
437 Local/Makefile (with some defaults set). These are used in built scripts
438 such as exicyclog, but they have never been used in the exim_install
439 script (though there are many overriding facilities there). I have
440 arranged that the exim_install script now takes note of these two
443 PH/35 Installed configuration files for Dragonfly.
445 PH/36 When a locally submitted message by a trusted user did not contain a
446 From: header, and the sender address was obtained from -f or from an SMTP
447 MAIL command, and the trusted user did not use -F to supply a sender
448 name, $originator_name was incorrectly used when constructing a From:
449 header. Furthermore, $originator_name was used for submission mode
450 messages from external hosts without From: headers in a similar way,
451 which is clearly wrong.
453 PH/37 Added control=suppress_local_fixups.
455 PH/38 When log_selector = +received_sender was set, and the addition of the
456 sender made the log line's construction buffer exactly full, or one byte
457 less than full, an overflow happened when the terminating "\n" was
460 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
461 when the result of a list match is failure because a DNS lookup failed.
463 PH/40 RM_COMMAND is now used in the building process.
465 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
466 the "build-* directories that it finds.
468 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
469 address in a domain literal was a prefix of an interface address.
471 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
472 when verifying a sender address, unless rewrite_headers is false.
474 PH/44 Wrote a long comment about why errors_to addresses are verified as
475 recipients, not senders.
477 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
478 the ratelimit ACL was added.
480 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
482 PH/46 Added extra information about PostgreSQL errors to the error string.
484 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
485 avoiding the need to install special zones in a real server. This is
486 backwards compatible; if it can't find the fake resolver, it drops back.
487 Thus, both old and new test suites can be run.
489 TF/02 Added util/ratelimit.pl
491 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
492 clock is set back in time.
494 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
495 Candler <B.Candler@pobox.com>.
497 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
498 OK for addresses that are the result of redirections.
500 PH/48 A number of further additions for the benefit of the new test suite,
501 including a fake gethostbyname() that interfaces to the fake DNS resolver
504 TF/06 The fix for widen_domains has also been applied to qualify_single and
505 search_parents which are the other dnslookup options that can cause
508 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
511 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
513 TF/07 Exim produced the error message "an SRV record indicated no SMTP
514 service" if it encountered an MX record with an empty target hostname.
515 The message is now "an MX or SRV record indicated no SMTP service".
517 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
518 if the DNS of the sending site is misconfigured. This is quite a
519 common situation. This change restores the behaviour of treating a
520 helo verification defer as a failure.
522 PH/51 If self=fail was set on a router, the bounce message did not include the
523 actual error message.
529 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
531 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
532 command itself ended in error, the underlying message about the transport
533 filter timeout was being overwritten with the pipe command error. Now the
534 underlying error message should be appended to the second error message.
536 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
538 PH/02 Reduce the amount of output that "make" produces by default. Full output
539 can still be requested.
541 PH/03 The warning log line about a condition test deferring for a "warn" verb
542 was being output only once per connection, rather than after each
543 occurrence (because it was using the same function as for successful
544 "warn" verbs). This seems wrong, so I have changed it.
546 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
547 it should not have, which might have caused a crash in the right
548 circumstances, but probably never did.
550 PH/04 Installed a modified version of Tony Finch's patch to make submission
551 mode fix the return path as well as the Sender: header line, and to
552 add a /name= option so that you can make the user's friendly name appear
555 TF/03 Added the control = fakedefer ACL modifier.
557 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
558 Mark Lowes for thorough testing.
560 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
562 TK/03 Merged latest SRS patch from Miles Wilton.
564 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
565 with the definition in sysexits.h (which is #included earlier).
566 Fortunately, Exim does not actually use EX_OK. The code used to try to
567 preserve the sysexits.h value, by assumimg that macro definitions were
568 scanned for macro replacements. I have been disabused of this notion,
569 so now the code just undefines EX_OK before #including unistd.h.
571 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
572 in the smtp transport. When a block could not be written in a single
573 write() function, the timeout was being re-applied to each part-write.
574 This seems wrong - if the receiver was accepting one byte at a time it
575 would take for ever. The timeout is now adjusted when this happens. It
576 doesn't have to be particularly precise.
578 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
579 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
581 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
582 without a domain if the check to <postmaster@domain> fails.
584 SC/01 Eximstats: added -xls and the ability to specify output files
585 (patch written by Frank Heydlauf).
587 SC/02 Eximstats: use FileHandles for outputing results.
589 SC/03 Eximstats: allow any combination of xls, txt, and html output.
591 SC/04 Eximstats: fixed display of large numbers with -nvr option
593 SC/05 Eximstats: fixed merging of reports with empty tables.
595 SC/06 Eximstats: added the -include_original_destination flag
597 SC/07 Eximstats: removed tabs and trailing whitespace.
599 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
601 TK/06 MBOX spool code: Add real "From " MBOX separator line
602 so the .eml file is really in mbox format (even though
603 most programs do not really care). Patch from Alex Miller.
605 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
606 The latter is generated from $received_to and is only set if the
607 message has one envelope recipient. SA can use these headers,
608 obviously out-of-the-box. Patch from Alex Miller.
610 PH/08 The ${def test on a variable was returning false if the variable's
611 value was "0", contrary to what the specification has always said!
612 The result should be true unless the variable is empty.
614 PH/09 The syntax error of a character other than { following "${if
615 def:variable_name" (after optional whitespace) was not being diagnosed.
616 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
617 accidental colon was present, for example, could give incorrect results.
619 PH/10 Tidied the code in a number of places where the st_size field of a stat()
620 result is used (not including appendfile, where other changes are about
623 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
624 This involved changing a lot of size variables from int to off_t. It
625 should work with maildirs and everything.
627 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
628 spamd dying while we are connected to it.
630 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
633 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
634 function for BDB 4.3.
636 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
638 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
639 the sources for PCRE have been re-arranged and more files are now
642 PH/15 The code I had for printing potentially long long variables in PH/11
643 above was not the best (it lost precision). The length of off_t variables
644 is now inspected at build time, and an appropriate printing format (%ld
645 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
646 to be "long long int" or "long int". This is needed for the internal
647 formatting function string_vformat().
649 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
650 the configuration file to be ":syslog", then the script "guesses" where
651 the logs files are, rather than using the compiled in default. In our
652 case the guess is not the same as the compiled default, so the script
653 suddenly stopped working when I started to use syslog. The patch checks
654 to see if log_file_path is "". If so, it attempts to read it from exim
655 with no configuration file to get the compiled in version, before it
656 falls back to the previous guessing code."
658 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
659 implementing BATV in an Exim configuration. See NewStuff for the gory
662 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
663 Makefile that are specific to HP-UX.
665 PH/18 If the "use_postmaster" option was set for a recipient callout together
666 with the "random" option, the postmaster address was used as the MAIL
667 FROM address for the random test, but not for the subsequent recipient
668 test. It is now used for both.
670 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
671 patch removes a few documentation additions to RFC 3028, because the
672 latest draft now contains them. It adds the new en;ascii-case comparator
673 and a new error check for 8bit text in MIME parts. Comparator and
674 require names are now matched exactly. I enabled the subaddress
675 extension, but it is not well tested yet (read: it works for me)."
677 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
678 rework some of the code of TK/09 above to avoid the hardwired use of
679 "%lld" and "long long". Replaced the call to snprintf() with a call to
682 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
683 records point to non-existent hosts", "retry timeout exceeded", and
684 "retry time not reached for any host after a long failure period".
686 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
687 experimental DomainKeys support:
689 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
690 (2) On an error such as an illegally used "control", the wrong name for
691 the control was given.
693 These problems did NOT occur unless DomainKeys support was compiled.
695 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
697 PH/24 Added ${if match_ip condition.
699 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
700 databases so that it will be absolutely obvious if a crash occurs in the
701 DB library. This is a regular occurrence (often caused by mis-matched
704 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
705 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
706 picked up on a user's system that detects such things. There doesn't seem
707 to be a gcc warning option for this - only an attribute that has to be
708 put on the function's prototype. It seems that in Fedora Core 4 they have
709 set this on a number of new functions. No doubt there will be more in due
712 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
713 specify a transport. However, if an address that was verified by such a
714 router was the subject of a callout, Exim crashed because it tried to
715 read the rcpt_include_affixes from the non-existent transport. Now it
716 just assumes that the setting of that option is false. This bug was
717 introduced by 4.51/PH/31.
719 PH/28 Changed -d+all to exclude +memory, because that information is very
720 rarely of interest, but it makes the output a lot bigger. People tend to
721 do -d+all out of habit.
723 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
724 code in os-type was giving problems when libc.so lives in lib64, like on
727 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
728 aren't the modern standard, and it seems that some systems' include files
729 don't always have them. Exim was already checking for some of the newer
730 ones like T_AAAA, and defining it itself. I've added checks for all the
731 record types that Exim uses.
733 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
734 not automatically generating a new one, as it is supposed to. This
735 prevented TLS from working. If the file did exist, but contained invalid
736 data, a new version was generated, as expected. It was only the case of a
737 non-existent file that was broken.
739 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
740 with a change in libdomainkeys > 0.64.
742 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
743 from DNS. If the selector record carries the flag, it now has
744 precedence over the domain-wide flag.
746 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
748 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
749 the use of a port name that isn't defined in /etc/services) occurred, the
750 message was deferred as in a normal delivery, and thus remained on the
751 spool, instead of being failed because of the mua_wrapper setting. This
752 is now fixed, and I tidied up some of the mua_wrapper messages at the
755 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
756 the messages in a hash of arrays rather than using individual hashes.
757 This is a bit cleaner and results in dramatic memory savings, albeit
758 at a slight CPU cost.
760 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
761 as requested by Marc Sherman.
763 SC/10 Eximstats: added histograms for user specified patterns as requested
766 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
768 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
769 fopen() in the content-scanning modules that did not already have it.
775 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
776 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
778 TK/02 Fix ACL "control" statement not being available in MIME ACL.
780 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
782 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
783 to test Sieve filters that use "vacation".
785 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
786 that changes the way the GnuTLS parameters are stored in the cache file.
787 The new format can be generated externally. For backward compatibility,
788 if the data in the cache doesn't make sense, Exim assumes it has read an
789 old-format file, and it generates new data and writes a new file. This
790 means that you can't go back to an older release without removing the
793 PH/03 A redirect router that has both "unseen" and "one_time" set does not
794 work if there are any delivery delays because "one_time" forces the
795 parent to be marked "delivered", so its unseen clone is never tried
796 again. For this reason, Exim now forbids the simultaneous setting of
799 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
800 redirected to themselves ("homonym" addresses). Read the long ChangeLog
801 entry if you want to know the details. The fix, however, neglected to
802 consider the case when local delivery batching is involved. The test for
803 "previously delivered" was not happening when checking to see if an
804 address could be batched with a previous (undelivered) one; under
805 certain circumstances this could lead to multiple deliveries to the same
808 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
809 in its include files, and this causes problems building Exim.
811 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
812 header_syntax) but Exim was just ignoring anything given after a slash.
813 In particular, this caused confusion with an attempt to use "verify =
814 reverse_host_lookup/defer_ok". An error is now given when options are
815 supplied for verify items that do not have them. (Maybe reverse_host_
816 lookup should have a defer_ok option, but that's a different point.)
818 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
819 defined by RFC 821) to 2048, because there were problems with some AUTH
820 commands, and RFC 1869 says the size should be increased for extended
821 SMTP commands that take arguments.
823 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
826 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
827 "unknown" error; now it says that the functionality isn't in the binary.
829 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
830 an address' error message when a string expansion fails (syntax or
831 whatever). Otherwise the password may appear in the log. Following change
832 PH/42 below, there is no longer a chance of it appearing in a bounce
835 PH/11 Installed exipick version 20050225.0 from John Jetmore.
837 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
838 of its addresses was ever tried. (Bugzilla bug #2.)
840 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
841 the result incorrectly in the debug output. (It correctly added a newline
842 to what was transported.)
844 TF/01 Added $received_time.
846 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
847 commented out examples of how to interface to a virus scanner and to
848 SpamAssassin. Also added commented examples of av_scanner and
849 spamd_address settings.
851 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
852 and controls are allowed in which ACLs. There were a couple of minor
853 errors. Some of the entries in the conditions table (which is a table of
854 where they are NOT allowed) were getting very unwieldy; rewrote them as a
855 negation of where the condition IS allowed.
857 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
859 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
860 header file does not have a version number, so I've had to invent a new
861 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
862 API. The code is untested by me (my Linux distribution still has 0.3.2 of
863 radiusclient), but it was contributed by a Radius user.
865 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
866 files or directories, for OpenSSL.
868 PH/19 When an Exim process that is running as root has to create an Exim log
869 file, it does so in a subprocess that runs as exim:exim so as to get the
870 ownership right at creation (otherwise, other Exim processes might see
871 the file with the wrong ownership). There was no test for failure of this
872 fork() call, which would lead to the process getting stuck as it waited
873 for a non-existent subprocess. Forks do occasionally fail when resources
874 run out. I reviewed all the other calls to fork(); they all seem to check
877 PH/20 When checking for unexpected SMTP input at connect time (before writing
878 the banner), Exim was not dealing correctly with a non-positive return
879 from the read() function. If the client had disconnected by this time,
880 the result was a log entry for a synchronization error with an empty
881 string after "input=" when read() returned zero. If read() returned -1
882 (an event I could not check), uninitialized data bytes were printed.
883 There were reports of junk text (parts of files, etc) appearing after
886 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
888 PH/22 Added support for macro redefinition, and (re)definition in between
889 driver and ACL definitions.
891 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
892 forgetting to use the resulting value; it was using the unexpanded value.
894 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
895 hadn't been configured. The fix is from Juergen Kreileder, who
896 understands it better than I do:
898 "Here's what I see happening with three configured cyrus_sasl
899 authenticators configured (plain, login, cram-md5):
901 On startup auth_cyrus_sasl_init() gets called for each of these.
902 This means three calls to sasl_listmech() without a specified mech_list.
903 => SASL tests which mechs of all available mechs actually work
904 => three warnings about OTP not working
905 => the returned list contains: plain, login, cram-md5, digest-md5, ...
907 With the patch, sasl_listmech() also gets called three times. But now
908 SASL's mech_list option is set to the server_mech specified in the the
909 authenticator. Or in other words, the answer from sasl_listmech()
910 gets limited to just the mech you're testing for (which is different
912 => the return list contains just 'plain' or 'login', 'cram-md5' or
913 nothing depending on the value of ob->server_mech.
915 I've just tested the patch: Authentication still works fine,
916 unavailable mechs specified in the exim configuration are still
917 caught, and the auth.log warnings about OTP are gone."
919 PH/25 When debugging is enabled, the contents of the command line are added
920 to the debugging output, even when log_selector=+arguments is not
923 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
924 answer is "GNU", and only if the return is "GNU/something" is the answer
927 PH/27 $acl_verify_message is now set immediately after the failure of a
928 verification in an ACL, and so is available in subsequent modifiers. In
929 particular, the message can be preserved by coding like this:
931 warn !verify = sender
932 set acl_m0 = $acl_verify_message
934 Previously, $acl_verify_message was set only while expanding "message"
935 and "log_message" when a very denied access.
937 PH/28 Modified OS/os.c-Linux with
939 -#ifndef OS_LOAD_AVERAGE
940 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
942 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
943 nomenclature these days.)
945 PH/29 Installed patch from the Sieve maintainer that adds the options
946 sieve_useraddress and sieve_subaddress to the redirect router.
948 PH/30 In these circumstances:
949 . Two addresses routed to the same list of hosts;
950 . First host does not offer TLS;
951 . First host accepts first address;
952 . First host gives temporary error to second address;
953 . Second host offers TLS and a TLS session is established;
954 . Second host accepts second address.
955 Exim incorrectly logged both deliveries with the TLS parameters (cipher
956 and peerdn, if requested) that were in fact used only for the second
959 PH/31 When doing a callout as part of verifying an address, Exim was not paying
960 attention to any local part prefix or suffix that was matched by the
961 router that accepted the address. It now behaves in the same way as it
962 does for delivery: the affixes are removed from the local part unless
963 rcpt_include_affixes is set on the transport.
965 PH/32 Add the sender address, as F=<...>, to the log line when logging a
966 timeout during the DATA phase of an incoming message.
968 PH/33 Sieve envelope tests were broken for match types other than :is. I have
969 applied a patch sanctioned by the Sieve maintainer.
971 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
972 the uid or gid is negative. A case of a negative gid caused this to be
973 noticed. The fix allows for either to be negative.
975 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
976 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
979 PH/36 The change PH/12 above was broken. Fixed it.
981 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
982 the grounds that routing the same address twice would always produce the
983 same answer. This might have been true once, but it is certainly no
984 longer true now. Routing a child address may depend on the previous
985 routing that produced that child. Some complicated redirection strategies
986 went wrong when messages had multiple recipients, and made Exim's
987 behaviour dependent on the order in which the addresses were given.
989 I have moved the duplicate checking until after the routing is complete.
990 Exim scans the addresses that are assigned to local and remote
991 transports, and removes any duplicates. This means that more work will be
992 done, as duplicates will always all be routed, but duplicates are
993 presumably rare, so I don't expect this is of any significance.
995 For deliveries to pipes, files, and autoreplies, the duplicate checking
996 still happens during the routing process, since they are not going to be
999 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1000 It corrects a timeout issue with spamd. This is Ian's comment: "The
1001 background is that sometimes spamd either never reads data from a
1002 connection it has accepted, or it never writes response data. The exiscan
1003 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1004 blindly assumes that writes won't block so it may never time out."
1006 PH/39 Allow G after quota size as well as K and M.
1008 PH/40 The value set for $authenticated_id in an authenticator may not contain
1009 binary zeroes or newlines because the value is written to log lines and
1010 to spool files. There was no check on this. Now the value is run through
1011 the string_printing() function so that such characters are converted to
1012 printable escape sequences.
1014 PH/41 $message_linecount is a new variable that contains the total number of
1015 lines in the message. Compare $body_linecount, which is the count for the
1018 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1019 bounce and delay warning messages, except in certain special cases, which
1022 (a) An SMTP error message from a remote host;
1023 (b) A message specified in a :fail: redirection;
1024 (c) A message specified in a "fail" command in a system filter;
1025 (d) A message specified in a FAIL return from the queryprogram router;
1026 (e) A message specified by the cannot_route_message router option.
1028 In these cases only, Exim does include the error details in bounce and
1029 warning messages. There are also a few cases where bland messages such
1030 as "unrouteable address" or "local delivery error" are given.
1032 PH/43 $value is now also set for the "else" part of a ${run expansion.
1034 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1035 being worked on, but at least Exim now implements the latest version to
1038 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1039 process to complete was treated as a delivery failure, a timeout while
1040 writing the message to the pipe was logged, but erroneously treated as a
1041 successful delivery. Such timeouts include transport filter timeouts. For
1042 consistency with the overall process timeout, these timeouts are now
1043 treated as errors, giving rise to delivery failures by default. However,
1044 there is now a new Boolean option for the pipe transport called
1045 timeout_defer, which, if set TRUE, converts the failures into defers for
1046 both kinds of timeout. A transport filter timeout is now identified in
1049 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1050 systems where "make" and "gmake" are different, calling "gmake" at top
1051 level broke things. I've arranged for the value of $(MAKE) to be passed
1052 from the Makefile to this script so that it can call the same version of
1056 A note about Exim versions 4.44 and 4.50
1057 ----------------------------------------
1059 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1060 changes of various kinds. As a consequence, a big documentation update was
1061 needed. This delayed the release for rather longer than seemed good, especially
1062 in the light of a couple of (minor) security issues. Therefore, the changes
1063 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1064 release. So 4.44 and 4.50 are in effect two different branches that both start
1067 I have left the 4.50 change log unchanged; it contains all the changes since
1068 4.43. The change log for 4.44 is below; many of its items are identical to
1069 those for 4.50. This seems to be the most sensible way to preserve the
1070 historical information.
1076 1. Minor wording change to the doc/README.SIEVE file.
1078 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1079 computation of the current number of files was incorrect.
1081 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1082 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1085 4. Give more explanation in the error message when the command for a transport
1086 filter fails to execute.
1088 5. There are several places where Exim runs a non-Exim command in a
1089 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1090 was being done only for the command run by the queryprogram router. It is
1091 now done for all such subprocesses. The other cases are: ${run, transport
1092 filters, and the commands run by the lmtp and pipe transports.
1094 6. Added CONFIGURE_GROUP build-time option.
1096 7. Some older OS have a limit of 256 on the maximum number of file
1097 descriptors. Exim was using setrlimit() to set 1000 as a large value
1098 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1099 systems. I've change it so that if it can't get 1000, it tries for 256.
1101 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1102 was an oversight, and furthermore, ever since the addition of extra
1103 controls (e.g. 4.43/32), the checks on when to allow different forms of
1104 "control" were broken. There should now be diagnostics for all cases when a
1105 control that does not make sense is encountered.
1107 9. Added the /retain_sender option to "control=submission".
1109 10. $recipients is now available in the predata ACL (oversight).
1111 11. Tidy the search cache before the fork to do a delivery from a message
1112 received from the command line. Otherwise the child will trigger a lookup
1113 failure and thereby defer the delivery if it tries to use (for example) a
1114 cached ldap connection that the parent has called unbind on.
1116 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1117 of $address_data from the recipient verification was clobbered by the
1118 sender verification.
1120 13. The value of address_data from a sender verification is now available in
1121 $sender_address_data in subsequent conditions in the ACL statement.
1123 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1125 15. Added a new option "connect=<time>" to callout options, to set a different
1128 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1129 was its contents. (It was OK if the option was not defined at all.)
1131 17. A "Completed" log line is now written for messages that are removed from
1132 the spool by the -Mrm option.
1134 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1135 information about exactly what failed.
1137 19. Added -dd to debug only the daemon process.
1139 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1140 handles timeouts, both on the server side and network timeouts. Renamed the
1141 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1143 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1144 It is now set to "smtps".
1146 22. $host_address is now set to the target address during the checking of
1147 ignore_target_hosts.
1149 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1150 being passed; this would have caused $sender_host_name to have been used if
1151 matching the list had actually called for a host name (not very likely,
1152 since this list is usually IP addresses). A host name is now passed as
1155 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1156 code that specifies a non-restarting handler (typically sigaction() in
1157 modern systems) in an attempt to fix a rare and obscure crash bug.
1159 25. Narrowed the window for a race in the daemon that could cause it to ignore
1160 SIGCHLD signals. This is not a major problem, because they are used only to
1161 wake it up if nothing else does.
1163 26. A malformed maildirsize file could cause Exim to calculate negative values
1164 for the mailbox size or file count. Odd effects could occur as a result.
1165 The maildirsize information is now recalculated if the size or filecount
1168 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1169 support for a long time. Removed HAVE_SYS_VFS_H.
1171 28. Installed the latest version of exipick from John Jetmore.
1173 29. In an address list, if the pattern was not a regular expression, an empty
1174 subject address (from a bounce message) matched only if the pattern was an
1175 empty string. Non-empty patterns were not even tested. This was the wrong
1176 because it is perfectly reasonable to use an empty address as part of a
1177 database query. An empty address is now tested by patterns that are
1178 lookups. However, all the other forms of pattern expect the subject to
1179 contain a local part and a domain, and therefore, for them, an empty
1180 address still always fails if the pattern is not itself empty.
1182 30. Exim went into a mad DNS loop when attempting to do a callout where the
1183 host was specified on an smtp transport, and looking it up yielded more
1184 than one IP address.
1186 31. Re-factored the code for checking spool and log partition space into a
1187 function that finds that data and another that does the check. The former
1188 is then used to implement four new variables: $spool_space, $log_space,
1189 $spool_inodes, and $log_inodes.
1191 32. The RFC2047 encoding function was originally intended for short strings
1192 such as real names; it was not keeping to the 75-character limit for
1193 encoded words that the RFC imposes. It now respects the limit, and
1194 generates multiple encoded words if necessary. To be on the safe side, I
1195 have increased the buffer size for the ${rfc2047: expansion operator from
1198 33. It is now permitted to omit both strings after an "if" condition; if the
1199 condition is true, the result is "true". As before, when the second string
1200 is omitted, a false condition yields an empty string. This makes it less
1201 cumbersome to write custom ACL and router conditions.
1203 34. Failure to deliver a bounce message always caused it to be frozen, even if
1204 there was an errors_to setting on the router. The errors_to setting is now
1207 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1208 canonical form (fully expanded) before being placed in
1209 $sender_host_address.
1211 36. The table in the code that translates DNS record types into text (T_A to
1212 "A" for instance) was missing entries for NS and CNAME. It is just possible
1213 that this could have caused confusion if both these types were looked up
1214 for the same domain, because the text type is used as part of Exim's
1215 per-process caching. But the chance of anyone hitting this buglet seems
1218 37. The dnsdb lookup has been extended in a number of ways.
1220 (1) There is a new type, "zns", which walks up the domain tree until it
1221 finds some nameserver records. It should be used with care.
1223 (2) There is a new type, "mxh", which is like "mx" except that it returns
1224 just the host names, not the priorities.
1226 (3) It is now possible to give a list of domains (or IP addresses) to be
1227 looked up. The behaviour when one of the lookups defers can be
1228 controlled by a keyword.
1230 (4) It is now possible to specify the separator character for use when
1231 multiple records are returned.
1233 38. The dnslists ACL condition has been extended: it is now possible to supply
1234 a list of IP addresses and/or domains to be looked up in a particular DNS
1237 39. Added log_selector=+queue_time_overall.
1239 40. When running the queue in the test harness, wait just a tad after forking a
1240 delivery process, to get repeatability of debugging output.
1242 41. Include certificate and key file names in error message when GnuTLS fails
1243 to set them up, because the GnuTLS error message doesn't include the name
1244 of the failing file when there is a problem reading it.
1246 42. Allow both -bf and -bF in the same test run.
1248 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1250 44. The "Exiscan patch" is now merged into the mainline Exim source.
1252 45. Sometimes the final signoff response after QUIT could fail to get
1253 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1254 < 0 before doing a fflush(). This bug looks as though it goes back to the
1255 introduction of TLS in release 3.20, but "sometimes" must have been rare
1256 because the tests only now provoked it.
1258 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1259 (this can affect the format of dates).
1261 47. exim_tidydb, when checking for the continued existence of a message for
1262 which it has found a message-specific retry record, was not finding
1263 messages that were in split spool directories. Consequently, it was
1264 deleting retry records that should have stayed in existence.
1266 48. Steve fixed some bugs in eximstats.
1268 49. The SPA authentication driver was not abandoning authentication and moving
1269 on to the next authenticator when an expansion was forced to fail,
1270 contradicting the general specification for all authenticators. Instead it
1271 was generating a temporary error. It now behaves as specified.
1273 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1274 (the order specifies the preference for clients). The order is now AES256,
1275 AES128, 3DES, ARCFOUR128.
1277 51. Small patch to Sieve code - explicitly set From: when generating an
1280 52. Exim crashed if a remote delivery caused a very long error message to be
1281 recorded - for instance if somebody sent an entire SpamAssassin report back
1282 as a large number of 550 error lines. This bug was coincidentally fixed by
1283 increasing the size of one of Exim's internal buffers (big_buffer) that
1284 happened as part of the Exiscan merge. However, to be on the safe side, I
1285 have made the code more robust (and fixed the comments that describe what
1288 53. Now that there can be additional text after "Completed" in log lines (if
1289 the queue_time_overall log selector is set), a one-byte patch to exigrep
1290 was needed to allow it to recognize "Completed" as not the last thing in
1293 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1294 patch that reportedly fixes this has been added. I am not expert enough to
1295 create a test for it. This is what the patch creator wrote:
1297 "I found a little strange behaviour of ldap code when working with
1298 Windows 2003 AD Domain, where users was placed in more than one
1299 Organization Units. When I tried to give exim partial DN, the exit code
1300 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1301 But simultaneously result of request was absolutely normal ldap result,
1302 so I produce this patch..."
1304 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1305 REFERENCE, so I have modified the code to exclude the patch when that macro
1308 55. Some experimental protocols are using DNS PTR records for new purposes. The
1309 keys for these records are domain names, not reversed IP addresses. The
1310 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1311 leaves it alone. Component reversal etc. now happens only for IP addresses.
1314 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1316 57. Double the size of the debug message buffer (to 2048) so that more of very
1317 long debug lines gets shown.
1319 58. The exicyclog utility now does better if the number of log files to keep
1320 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1322 59. Two changes related to the smtp_active_hostname option:
1324 (1) $smtp_active_hostname is now available as a variable.
1325 (2) The default for smtp_banner uses $smtp_active_hostname instead
1326 of $primary_hostname.
1328 60. The host_aton() function is supposed to be passed a string that is known
1329 to be a valid IP address. However, in the case of IPv6 addresses, it was
1330 not checking this. This is a hostage to fortune. Exim now panics and dies
1331 if the condition is not met. A case was found where this could be provoked
1332 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1333 components; fortuitously, this particular loophole had already been fixed
1334 by change 4.50/55 above.
1336 If there are any other similar loopholes, the new check in host_aton()
1337 itself should stop them being exploited. The report I received stated that
1338 data on the command line could provoke the exploit when Exim was running as
1339 exim, but did not say which command line option was involved. All I could
1340 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1341 running as the user.
1344 61. There was a buffer overflow vulnerability in the SPA authentication code
1345 (which came originally from the Samba project). I have added a test to the
1346 spa_base64_to_bits() function which I hope fixes it.
1349 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1350 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1352 63. The daemon start-up calls getloadavg() while still root for those OS that
1353 need the first call to be done as root, but it missed one case: when
1354 deliver_queue_load_max is set with deliver_drop_privilege. This is
1355 necessary for the benefit of the queue runner, because there is no re-exec
1356 when deliver_drop_privilege is set.
1358 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1359 This has been fixed.
1361 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1362 was in use, was not putting the data itself into the right store pool;
1363 consequently, it could be overwritten for a subsequent message in the same
1364 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1367 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1369 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1370 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1371 the function were treating the return as a boolean value, which happened to
1372 work because 0=false and not-0=true, but is not correct code.
1374 68. The host_aton() function was not handling scoped IPv6 addresses (those
1375 with, for example, "%eth0" on the end) correctly.
1377 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1378 negated items (that is, ~something) in unsigned ints. Some compilers
1379 apparently mutter when there is no cast.
1381 70. If an address verification called from an ACL failed, and did not produce a
1382 user-specific message (i.e. there was only a "system" message), nothing was
1383 put in $acl_verify_message. In this situation, it now puts the system
1386 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1387 session; change 4.31/43 added the unwanted input to the log line - except
1388 that it did not do this in the start of session case. It now does.
1390 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1391 This is wrong and can cause the other end to generate a synchronization
1392 error if it is another Exim or anything else that does the synchronization
1393 check. A QUIT command is no longer sent after a timeout.
1395 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1396 during host lookups.
1398 74. The defer_ok option of callout verification was not working if it was used
1399 when verifying addresses in header lines, that is, for this case:
1401 verify = header_sender/callout=defer_ok
1403 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1404 those file descriptors could be used for SMTP connections. If anything
1405 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1406 could be sent to the SMTP client, causing chaos. The daemon now opens
1407 stdin, stdout, and stderr to /dev/null when it puts itself into the
1410 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1411 log by default. The user can override this with suitable Perl magic.
1413 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1414 the log message when discard triggers, was not working for the DATA ACL or
1415 for the non-SMTP ACL.
1417 78. Error message wording change in sieve.c.
1419 79. If smtp_accept_max_per_host was set, the number of connections could be
1420 restricted to fewer than expected, because the daemon was trying to set up
1421 a new connection before checking whether the processes handling previous
1422 connections had finished. The check for completed processes is now done
1423 earlier. On busy systems, this bug wouldn't be noticed because something
1424 else would have woken the daemon, and it would have reaped the completed
1427 80. If a message was submitted locally by a user whose login name contained one
1428 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1429 It caused a spool format error. I have fixed the spool reading code. A
1430 related problem was that the "from" clause in the Received: line became
1431 illegal because of the space(s). It is now covered by ${quote_local_part.
1433 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1436 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1438 83. Patch from Sieve maintainer to fix unterminated string problem in
1439 "vacation" handling.
1441 84. Some minor changes to the Linux configuration files to help with other
1442 OS variants using glibc.
1444 85. One more patch for Sieve to update vacation handling to latest spec.
1447 ----------------------------------------------------
1448 See the note above about the 4.44 and 4.50 releases.
1449 ----------------------------------------------------
1455 1. Change 4.43/35 introduced a bug that caused file counts to be
1456 incorrectly computed when quota_filecount was set in an appendfile
1459 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1460 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1463 3. Give more explanation in the error message when the command for a transport
1464 filter fails to execute.
1466 4. There are several places where Exim runs a non-Exim command in a
1467 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1468 was being done only for the command run by the queryprogram router. It is
1469 now done for all such subprocesses. The other cases are: ${run, transport
1470 filters, and the commands run by the lmtp and pipe transports.
1472 5. Some older OS have a limit of 256 on the maximum number of file
1473 descriptors. Exim was using setrlimit() to set 1000 as a large value
1474 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1475 systems. I've change it so that if it can't get 1000, it tries for 256.
1477 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1478 was an oversight, and furthermore, ever since the addition of extra
1479 controls (e.g. 4.43/32), the checks on when to allow different forms of
1480 "control" were broken. There should now be diagnostics for all cases when a
1481 control that does not make sense is encountered.
1483 7. $recipients is now available in the predata ACL (oversight).
1485 8. Tidy the search cache before the fork to do a delivery from a message
1486 received from the command line. Otherwise the child will trigger a lookup
1487 failure and thereby defer the delivery if it tries to use (for example) a
1488 cached ldap connection that the parent has called unbind on.
1490 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1491 of $address_data from the recipient verification was clobbered by the
1492 sender verification.
1494 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1495 was its contents. (It was OK if the option was not defined at all.)
1497 11. A "Completed" log line is now written for messages that are removed from
1498 the spool by the -Mrm option.
1500 12. $host_address is now set to the target address during the checking of
1501 ignore_target_hosts.
1503 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1504 being passed; this would have caused $sender_host_name to have been used if
1505 matching the list had actually called for a host name (not very likely,
1506 since this list is usually IP addresses). A host name is now passed as
1509 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1510 code that specifies a non-restarting handler (typically sigaction() in
1511 modern systems) in an attempt to fix a rare and obscure crash bug.
1513 15. Narrowed the window for a race in the daemon that could cause it to ignore
1514 SIGCHLD signals. This is not a major problem, because they are used only to
1515 wake it up if nothing else does.
1517 16. A malformed maildirsize file could cause Exim to calculate negative values
1518 for the mailbox size or file count. Odd effects could occur as a result.
1519 The maildirsize information is now recalculated if the size or filecount
1522 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1523 support for a long time. Removed HAVE_SYS_VFS_H.
1525 18. Updated exipick to current release from John Jetmore.
1527 19. Allow an empty sender to be matched against a lookup in an address list.
1528 Previously the only cases considered were a regular expression, or an
1531 20. Exim went into a mad DNS lookup loop when doing a callout where the
1532 host was specified on the transport, if the DNS lookup yielded more than
1535 21. The RFC2047 encoding function was originally intended for short strings
1536 such as real names; it was not keeping to the 75-character limit for
1537 encoded words that the RFC imposes. It now respects the limit, and
1538 generates multiple encoded words if necessary. To be on the safe side, I
1539 have increased the buffer size for the ${rfc2047: expansion operator from
1542 22. Failure to deliver a bounce message always caused it to be frozen, even if
1543 there was an errors_to setting on the router. The errors_to setting is now
1546 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1547 canonical form (fully expanded) before being placed in
1548 $sender_host_address.
1550 24. Updated eximstats to version 1.33
1552 25. Include certificate and key file names in error message when GnuTLS fails
1553 to set them up, because the GnuTLS error message doesn't include the name
1554 of the failing file when there is a problem reading it.
1556 26. Expand error message when OpenSSL has problems setting up cert/key files.
1559 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1560 (this can affect the format of dates).
1562 28. exim_tidydb, when checking for the continued existence of a message for
1563 which it has found a message-specific retry record, was not finding
1564 messages that were in split spool directories. Consequently, it was
1565 deleting retry records that should have stayed in existence.
1567 29. eximstats updated to version 1.35
1568 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1569 1.35 - bugfix such that pie charts by volume are generated correctly
1571 30. The SPA authentication driver was not abandoning authentication and moving
1572 on to the next authenticator when an expansion was forced to fail,
1573 contradicting the general specification for all authenticators. Instead it
1574 was generating a temporary error. It now behaves as specified.
1576 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1577 (the order specifies the preference for clients). The order is now AES256,
1578 AES128, 3DES, ARCFOUR128.
1580 31. Small patch to Sieve code - explicitly set From: when generating an
1583 32. Exim crashed if a remote delivery caused a very long error message to be
1584 recorded - for instance if somebody sent an entire SpamAssassin report back
1585 as a large number of 550 error lines. This bug was coincidentally fixed by
1586 increasing the size of one of Exim's internal buffers (big_buffer) that
1587 happened as part of the Exiscan merge. However, to be on the safe side, I
1588 have made the code more robust (and fixed the comments that describe what
1591 33. Some experimental protocols are using DNS PTR records for new purposes. The
1592 keys for these records are domain names, not reversed IP addresses. The
1593 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1594 leaves it alone. Component reversal etc. now happens only for IP addresses.
1597 34. The host_aton() function is supposed to be passed a string that is known
1598 to be a valid IP address. However, in the case of IPv6 addresses, it was
1599 not checking this. This is a hostage to fortune. Exim now panics and dies
1600 if the condition is not met. A case was found where this could be provoked
1601 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1602 components; fortuitously, this particular loophole had already been fixed
1603 by change 4.50/55 or 4.44/33 above.
1605 If there are any other similar loopholes, the new check in host_aton()
1606 itself should stop them being exploited. The report I received stated that
1607 data on the command line could provoke the exploit when Exim was running as
1608 exim, but did not say which command line option was involved. All I could
1609 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1610 running as the user.
1613 35. There was a buffer overflow vulnerability in the SPA authentication code
1614 (which came originally from the Samba project). I have added a test to the
1615 spa_base64_to_bits() function which I hope fixes it.
1618 36. The daemon start-up calls getloadavg() while still root for those OS that
1619 need the first call to be done as root, but it missed one case: when
1620 deliver_queue_load_max is set with deliver_drop_privilege. This is
1621 necessary for the benefit of the queue runner, because there is no re-exec
1622 when deliver_drop_privilege is set.
1624 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1625 was in use, was not putting the data itself into the right store pool;
1626 consequently, it could be overwritten for a subsequent message in the same
1627 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1630 38. Sometimes the final signoff response after QUIT could fail to get
1631 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1632 < 0 before doing a fflush(). This bug looks as though it goes back to the
1633 introduction of TLS in release 3.20, but "sometimes" must have been rare
1634 because the tests only now provoked it.
1640 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1641 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1642 Now it returns an integer. A number of calls were still expecting a T/F
1643 return. Fortuitously, in all cases, the tests worked in OK situations,
1644 which is the norm. However, things would have gone wrong on any write
1645 failures on the smtp file descriptor. This function is used when sending
1646 messages over SMTP and also when doing verify callouts.
1648 2. When Exim is called to do synchronous delivery of a locally submitted
1649 message (the -odf or -odi options), it no longer closes stderr before doing
1652 3. Implemented the mua_wrapper option.
1654 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1656 5. Implemented the functions header_remove(), header_testname(),
1657 header_add_at_position(), and receive_remove_recipient(), and exported them
1660 6. If an ACL "warn" statement specified the addition of headers, Exim already
1661 inserted X-ACL-Warn: at the start if there was no header name. However, it
1662 was not making this test for the second and subsequent header lines if
1663 there were newlines in the string. This meant that an invalid header could
1664 be inserted if Exim was badly configured.
1666 7. Allow an ACL "warn" statement to add header lines at the start or after all
1667 the Received: headers, as well as at the end.
1669 8. Added the rcpt_4xx retry error code.
1671 9. Added postmaster_mailfrom=xxx to callout verification option.
1673 10. Added mailfrom=xxxx to the callout verification option, for verify=
1676 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1677 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1679 12. Inserted some casts to stop certain compilers warning when using pointer
1680 differences as field lengths or precisions in printf-type calls (mostly
1681 affecting debugging statements).
1683 13. Added optional readline() support for -be (dynamically loaded).
1685 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1686 same clock tick as a message's arrival, so that its received time was the
1687 same as the "first fail" time on the retry record, and that message
1688 remained on the queue past the ultimate address timeout, every queue runner
1689 would try a delivery (because it was past the ultimate address timeout) but
1690 after another failure, the ultimate address timeout, which should have then
1691 bounced the address, did not kick in. This was a "< instead of <=" error;
1692 in most cases the first failure would have been in the next clock tick
1693 after the received time, and all would be well.
1695 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1696 being recognized when the domain list was tested by the match_domain
1697 condition in an expansion string.
1699 16. Added the ${str2b64: operator.
1701 17. Exim was always calling setrlimit() to set a large limit for the number of
1702 processes, without checking whether the existing limit was already
1703 adequate. (It did check for the limit on file descriptors.) Furthermore,
1704 errors from getrlimit() and setrlimit() were being ignored. Now they are
1705 logged to the main and panic logs, but Exim does carry on, to try to do its
1706 job under whatever limits there are.
1708 18. Imported PCRE 5.0.
1710 19. Trivial typo in log message " temporarily refused connection" (the leading
1713 20. If the log selector return_path_on_delivery was set and an address was
1714 redirected to /dev/null, the delivery process crashed because it assumed
1715 that a return path would always be set for a "successful" delivery. In this
1716 case, the whole delivery is bypassed as an optimization, and therefore no
1719 21. Internal re-arrangement: the function for sending a challenge and reading
1720 a response while authentication was assuming a zero-terminated challenge
1721 string. It's now changed to take a pointer and a length, to allow for
1722 binary data in such strings.
1724 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1726 23. Exim was not respecting finduser_retries when seeking the login of the
1727 uid under which it was called; it was always trying 10 times. (The default
1728 setting of finduser_retries is zero.) Also, it was sleeping after the final
1729 failure, which is pointless.
1731 24. Implemented tls_on_connect_ports.
1733 25. Implemented acl_smtp_predata.
1735 26. If the domain in control=submission is set empty, Exim assumes that the
1736 authenticated id is a complete email address when it generates From: or
1737 Sender: header lines.
1739 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1740 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1741 chown and chgrp in /bin and hostname in /usr/bin.
1743 28. Exim was keeping the "process log" file open after each use, just as it
1744 does for the main log. This opens the possibility of it remaining open for
1745 long periods when the USR1 signal hits a daemon. Occasional processlog
1746 errors were reported, that could have been caused by this. Anyway, it seems
1747 much more sensible not to leave this file open at all, so that is what now
1750 29. The long-running daemon process does not normally write to the log once it
1751 has entered its main loop, and it closes the log before doing so. This is
1752 so that log files can straightforwardly be renamed and moved. However,
1753 there are a couple of unusual error situations where the daemon does write
1754 log entries, and I had neglected to close the log afterwards.
1756 30. The text of an SMTP error response that was received during a remote
1757 delivery was being truncated at 512 bytes. This is too short for some of
1758 the long messages that one sometimes sees. I've increased the limit to
1761 31. It is now possible to make retry rules that apply only when a message has a
1762 specific sender, in particular, an empty sender.
1764 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1765 it possible to be selective about when SMTP synchronization is enforced.
1767 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1769 32. Implemented hosts_connection_nolog.
1771 33. Added an ACL for QUIT.
1773 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1776 35. Added mailbox_size and mailbox_filecount to appendfile.
1778 36. Added control = no_multiline_responses to ACLs.
1780 37. There was a bug in the logic of the code that waits for the clock to tick
1781 in the case where the clock went backwards by a substantial amount such
1782 that the microsecond fraction of "now" was more than the microsecond
1783 fraction of "then" (but the whole seconds number was less).
1785 38. Added support for the libradius Radius client library this is found on
1786 FreeBSD (previously only the radiusclient library was supported).
1792 1. When certain lookups returned multiple values in the form name=value, the
1793 quoting of the values was not always being done properly. Specifically:
1794 (a) If the value started with a double quote, but contained no whitespace,
1796 (b) If the value contained whitespace other than a space character (i.e.
1797 tabs or newlines or carriage returns) it was not quoted.
1798 This fix has been applied to the mysql and pgsql lookups by writing a
1799 separate quoting function and calling it from the lookup code. The fix
1800 should probably also be applied to nisplus, ibase and oracle lookups, but
1801 since I cannot test any of those, I have not disturbed their existing code.
1803 2. A hit in the callout cache for a specific address caused a log line with no
1804 reason for rejecting RCPT. Now it says "Previous (cached) callout
1805 verification failure".
1807 3. There was an off-by-one bug in the queryprogram router. An over-long
1808 return line was truncated at 256 instead of 255 characters, thereby
1809 overflowing its buffer with the terminating zero. As well as fixing this, I
1810 have increased the buffer size to 1024 (and made a note to document this).
1812 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1813 when Exim is waiting for an SMTP response from a remote server, Exim
1814 restarts its select() call on the socket, thereby resetting its timeout.
1815 This is not a problem when such interrupts are rare. Somebody set up a cron
1816 job to run exiwhat every 2 minutes, which is less than the normal select()
1817 timeout (5 or 10 minutes). This meant that the select() timeout never
1818 kicked in because it was always reset. I have fixed this by comparing the
1819 time when an interrupt arrives with the time at the start of the first call
1820 to select(). If more time than the timeout has elapsed, the interrupt is
1821 treated as a timeout.
1823 5. Some internal re-factoring in preparation for the addition of Sieve
1824 extensions (by MH). In particular, the "personal" test is moved to a
1825 separate function, and given an option for scanning Cc: and Bcc: (which is
1826 not set for Exim filters).
1828 6. When Exim created an email address using the login of the caller as the
1829 local part (e.g. when creating a From: or Sender: header line), it was not
1830 quoting the local part when it contained special characters such as @.
1832 7. Installed new OpenBSD configuration files.
1834 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1835 try to make them clearer.
1837 9. Callout options, other than the timeout value, were being ignored when
1838 verifying sender addresses in header lines. For example, when using
1840 verify = header_sender/callout=no_cache
1842 the cache was (incorrectly) being used.
1844 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1845 only the Cygwin environment.
1847 11. When return_path_on_delivery was set as a log selector, if different remote
1848 addresses in the same message used different return paths and parallel
1849 remote delivery occurred, the wrong values would sometimes be logged.
1850 (Whenever a remote delivery process finished, the return path value from
1851 the most recently started remote delivery process was logged.)
1853 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1854 lines when AUTH and/or TLS are in use. This is the "received protocol"
1855 field. Exim used to use "asmtp" for authenticated SMTP, without any
1856 indication (in the protocol name) for TLS use. Now it follows the RFC and
1857 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1858 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1859 names appear in log lines as well as in Received: header lines.
1861 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1862 extensions, and comparison tests, and to fix some bugs.
1864 14. Changes to the "personal" filter test:
1866 (1) The test was buggy in that it was just doing the equivalent of
1867 "contains" tests on header lines. For example, if a user's address was
1868 anne@some.where, the "personal" test would incorrectly be true for
1870 To: susanne@some.where
1872 This test is now done by extracting each address from the header in turn,
1873 and checking the entire address. Other tests that are part of "personal"
1874 are now done using regular expressions (for example, to check local parts
1875 of addresses in From: header lines).
1877 (2) The list of non-personal local parts in From: addresses has been
1878 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1879 taken from the Sieve specification recommendations.
1881 (3) If the message contains any header line starting with "List-" it is
1882 treated as non-personal.
1884 (4) The test for "circular" in the Subject: header line has been removed
1885 because it now seems ill-conceived.
1887 15. Minor typos in src/EDITME comments corrected.
1889 16. Installed latest exipick from John Jetmore.
1891 17. If headers_add on a router specified a text string that was too long for
1892 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1893 of string_sprintf() is now avoided.
1895 18. $message_body_size was not set (it was always zero) when running the DATA
1896 ACL and the local_scan() function.
1898 19. For the "mail" command in an Exim filter, no default was being set for
1899 the once_repeat time, causing a random time value to be used if "once" was
1900 specified. (If the value happened to be <= 0, no repeat happened.) The
1901 default is now 0s, meaning "never repeat". The "vacation" command was OK
1902 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1903 (I found it when inspecting the code).
1905 20. There is now an overall timeout for performing a callout verification. It
1906 defaults to 4 times the callout timeout, which applies to individual SMTP
1907 commands during the callout. The overall timeout applies when there is more
1908 than one host that can be tried. The timeout is checked before trying the
1909 next host. This prevents very long delays if there are a large number of
1910 hosts and all are timing out (e.g. when the network connections are timing
1911 out). The value of the overall timeout can be changed by specifying an
1912 additional sub-option for "callout", called "maxwait". For example:
1914 verify = sender/callout=5s,maxwait=20s
1916 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1917 to the end before writing, but this should make it even safer).
1919 22. Exim was forgetting that it had advertised PIPELINING for the second and
1920 subsequent messages on an SMTP connection. It was also not resetting its
1921 memory on STARTTLS and an internal HELO.
1923 23. When Exim logs an SMTP synchronization error within a session, it now
1924 records whether PIPELINING has been advertised or not.
1926 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1927 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1928 rather than long int.
1930 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1932 26. Added the never_mail option to autoreply.
1938 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1939 crash if the getsockname() call failed; this can happen if a connection is
1940 closed very soon after it is established. The problem was simply in the
1941 order in which certain operations were done, causing Exim to try to write
1942 to the SMTP stream before it had set up the file descriptor. The bug has
1943 been fixed by making things happen in the correct order.
1949 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1950 before the connection was closed, thus losing the rejection response.
1952 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1953 some early Solaris releases, but causes trouble in current releases where
1954 socklen_t is defined.
1956 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1959 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1962 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1963 that releasing the top of it at the end releases what was used for sub-
1964 expansions (unless the block got too big). However, discard this block if
1965 the first thing is a variable or header, so that we can use its block when
1966 it is dynamic (useful for very large $message_headers, for example).
1968 6. Lookups now cache *every* query, not just the most recent. A new, separate
1969 store pool is used for this. It can be recovered when all lookup caches are
1970 flushed. Lookups now release memory at the end of their result strings.
1971 This has involved some general refactoring of the lookup sources.
1973 7. Some code has been added to the store_xxx() functions to reduce the amount
1974 of flapping under certain conditions.
1976 8. log_incoming_interface used to affect only the <= reception log lines. Now
1977 it causes the local interface and port to be added to several more SMTP log
1978 lines, for example "SMTP connection from", and rejection lines.
1980 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1982 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1984 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1985 could be overwritten at the end of the current message (or the start of a
1986 new message if it was set in a HELO ACL). The value is now preserved for
1987 the duration of the SMTP connection.
1989 12. If a transport had a headers_rewrite setting, and a matching header line
1990 contained an unqualified address, that address was qualified, even if it
1991 did not match any rewriting rules. The underlying bug was that the values
1992 of the flags that permit the existence of unqualified sender and recipient
1993 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1994 non-local messages, and by -bnq for local messages) were not being
1995 preserved with the message after it was received.
1997 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1998 "next input=" as part of the text comprising the host identity instead of
1999 the correct text. The code was using the same buffer for two different
2000 strings. However, depending on which order the printing function evaluated
2001 its arguments, the bug did not always show up. Under Linux, for example, my
2002 test suite worked just fine.
2004 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2005 doesn't work with some older versions of Perl. It has been changed to "my",
2006 which in any case is probably the better facility to use.
2008 15. A really picky compiler found some instances of statements for creating
2009 error messages that either had too many or two few arguments for the format
2012 16. The size of the buffer for calls to the DNS resolver has been increased
2013 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2014 for addresses that have a lot of PTR records. This alleviates a problem; it
2015 does not fully solve it.
2017 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2018 buffer now truncates the list and logs the incident, which is the same
2019 action as happens when Exim is looking up a host name and its aliases.
2020 Previously in this situation something unpredictable would happen;
2021 sometimes it was "internal error: store_reset failed".
2023 18. If a server dropped the connection unexpectedly when an Exim client was
2024 using GnuTLS and trying to read a response, the client delivery process
2025 crashed while trying to generate an error log message.
2027 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2028 string, for example:
2030 warn message = H1: something\nH2: something
2032 the text was added as a single header line from Exim's point of view
2033 though it ended up OK in the delivered message. However, searching for the
2034 second and subsequent header lines using $h_h2: did not work. This has been
2035 fixed. Similarly, if a system filter added multiple headers in this way,
2036 the routers could not see them.
2038 20. Expanded the error message when iplsearch is called with an invalid key to
2039 suggest using net-iplsearch in a host list.
2041 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2042 ACLs are no longer actually imposed (and a message to that effect is
2045 22. If a "gecos" field in a passwd entry contained escaped characters, in
2046 particular, if it contained a \" sequence, Exim got it wrong when building
2047 a From: or a Sender: header from that name. A second bug also caused
2048 incorrect handling when an unquoted " was present following a character
2049 that needed quoting.
2051 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2052 was not being matched caselessly.
2054 24. Arranged for all hyphens in the exim.8 source to be escaped with
2057 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2058 properly. Recipient callout cache records were still being keyed to include
2059 the sender, even when use_sender was set false. This led to far more
2060 callouts that were necessary. The sender is no longer included in the key
2061 when use_sender is false.
2063 26. Added "control = submission" modifier to ACLs.
2065 27. Added the ${base62d: operator to decode base 62 numbers.
2067 28. dnsdb lookups can now access SRV records.
2069 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2070 the configuration file.
2072 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2073 (-v) mode. This makes the output for a verbose queue run more intelligible.
2075 31. Added a use_postmaster feature to recipient callouts.
2077 32. Added the $body_zerocount variable, containing the number of binary zero
2078 bytes in the message body.
2080 33. The time of last modification of the "new" subdirectory is now used as the
2081 "mailbox time last read" when there is a quota error for a maildir
2084 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2086 35. Added +ignore_unknown as a special item in host lists.
2088 36. Code for decoding IPv6 addresses in host lists is now included, even if
2089 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2090 address was recognized as an IP address, but was then not correctly decoded
2091 into binary, causing unexpected and incorrect effects when compared with
2098 1. Very minor rewording of debugging text in manualroute to say "list of
2099 hosts" instead of "hostlist".
2101 2. If verify=header_syntax was set, and a header line with an unqualified
2102 address (no domain) and a large number of spaces between the end of the
2103 name and the colon was received, the reception process suffered a buffer
2104 overflow, and (when I tested it) crashed. This was caused by some obsolete
2105 code that should have been removed. The fix is to remove it!
2107 3. When running in the test harness, delay a bit after writing a bounce
2108 message to get a bit more predictability in the log output.
2110 4. Added a call to search_tidyup() just before forking a reception process. In
2111 theory, someone could use a lookup in the expansion of smtp_accept_max_
2112 per_host which, without the tidyup, could leave open a database connection.
2114 5. Added the variables $recipient_data and $sender_data which get set from a
2115 lookup success in an ACL "recipients" or "senders" condition, or a router
2116 "senders" option, similar to $domain_data and $local_part_data.
2118 6. Moved the writing of debug_print from before to after the "senders" test
2121 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2122 problems for message scanning, either using a data ACL, or using
2123 local_scan() because the Received: header was not generated till after they
2124 were called (in order to set the time as the time of reception completion).
2125 I have revised the way this works. The header is now generated after the
2126 body is received, but before the ACL or local_scan() are called. After they
2127 are run, the timestamp in the header is updated.
2133 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2134 before starting a queue runner without re-exec. This happened only when
2135 deliver_drop_privilege was set or when the Exim user was set to root. The
2136 effect of the bug was that timeouts during subsequent deliveries caused
2137 crashes instead of being properly handled. The handler is now left at its
2138 default (and expected) setting.
2140 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2141 message, again when deliver_drop_privilege is set or Exim is run as root.
2142 The bug described in (1) was not present in this case, but the tidying up
2143 of the other signals was missing. I have made the two cases consistent.
2145 3. The ignore_target_hosts setting on a manualroute router was being ignored
2146 for hosts that were looked up using the /MX notation.
2148 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2151 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2152 operated on the sender address. After changing the $sender_address to <>
2153 for the sender address verify, Exim was re-instated it as the original
2154 (before rewriting) address, but remembering that it had rewritten it, so it
2155 wasn't rewriting it again. This bug also had the effect of breaking the
2156 sender address verification caching when the sender address was rewritten.
2158 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2159 This has been changed so that if the ip literal address matches
2160 ignore_target_hosts, the router declines.
2162 7. Added expansion conditions match_domain, match_address, and match_local_
2163 part (NOT match_host).
2165 8. The placeholder for the Received: header didn't have a length field set.
2167 9. Added code to Exim itself and to exim_lock to test for a specific race
2168 condition that could lead to file corruption when using MBX delivery. The
2169 issue is with the lockfile that is created in /tmp. If this file is removed
2170 after a process has opened it but before that process has acquired a lock,
2171 there is the potential for a second process to recreate the file and also
2172 acquire a lock. This could lead to two Exim processes writing to the file
2173 at the same time. The added code performs the same test as UW imapd; it
2174 checks after acquiring the lock that its file descriptor still refers to
2175 the same named file.
2177 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2178 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2181 11. Added the smtp_active_hostname option. If used, this will typically be made
2182 to depend on the incoming interface address. Because $interface_address is
2183 not set up until the daemon has forked a reception process, error responses
2184 that can happen earlier (such as "too many connections") no longer contain
2187 12. If an expansion in a condition on a "warn" statement fails because a lookup
2188 defers, the "warn" statement is abandoned, and the next ACL statement is
2189 processed. Previously this caused the whole ACL to be aborted.
2191 13. Added the iplsearch lookup type.
2193 14. Added ident_timeout as a log selector.
2195 15. Added tls_certificate_verified as a log selector.
2197 16. Added a global option tls_require_ciphers (compare the smtp transport
2198 option of the same name). This controls incoming TLS connections.
2200 17. I finally figured out how to make tls_require_ciphers do a similar thing
2201 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2202 before starting the TLS session.
2204 18. Tabs are now shown as \t in -bP output.
2206 19. If the log selector return_path_on_delivery was set, Exim crashed when
2207 bouncing a message because it had too many Received: header lines.
2209 20. If two routers both had headers_remove settings, and the first one included
2210 a superfluous trailing colon, the final name in the first list and the
2211 first name in the second list were incorrectly joined into one item (with a
2212 colon in the middle).
2218 1. Added -C and -D options to the exinext utility, mainly to make it easier
2219 to include in the automated testing, but these could be helpful when
2220 multiple configurations are in use.
2222 2. The exinext utility was not formatting the output nicely when there was
2223 an alternate port involved in the retry record key, nor when there was a
2224 message id as well (for retries that were specific to a specific message
2225 and a specific host). It was also confused by IPv6 addresses, because of
2226 the additional colons they contain. I have fixed the IPv4 problem, and
2227 patched it up to do a reasonable job for IPv6.
2229 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2230 delivery, the log line now contains "pipelined" if PIPELINING was used.
2232 4. An SMTP transport process used to panic and die if the bind() call to set
2233 an explicit outgoing interface failed. This has been changed; it is now
2234 treated in the same way as a connect() failure.
2236 5. A reference to $sender_host_name in the part of a conditional expansion
2237 that was being skipped was still causing a DNS lookup. This no longer
2240 6. The def: expansion condition was not recognizing references to header lines
2241 that used bh_ and bheader_.
2243 7. Added the _cache feature to named lists.
2245 8. The code for checking quota_filecount in the appendfile transport was
2246 allowing one more file than it should have been.
2248 9. For compatibility with Sendmail, the command line option
2256 and sets the incoming protocol and host name (for trusted callers). The
2257 host name and its colon can be omitted when only the protocol is to be set.
2258 Note the Exim already has two private options, -pd and -ps, that refer to
2259 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2260 "s", but I don't think that's a major issue.
2262 10. A number of refactoring changes to the code, none of which should affect
2265 (a) The number of logging options was getting close to filling up the
2266 32-bit word that was used as a bit map. I have split them into two classes:
2267 those that are passed in the argument to log_write(), and those that are
2268 only ever tested independently outside of that function. These are now in
2269 separate 32-bit words, so there is plenty of room for expansion again.
2270 There is no change in the user interface or the logging behaviour.
2272 (b) When building, for example, log lines, the code previously used a
2273 macro that called string_cat() twice, in order to add two strings. This is
2274 not really sufficiently general. Furthermore, there was one instance where
2275 it was actually wrong because one of the argument was used twice, and in
2276 one call a function was used. (As it happened, calling the function twice
2277 did not affect the overall behaviour.) The macro has been replaced by a
2278 function that can join an arbitrary number of extra strings onto a growing
2281 (c) The code for expansion conditions now uses a table and a binary chop
2282 instead of a serial search (which was left over from when there were very
2283 few conditions). Also, it now recognizes conditions like "pam" even when
2284 the relevant support is not compiled in: a suitably worded error message is
2285 given if an attempt is made to use such a condition.
2287 11. Added ${time_interval:xxxxx}.
2289 12. A bug was causing one of the ddress fields not to be passed back correctly
2290 from remote delivery subprocesses. The field in question was not being
2291 subsequently used, so this caused to problems in practice.
2293 13. Added new log selectors queue_time and deliver_time.
2295 14. Might have fixed a bug in maildirsizefile handling that threw up
2296 "unexpected character" debug warnings, and recalculated the data
2297 unnecessarily. In any case, I expanded the warning message to give more
2300 15. Added the message "Restricted characters in address" to the statements in
2301 the default ACL that block characters like @ and % in local parts.
2303 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2304 Three changes have been made:
2306 (a) There was a serious bug; a negative response to MAIL caused the whole
2307 recipient domain to be cached as invalid, thereby blocking all messages
2308 to all local parts at the same domain, from all senders. This bug has
2309 been fixed. The domain is no longer cached after a negative response to
2310 MAIL if the sender used is not empty.
2312 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2315 (c) A new callout option, "use_sender" has been added for people who want
2316 the modified behaviour.
2322 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2325 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2326 indeed breaks things for older releases.
2328 3. Added additional logging to the case where there is a problem reading data
2329 from a filter that is running in a subprocess using a pipe, in order to
2330 try to track down a specific problem.
2332 4. Testing facility fudge: when running in the test harness and attempting
2333 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2334 getting "No route to host". Convert this to a timeout.
2336 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2339 6. Some OS don't have socklen_t but use size_t instead. This affects the
2340 fifth argument of getsockopt() amongst other things. This is now
2341 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2342 can be set for individual OS. I have set it for SunOS5, OSF1, and
2343 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2344 some earlier ones do not.
2346 7. Change 4.30/15 was not doing the test caselessly.
2348 8. The standard form for an IPv6 address literal was being rejected by address
2349 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2350 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2351 this, as well as the form without the "IPv6" on the front (but only when
2352 address literals are enabled, of course).
2354 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2356 10. Exim crashed if a message with an empty sender address specified by -f
2357 encountered a router with an errors_to setting. This could be provoked only
2358 by a command such as
2362 where an empty string was supplied; "<>" did not hit this bug.
2364 11. Installed PCRE release 4.5.
2366 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2367 remained set. It is now erased.
2369 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2370 times from message ids (which are base 36 rather than the normal 62).
2372 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2373 were being counted as actual protocol errors, and logged if the log
2374 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2375 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2376 and DATA following a set of rejected RCPTs do not count as protocol errors.
2377 In other words, Exim assumes they were pipelined, though this may not
2378 actually be the case. Of course, in all cases the client gets an
2379 appropriate error code.
2381 15. If a lookup fails in an ACL condition, a message about the failure may
2382 be available; it is used if testing the ACL cannot continue, because most
2383 such messages specify what the cause of the deferral is. However, some
2384 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2385 that caused an old message to be retained and used if a later statement
2386 caused a defer, replacing the real cause of the deferral.
2388 16. If an IP address had so many PTR records that the DNS lookup buffer
2389 was not large enough to hold them, Exim could crash while trying to process
2390 the truncated data. It now detects and logs this case.
2392 17. Further to 4.21/58, another change has been made: if (and only if) the
2393 first line of a message (the first header line) ends with CRLF, a bare LF
2394 in a subsequent header line has a space inserted after it, so as not to
2395 terminate the header.
2397 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2398 unnecessarily, used atoi() instead of strtol(), and didn't check the
2399 termination when getting file sizes from file names by regex.
2401 19. Completely re-implemented the support for maildirsize files, in the light
2402 of a number of problems with the previous contributed implementation
2403 (4.30/29). In particular:
2405 . If the quota is zero, the maildirsize file is maintained, but no quota is
2408 . If the maildir directory does not exist, it is created before any attempt
2409 to write a maildirsize file.
2411 . The quota value in the file is just a cache; if the quota is changed in
2412 the transport, the new value overrides.
2414 . A regular expression is available for excluding directories from the
2417 20. The autoreply transport checks the characters in options that define the
2418 message's headers; it allows continued headers, but it was checking with
2419 isspace() after an embedded newline instead of explicitly looking for a
2422 21. If all the "regular" hosts to which an address was routed had passed their
2423 expiry times, and had not reached their retry times, the address was
2424 bounced, even if fallback hosts were defined. Now Exim should go on to try
2427 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2428 equivalent code in the SMTP transport. Some hosts send humungous responses
2429 to HELO/EHLO, more than 1024 it seems.
2431 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2432 gives compiler warnings in some environments. I've now done it "properly",
2435 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2436 (because of problems with the built-in one) was declared to return uschar *
2437 instead of char *, causing compiler failure.
2439 25. Fixed a file descriptor leak when processing alias/forward files.
2441 26. Fixed a minor format string issue in dbfn.c.
2443 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2445 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2446 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2449 29. When a sender address is verified, it is cached, to save repeating the test
2450 when there is more than one recipient in a message. However, when the
2451 verification involves a callout, it is possible for different callout
2452 options to be set for different recipients. It is too complicated to keep
2453 track of this in the cache, so now Exim always runs a verification when a
2454 callout is required, relying on the callout cache for the optimization.
2455 The overhead is duplication of the address routing, but this should not be
2458 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2459 to be verified with callout=postmaster, and the main callout worked but the
2460 postmaster check failed, the verification correctly failed. However, if a
2461 subsequent RCPT command asked for sender verification *without* the
2462 postmaster check, incorrect caching caused this verification also to fail,
2465 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2466 it was not caching the DNS options (qualify_single, search_parents) that
2467 were used when the lookup failed. A subsequent lookup with different
2468 options therefore always gave the same answer, though there were cases
2469 where it should not have. (Example: a "domains = !$mx_any" option on a
2470 dnslookup router: the "domains" option is always processed without any
2471 widening, but the router might have qualify_single set.) Now Exim uses the
2472 cached value only when the same options are set.
2474 32. Added John Jetmore's "exipick" utility to the distribution.
2476 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2477 than a timeout (e.g. a certificate is required, and is not provided), an
2478 Exim server now closes the connection immediately. Previously it waited for
2479 the client to close - but if the client is SSL, it seems that they each
2480 wait for each other, leading to a delay before one of them times out.
2482 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2483 maintained 0.8.x compatibility because I don't think many are using it, and
2484 it is clearly obsolete.
2486 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2489 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2490 client certificate was expired. A simple patch fixes this, though I don't
2491 understand the full logic of why the verify callback is called multiple
2494 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2495 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2496 which causes problems with some clients (such as the Certicom SSL Plus
2497 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2498 disables the coutermeasure allowing Eudora to connect."
2500 38. Exim was not checking that a write() to a log file succeeded. This could
2501 lead to Bad Things if a log got too big, in particular if it hit a file
2502 size limit. Exim now panics and dies if it cannot write to a log file, just
2503 as it does if it cannot open a log file.
2505 39. Modified OS/Makefile-Linux so that it now contains
2507 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2509 The two -D definitions ensure that Exim is compiled with large file
2510 support, which makes it possible to handle log files that are bigger than
2513 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2514 instance) a domain was checked against a named list that involved a lookup,
2515 causing $domain_data to be set, then another domain was checked against the
2516 same list, then the first domain was re-checked, the value of $domain_data
2517 after the final check could be wrong. In particular, if the second check
2518 failed, it could be set empty. This bug probably also applied to
2521 41. The strip_trailing_dot option was not being applied to the address given
2522 with the -f command-line option.
2524 42. The code for reading a message's header from the spool was incrementing
2525 $received_count, but never initializing it. This meant that the value was
2526 incorrect (doubled) while delivering a message in the same process in which
2527 it was received. In the most common configuration of Exim, this never
2528 happens - a fresh exec is done - but it can happen when
2529 deliver_drop_privilege is set.
2531 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2532 it now includes up to 150 characters of the unexpected data in the log
2535 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2536 and building data strings. The size of both of these buffers was 10 000
2537 bytes - far larger than anybody would *ever* want, thought I. Needless to
2538 say, somebody hit the limit. I have increased the maximum line length to
2539 20 000 and the maximum data length of concatenated lines to 100 000. I have
2540 also fixed two bugs, because there was no checking on these buffers. Tsk,
2541 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2542 buffer is too small.
2544 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2545 lsearch lookups. Now it does.
2547 46. When parsing a route_list item in a manualroute router, a fixed-length
2548 buffer was used for the list of hosts. I made this 1024 bytes long,
2549 thinking that nobody would ever have a list of hosts that long. Wrong.
2550 Somebody had a whole pile of complicated expansion conditions, and the
2551 string was silently truncated, leading to an expansion error. It turns out
2552 that it is easier to change to an unlimited length (owing to other changes
2553 that have happened since this code was originally written) than to build
2554 structure for giving a limitation error. The length of the item that
2555 expands into the list of hosts is now unlimited.
2557 47. The lsearch lookup could not handle data where the length of text line was
2558 more than 4095 characters. Such lines were truncated, leading to shortened
2559 data being returned. It should now handle lines of any length.
2561 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2562 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2565 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2566 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2567 no longer happens in comment lines. A list of replacements is now placed
2568 at the head of all of the source files, except those whose only change is
2569 to replace PERL_COMMAND in the very first #! line.
2571 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2572 messages on the queue, with a bottom-up merge sort, using code contributed
2573 by Michael Haardt. This should make operations like -bp somewhat faster on
2574 large queues. It won't affect queue runners, except when queue_run_in_order
2577 51. Installed eximstats 1.31 in the distribution.
2579 52. Added support for SRV lookups to the dnslookup router.
2581 53. If an ACL referred to $message_body or $message_body_end, the value was not
2582 reset for any messages that followed in the same SMTP session.
2584 54. The store-handling optimization for building very long strings was not
2585 differentiating between the different store pools. I don't think this
2586 actually made any difference in practice, but I've tidied it.
2588 55. While running the routers to verify a sender address, $sender_address
2589 was still set to the sender address. This is wrong, because when routing to
2590 send a bounce to the sender, it would be empty. Therefore, I have changed
2591 it so that, while verifying a sender address, $sender_address is set to <>.
2592 (There is no change to what happens when verifying a recipient address.)
2594 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2595 target A or AAAA records (if not already returned) without resetting the
2596 qualify_single or search_parents options of the DNS resolver. These are
2597 inappropriate in this case because the targets of MX and SRV records must
2598 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2599 target that, when qualified, matched something in the local domain. These
2600 two options are now turned off when doing these lookups.
2602 57. It seems that at least some releases of Reiserfs (which does not have the
2603 concept of a fixed number of inodes) returns zero and not -1 for the
2604 number of available inodes. This interacted badly with check_spool_inodes,
2605 which assumed that -1 was the "no such thing" setting. What I have done is
2606 to check that the total number of inodes is greater than zero before doing
2607 the test of how many are available.
2609 58. When a "warn" ACL statement has a log_message modifier, the message is
2610 remembered, and not repeated. This is to avoid a lot of repetition when a
2611 message has many recipients that cause the same warning to be written.
2612 Howewer, Exim was preserving the list of already written lines for an
2613 entire SMTP session, which doesn't seem right. The memory is now reset if a
2614 new message is started.
2616 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2617 debugging output unless log_rewrite was also set.
2619 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2620 of (int)(handle) when we know that handle contains (void *)(-1).
2622 61. The Exim daemon panic-logs an error return when it closes the incoming
2623 connection. However "connection reset by peer" seems to be common, and
2624 isn't really an error worthy of noting specially, so that particular error
2627 62. When Exim is trying to find all the local interfaces, it used to panic and
2628 die if the ioctl to get the interface flags failed. However, it seems that
2629 on at least one OS (Solaris 9) it is possible to have an interface that is
2630 included in the list of interfaces, but for which you get a failure error
2631 for this call. This happens when the interface is not "plumbed" into a
2632 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2633 failure of the "get flags" call assumes that the interface is down.
2635 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2636 makes life easier for people who are doing arithmetic on fields extracted
2637 from dates, where you often get leading zeros that should not be
2638 interpreted as octal.
2640 64. Added qualify_domain to the redirect router, to override the global
2643 65. If a pathologically long header line contained very many addresses (the
2644 report of this problem mentioned 10 000) and each of them was rewritten,
2645 Exim could use up a very large amount of memory. (It kept on making new
2646 copies of the header line as it rewrote, and never released the old ones.)
2647 At the expense of a bit more processing, the header rewriting function has
2648 been changed so that it no longer eats memory in this way.
2650 66. The generation of the Received: header has been moved from the time that a
2651 message starts to be received, to the time that it finishes. The timestamp
2652 in the Received: header should now be very close to that of the <= log
2653 line. There are two side-effects of this change:
2655 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2656 logged header lines no longer include the local Received: line, because
2657 it has not yet been created. The same applies to a copy of the message
2658 that is returned to a non-SMTP sender when a message is rejected.
2660 (b) When a filter file is tested using -bf, no additional Received: header
2661 is added to the test message. After some thought, I decided that this
2664 This change does not affect the value of $received_for. It is still set
2665 after address rewriting, but before local_scan() is called.
2667 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2669 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2670 gave an unhelpful panic error message, and a defer error. I have managed to
2671 change this behaviour so that it now rejects any supplied certificate,
2672 which seems right, as the list of acceptable certificates is empty.
2674 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2675 gave an unhelpful defer error. I have not managed to make this reject any
2676 supplied certificates, but the error message it gives is "no certificate
2677 supplied", which is not helpful.
2679 70. exigrep's output now also includes lines that are not associated with any
2680 message, but which match the given pattern. Implemented by a patch from
2681 Martin Sluka, which also tidied up the Perl a bit.
2683 71. Recipient callout verification, like sender verification, was using <> in
2684 the MAIL FROM command. This isn't really the right thing, since the actual
2685 sender may affect whether the remote host accepts the recipient or not. I
2686 have changed it to use the actual sender in the callout; this means that
2687 the cache record is now keyed on a recipient/sender pair, not just the
2688 recipient address. There doesn't seem to be a real danger of callout loops,
2689 since a callout by the remote host to check the sender would use <>.
2690 [SEE ABOVE: changed after hitting problems.]
2692 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2693 temporary errors. However, in the case of such a code being given after
2694 the end of a data transmission (i.e. after ".") Exim was failing to write
2695 a retry record for the message. (Yes, there was some broken host that was
2696 actually sending 8xx at this point.)
2698 73. An unknown lookup type in a host list could cause Exim to panic-die when
2699 the list was checked. (An example that provoked this was putting <; in the
2700 middle of a list instead of at the start.) If this happened during a DATA
2701 ACL check, a -D file could be left lying around. This kind of configuration
2702 error no longer causes Exim to die; instead it causes a defer errror. The
2703 incident is still logged to the main and panic logs.
2705 74. Buglet left over from Exim 3 conversion. The message "too many messages
2706 in one connection" was written to the rejectlog but not the mainlog, except
2707 when address rewriting (yes!) was being logged.
2709 75. Added write_rejectlog option.
2711 76. When a system filter was run not as root (that is, when system_filter_user
2712 was set), the values of the $n variables were not being returned to the
2713 main process; thus, they were not subsequently available in the $sn
2716 77. Added +return_path_on_delivery log selector.
2718 78. A connection timeout was being treated differently from recipients deferred
2719 when testing hosts_max_try with a message that was older than the host's
2720 retry timeout. (The host should not be counted, thus allowing all hosts to
2721 be tried at least once before bouncing.) This may have been the cause of an
2722 occasionally reported bug whereby a message would remain on the queue
2723 longer than the retry timeout, but would be bounced if a delivery was
2724 forced. I say "may" because I never totally pinned down the problem;
2725 setting up timeout/retry tests is difficult. See also the next item.
2727 79. The ultimate address timeout was not being applied to errors that involved
2728 a combination of host plus message (for example, a timeout on a MAIL
2729 command). When an address resolved to a number of possible hosts, and they
2730 were not all tried for each delivery (e.g. because of hosts_max_try), a
2731 message could remain on the queue longer than the retry timeout.
2733 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2736 81. Fixed an obscure SMTP outgoing bug which required at least the following
2737 conditions: (a) there was another message waiting for the same server;
2738 (b) the server returned 5xx to all RCPT commands in the first message so
2739 that the message was not completed; (c) the server dropped the connection
2740 or gave a negative response to the RSET that Exim sends to abort the
2741 transaction. The observed case was a dropped connection after DATA that had
2742 been sent in pipelining mode. That is, the server had advertised PIPELINING
2743 but was not implementing it correctly. The effect of the bug was incorrect
2744 behaviour, such as trying another host, and this could lead to a crash.
2750 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2751 and daemon.c were passed as pointers to ints; they should have been
2752 pointers to socklen_t variables (which are typically unsigned ints).
2754 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2757 3. Fixed a really odd bug that affected only the testing scheme; patching a
2758 certain fixed string in the binary changed the value of another string that
2759 happened to be identical to the end of the original first string.
2761 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2762 name", it returns that address as the IP address. On some operating
2763 systems (e.g. Solaris), it also passes back the IP address string as the
2764 "host name". However, on others (e.g. Linux), it passes back an empty
2765 string. Exim wasn't checking for this, and was changing the host name to an
2766 empty string, assuming it had been canonicized.
2768 5. Although rare, it is permitted to have more than one PTR record for a given
2769 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2770 all the names associated with an address, because they do in Solaris.
2771 However, it seems that they do not in Linux for data that comes from the
2772 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2773 I found this out when I moved to a new Linux workstation and tried to run
2774 the Exim test suite.
2776 To get round this problem I have changed the code so that it now does its
2777 own call to the DNS to look up PTR records when searching for a host name.
2778 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2779 addresses that are only in /etc/hosts are still found.
2781 This behaviour is, however, controlled by an option called host_lookup_
2782 order, which defaults to "bydns:byaddr". If people want to use the other
2783 order, or indeed, just use one or the other means of lookup, they can
2784 specify it in this variable.
2786 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2787 some operating systems, this comes back from gethostbyaddr() as an empty
2788 string, and this is what Exim used to test for. However, it seems that in
2789 other systems, "." is yielded. Exim now tests for this case too.
2791 7. The values of check_spool_space and check_log_space are now held internally
2792 as a number of kilobytes instead of an absolute number of bytes. If a
2793 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2794 kilobyte. This means that much larger values can be stored.
2796 8. Exim monitor: an attempt to get the action menu when not actually pointing
2797 at a message produces an empty menu entitled "No message selected". This
2798 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2799 no entries in it ("Shell widget menu has zero width and/or height"). So I
2800 have added a single, blank menu entry in this case.
2802 9. Added ${quote_local_part.
2804 10. MIME decoding is now applied to the contents of Subject: header lines when
2807 11. Now that a reference to $sender_host_address automatically causes a reverse
2808 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2809 host lookup before query-style lookups in lists that might use this
2810 variable. This has therefore been abolished, and the "net-" prefix is no
2811 longer necessary for query-style lookups.
2813 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2814 have been a typo for LFLAGS, so it has been changed.
2816 13. The install script calls Exim with "-C /dev/null" in order to find the
2817 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2818 to be output. Howeve, since Exim outputs its version number before the
2819 error, it didn't break the script. It just looked ugly. I fixed this by
2820 always allowing "-C /dev/null" if the caller is root.
2822 14. Ignore overlarge ACL variable number when reading spool file - insurance
2823 against a later release with more variables having written the file.
2825 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2826 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2827 this, as well as the form without the "IPv6" on the front.
2829 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2830 OS/Makefile-Darwin file.
2832 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2833 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2835 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2836 variable. However, some OS use ints for inodes. I've added cast to long int
2837 to get rid of the compiler warning.
2839 19. I had forgotten to lock out "/../" in configuration file names when
2840 ALT_CONFIG_PREFIX was set.
2842 20. Routers used for verification do not need to specify transports. However,
2843 if such a router generated a host list, and callout was configured, Exim
2844 crashed, because it could not find a port number from the (non-existent)
2845 transport. It now assumes port 25 in this circumstance.
2847 21. Added the -t option to exigrep.
2849 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2850 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2851 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2852 them is set, LOOKUP_LSEARCH is forced.
2854 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2856 24. Added sender and host information to the "rejected by local_scan()" log
2857 line; previously there was no indication of these.
2859 25. Added .include_if_exists.
2861 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2862 Linux. It turns out that not all file systems support this. Apparently some
2863 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2864 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2865 not supported on the file descriptor, is now ignored when Exim is trying to
2866 sync a directory. This applies only to Linux.
2868 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2870 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2871 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2874 29. Added support for maildirsize files from supplied patch (modified a bit).
2876 30. The use of :fail: followed by an empty string could lead Exim to respond to
2877 sender verification failures with (e.g.):
2879 550 Verification failed for <xxx>
2880 550 Sender verify failed
2882 where the first response line was missing the '-' that indicates it is not
2883 the final line of the response.
2885 31. The loop for finding the name of the user that called Exim had a hardwired
2886 limit of 10; it now uses the value of finduser_retries, which is used for
2887 all other user lookups.
2889 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2892 33. Exim was neglecting to zero errno before one call of strtol() when
2893 expanding a string and expecting an integer value. On some systems this
2894 resulted in spurious "integer overflow" errors. Also, it was casting the
2895 result into an int without checking.
2897 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2898 did not work. The code looks as if it has *never* worked, though it appears
2899 to have been documented since at least releast 1.62. I have made it work.
2901 35. The "timeout_DNS" error in retry rules, also documented since at least
2902 1.62, also never worked. As it isn't clear exactly what this means, and
2903 clearly it isn't a major issue, I have abolished the feature by treating it
2904 as "timeout", and writing a warning to the main and panic logs.
2906 36. The display of retry rules for -brt wasn't always showing the error code
2909 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2910 timeout_connect_A, timeout_connect_MX.
2912 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2913 to the empty sender.
2915 39. The daemon was not analysing the content of -oX till after it had closed
2916 stderr and disconnected from the controlling terminal. This meant that any
2917 syntax errors were only noted on the panic log, and the return code from
2918 the command was 0. By re-arranging the code a little, I've made the
2919 decoding happen first, so such errors now appear on stderr, and the return
2920 code is 1. However, the actual setting up of the sockets still happens in
2921 the disconnected process, so errors there are still only recorded on the
2924 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2925 connections (as happens on some IP stacks) was logged at start up time as
2926 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2927 it from "IPv6 and IPv4", which means that two separate sockets are being
2930 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2931 says whether AF_INET or AF_INET6 was passed as an argument.
2933 42. Exiwhat output was messed up when time zones were included in log
2936 43. Exiwhat now gives more information about the daemon's listening ports,
2937 and whether -tls-on-connect was used.
2939 44. The "port" option of the smtp transport is now expanded.
2941 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2942 silently ignored. Now an error message is written to the main and panic
2945 46. There's a new ACL modifier called "logwrite" which writes to a log file
2946 as soon as it is encountered.
2948 47. Added $local_user_uid and $local_user_gid at routing time.
2950 48. Exim crashed when trying to verify a sender address that was being
2953 49. Exim was recognizing only a space character after ".include". It now also
2954 recognizes a tab character.
2956 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2957 extracting the relevant information from the specification. The man page no
2958 longer contains scrambled data for the -d option, and I've added a section
2959 at the front about calling Exim under different names.
2961 51. Added "extra_headers" argument to the "mail" command in filter files.
2963 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2966 53. Installed eximstats 1.29.
2968 54. Added transport_filter_timeout as a generic transport option.
2970 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2971 Cc: header lines. This was required by RFC 822, but it not required by RFC
2974 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2975 incoming messages that did not have them. Now it does so only if the
2976 message originates locally, that is, if there is no associated remote host
2977 address. When Resent- header lines are present, this applies to the Resent-
2978 lines rather than the non-Resent- lines.
2980 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2981 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2983 58. Messages for configuration errors now include the name of the main
2984 configuration file - useful now that there may be more than one file in a
2985 list (.included file names were always shown).
2987 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2988 for those rare installations that do not start the daemon as root or run it
2989 setuid root. I've cut out the call to initgroups() if the daemon is not
2992 60. The Exim user and group can now be bound into the binary as text strings
2993 that are looked up at the start of Exim's processing.
2995 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2997 62. Added $mailstore_basename variable.
2999 63. Installed patch to sieve.c from Michael Haardt.
3001 64. When Exim failed to open the panic log after failing to open the main log,
3002 the original message it was trying to log was written to stderr and debug
3003 output, but if they were not available (the usual case in production), it
3004 was lost. Now it is written to syslog before the two lines that record the
3005 failures to open the logs.
3007 65. Users' Exim filters run in subprocesses under the user's uid. It is
3008 possible for a "deliver" command or an alias in a "personal" command to
3009 provoke an address rewrite. If logging of address rewriting is configured,
3010 this fails because the process is not running as root or exim. There may be
3011 a better way of dealing with this, but for the moment (because 4.30 needs
3012 to be released), I have disabled address rewrite logging when running a
3013 filter in a non-root, non-exim process.
3019 1. The buildconfig auxiliary program wasn't quoting the value set for
3020 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3021 not defined. This bug was masked in 4.22 by the effect that was fixed in
3024 2. Some messages that were rejected after a message id was allocated were
3025 shown as "incomplete" by exigrep. It no longer does this for messages that
3026 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3028 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3029 have allow_domain_literals set, the ID did not get logged in the <= line.
3030 Domain literals are now always recognized in Message-ID: header lines.
3032 4. The first argument for a ${extract expansion item is the key name or field
3033 number. Leading and trailing spaces in this item were not being ignored,
3034 causing some misleading effects.
3036 5. When deliver_drop_privilege was set, single queue runner processes started
3037 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3038 same command in the process it spins off) were not dropping privilege.
3040 6. When the daemon running as "exim" started a queue runner, it always
3041 re-executed Exim in the spun-off process. This is a waste of effort when
3042 deliver_drop_privilege is set. The new process now just calls the
3043 queue-runner function directly.
3049 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3052 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3053 ignored. Though the use of -oP was forcing the writing of a pid file, it
3054 was always written to the default place.
3056 3. If the message "no IP address found for host xxxx" is generated during
3057 incoming verification, it is now followed by identification of the incoming
3058 connection (so you can more easily find what provoked it).
3060 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3062 5. Added some features to "harden" Exim a bit more against certain attacks:
3064 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3065 be put in Local/Makefile. This is like the never_users runtime option,
3066 but it cannot be overridden. The default setting is "root".
3068 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3069 prefix string with which any file named in a -C command line option
3072 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3073 is retained for -C and -D only if the caller of Exim is root. Without
3074 it, the exim user may also use -C and -D and retain privilege.
3076 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3077 command line option is disabled.
3079 6. Macro names set by the -D option must start with an upper case letter, just
3080 like macro names defined in the configuration file.
3082 7. Added "dereference=" facility to LDAP.
3084 8. Two instances of the typo "uknown" in the source files are fixed.
3086 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3087 the Configure-Makefile script screwed up while processing it.
3089 10. Incorporated PCRE 4.4.
3091 11. The SMTP synchronization check was not operating right at the start of an
3092 SMTP session. For example, it could not catch a HELO sent before the client
3093 waited for the greeting. There is now a check for outstanding input at the
3094 point when the greeting is written. Because of the duplex, asynchronous
3095 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3096 way, but not yet received, when the check is performed.
3098 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3099 on TCP/IP sockets, because this apparently causes some broken clients to
3102 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3103 unchanged) from the Cygwin maintainer.
3105 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3106 was supported instead of testing for mbx. Effectively a typo.
3108 15. The spa authenticator server code was not checking that the input it
3109 received was valid base64.
3111 16. The debug output line for the "set" modifier in ACLs was not showing the
3112 name of the variable that was being set.
3114 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3116 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3117 lookup on its own. Something else was needed to trigger the lookup. For
3118 example, a match in host_lookup or the need for a host name in a host list.
3119 Now, if $sender_host_name is referenced and the host name has not yet been
3120 looked up, a lookup is performed. If the lookup fails, the variable remains
3121 empty, and $host_lookup_failed is set to "1".
3123 19. Added "eqi" as a case-independent comparison operator.
3125 20. The saslauthd authentication condition could segfault if neither service
3126 nor realm was specified.
3128 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3129 error message that was logged was misleading, and incoming SMTP
3130 connections were dropped. The message is now more accurate, and temporary
3131 errors are given to SMTP connections.
3133 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3134 (except RSET) with a 421 error, until QUIT is received. However, it was
3135 failing to send a response to QUIT.
3137 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3138 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3139 ACL is now run after the helo_try_verify_hosts code.
3141 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3142 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3143 case-independent, so other case variants are also recognized). Apparently
3144 some systems use these upper case variants.
3146 25. If more than two messages were waiting for the same host, and a transport
3147 filter was specified for the transport, Exim sent two messages over the
3148 same TCP/IP connection, and then failed with "socket operation on non-
3149 socket" when it tried to send the third.
3151 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3153 27. The extern definition of crypt16() in expand.c was not being excluded when
3154 the OS had its own crypt16() function.
3156 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3157 as a preferred synonym for return_size_limit, both as an option and as an
3160 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3162 30. Changed the default configuration ACL to relax the local part checking rule
3163 for addresses that are not in any local domains. For these addresses,
3164 slashes and pipe symbols are allowed within local parts, but the sequence
3165 /../ is explicitly forbidden.
3167 31. SPA server authentication was not clearing the challenge buffer before
3170 32. log_message in a "warn" ACL statement was writing to the reject log as
3171 well as to the main log, which contradicts the documentation and doesn't
3172 seem right (because no rejection is happening). So I have stopped it.
3174 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3175 However, I am unable to do any testing of this.
3177 34. Fixed an infelicity in the appendfile transport. When checking directories
3178 for a mailbox, to see if any needed to be created, it was accidentally
3179 using path names with one or more superfluous leading slashes; tracing
3180 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3182 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3183 contents are added to the log line that is written for every discarded
3184 recipient. (Previously a log_message setting was ignored.)
3186 36. The ${quote: operator now quotes the string if it is empty.
3188 37. The install script runs exim in order to find its version number. If for
3189 some reason other than non-existence or emptiness, which it checks, it
3190 could not run './exim', it was installing it with an empty version number,
3191 i.e. as "exim-". This error state is now caught, and the installation is
3194 38. An argument was missing from the function that creates an error message
3195 when Exim fails to connect to the socket for saslauthd authentication.
3196 This could cause Exim to crash, or give a corrupted message.
3198 39. Added isip, isip4, and isip6 to ${if conditions.
3200 40. The ACL variables $acl_xx are now saved with the message, and can be
3201 accessed later in routers, transports, and filters.
3203 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3204 strings in the file are not string-expanded.
3206 42. If a MAIL command specified a SIZE value that was too large to fit into an
3207 int variable, the check against message_size_limit failed. Such values are
3208 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3209 day this will have to be increased, but I don't think I want to be around
3210 when emails are that large.
3217 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3218 iconv() is not standard in FreeBSD.
3220 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3221 IPv6 enabled. The observed symptom was a segmentation fault on return from
3222 the function os_common_find_running_interfaces() in src/os.c.
3224 3. In the check_special_case() function in daemon.c I had used "errno" as an
3225 argument name, which causes warnings on some systems. This was basically a
3226 typo, since it was named "eno" in the comments!
3228 4. The code that waits for the clock to tick (at a resolution of some fraction
3229 of a second) so as to ensure message-id uniqueness was always waiting for
3230 at least one whole tick, when it could have waited for less. [This is
3231 almost certainly not relevant at current processor speeds, where it is
3232 unlikely to ever wait at all. But we try to future-proof.]
3234 5. The function that sleeps for a time interval that includes fractions of a
3235 second contained a race. It did not block SIGALRM between setting the
3236 timer, and suspending (a couple of lines later). If the interval was short
3237 and the sigsuspend() was delayed until after it had expired, the suspension
3238 never ended. On busy systems this could lead to processes getting stuck for
3241 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3242 process, before it forks any delivery processes. The open lookup caching
3243 mechanism meant that the open file or database connection was passed into
3244 the delivery process. The problem was that delivery processes always tidy
3245 up cached lookup data. This could cause a problem for the next delivery
3246 process started by the queue runner, because the external queue runner
3247 process does not know about the closure. So the next delivery process
3248 still has data in the lookup cache. In the case of a file lookup, there was
3249 no problem because closing a file descriptor in a subprocess doesn't affect
3250 the parent. However, if the lookup was caching a connection to a database,
3251 the connection was closed, and the second delivery process was likely to
3252 see errors such as "PGSQL: query failed: server closed the connection
3253 unexpectedly". The problem has been fixed by closing all cached lookups
3254 in a queue runner before running a delivery process.
3256 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3257 seem to have the "const" qualifier which it has on other OS. I've
3260 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3261 *of the same type* (client or server) with the same public name that an
3262 error should be diagnosed.
3264 9. When Exim looked up a host name for an IP address, but failed to find the
3265 original IP address when looking up the host name (a safety check), it
3266 output the message "<ip address> does not match any IP for NULL", which was
3267 confusing, to say the least. The bug was that the host name should have
3268 appeared instead of "NULL".
3270 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3271 user that is built into the binary, and the -C or -D options is used, root
3272 privilege is dropped before the configuration file is read. In addition,
3273 logging is switched to stderr instead of the normal log files. If the
3274 configuration then re-defines the Exim user, the unprivileged environment
3275 is probably not what is expected, so Exim logs a panic warning message (but
3278 However, if deliver_drop_privilege is set, the unprivileged state may well
3279 be exactly what is intended, so the warning has been cut out in that case,
3280 and Exim is allowed to try to write to its normal log files.
3286 1. smtp_return_error_details was not giving details for temporary sender
3287 or receiver verification errors.
3289 2. Diagnose a configuration error if two authenticators have the same public
3292 3. Exim used not to create the message log file for a message until the first
3293 delivery attempt. This could be confusing when incoming messages were held
3294 for policy or load reasons. The message log file is now created at the time
3295 the message is received, and an initial "Received" line is written to it.
3297 4. The automatically generated man page for command line options had a minor
3298 bug that caused no ill effects; however, a more serious problem was that
3299 the procedure for building the man page automatically didn't always
3300 operate. Consequently, release 4.20 contains an out-of-date version. This
3301 shouldn't happen again.
3303 5. When building Exim with embedded Perl support, the script that builds the
3304 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3305 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3307 6. The freeze_tell option was not being used for messages that were frozen on
3308 arrival, either by an ACL or by local_scan().
3310 7. Added the smtp_incomplete_transaction log selector.
3312 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3313 was accepting AUTH without a new EHLO.
3315 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3316 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3319 10. Logging of TCP/IP connections (when configured) now happens in the main
3320 daemon process instead of the child process, so that the TCP/IP connection
3321 count is more accurate (but it can never be perfect).
3323 11. The use of "drop" in a nested ACL was not being handled correctly in the
3324 outer ACL. Now, if condition failure induced by the nested "drop" causes
3325 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3326 or "require"), the connection is dropped.
3328 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3329 that yield "discard" can now be used with an "accept" or a "discard" verb,
3330 but an error is generated for any others (because I can't see a useful way
3331 to define what should happen).
3333 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3334 are now processed in the same way as lines in the Exim configuration file.
3335 In particular, continuation lines are supported.
3337 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3339 15. Added -ti meaning -t -i.
3341 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3342 domains, and warn by logging if others are found.
3344 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3345 returned from GIFCONF when Exim is trying to find the list of interfaces on
3346 a host. The code in os.c has been modified to copy each ifreq to an aligned
3347 structure in all cases.
3349 Also, in some cases, the returned ifreq's were being copied to a 'struct
3350 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3351 means the last couple of bytes of an IPv6 address could be chopped if the
3352 ifreq contained only a normal sockaddr (14 bytes storage).
3354 18. Named domain lists were not supported in the hosts_treat_as_local option.
3355 An entry such as +xxxx was not recognized, and was treated as a literal
3358 19. Ensure that header lines added by a DATA ACL are included in the reject log
3359 if the ACL subsequently rejects the message.
3361 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3362 MD5 (which is deprecated).
3364 21. When testing a filter file using -bf, Exim was writing a message when it
3365 took the sender from a "From " line in the message, but it was not doing so
3366 when it took $return_path from a Return-Path: header line. It now does.
3368 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3369 with a valid header line field name (a series of printing characters
3370 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3372 23. Changed "disc" in the source to "disk" to conform to the documentation and
3373 the book and for uniformity.
3375 24. Ignore Sendmail's -Ooption=value command line item.
3377 25. When execve() failed while trying to run a command in a pipe transport,
3378 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3379 could be confused with a return value of 69 from the command itself. This
3380 has been changed to 127, the value the shell returns if it is asked to run
3381 a non-existent command. The wording for the related log line suggests a
3382 non-existent command as the problem.
3384 26. If received_header_text expands to an empty string, do not add a Received:
3385 header line to the message. (Well, it adds a token one on the spool, but
3386 marks it "old" so that it doesn't get used or transmitted.)
3388 27. Installed eximstats 1.28 (addition of -nt option).
3390 28. There was no check for failure on the call to getsockname() in the daemon
3391 code. This can fail if there is a shortage of resources on the system, with
3392 ENOMEM, for example. A temporary error is now given on failure.
3394 29. Contrary to the C standard, it seems that in some environments, the
3395 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3396 program. Exim now does this explicitly; it affects the formatting of
3397 timestamps using strftime().
3399 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3400 complaints. I've now initialized all the variables, to avoid this.
3402 32. Header lines added by a system filter were not being "seen" during
3403 transport-time rewrites.
3405 33. The info_callback() function passed to OpenSSL is set up with type void
3406 (*)(SSL *, int, int), as described somewhere. However, when calling the
3407 function (actually a macro) that sets it up, the type void(*)() is
3408 expected. I've put in a cast to prevent warnings from picky compilers.
3410 34. If a DNS black list lookup found a CNAME record, but there were no A
3411 records associated with the domain it pointed at, Exim crashed.
3413 35. If a DNS black list lookup returned more than one A record, Exim ignored
3414 all but the first. It now scans all returned addresses if a particular IP
3415 value is being sought. In this situation, the contents of the
3416 $dnslist_value variable are a list of all the addresses, separated by a
3419 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3420 to accept a host name and all its aliases if the forward lookup for any of
3421 them yielded the IP address of the incoming connection. Now it accepts only
3422 those names whose forward lookup yields the correct IP address. Any other
3423 names are discarded. This closes a loophole whereby a rogue DNS
3424 administrator could create reverse DNS records to break through a
3425 wildcarded host restriction in an ACL.
3427 37. If a user filter or a system filter that ran in a subprocess used any of
3428 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3429 the wrong values were passed to the pipe command ($thisaddress had the
3430 value of $0, $0 had the value of $1, etc). This bug was introduced by
3431 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3433 38. Improved the line breaking for long SMTP error messages from ACLs.
3434 Previously, if there was no break point between 40 and 75 characters, Exim
3435 left the rest of the message alone. Two changes have been made: (a) I've
3436 reduced the minimum length to 35 characters; (b) if it can't find a break
3437 point between 35 and 75 characters, it looks ahead and uses the first one
3438 that it finds. This may give the occasional overlong line, but at least the
3439 remaining text gets split now.
3441 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3442 file descriptors might be low, and that setting 1000 would always raise it.
3443 It turns out that in some environments, the limit is already over 1000 and
3444 that lowering it causes trouble. So now Exim takes care not to decrease it.
3446 40. When delivering a message, the value of $return_path is set to $sender_
3447 address at the start of routing (routers may change the value). By an
3448 oversight, this default was not being set up when an address was tested by
3449 -bt or -bv, which affected the outcome if any router or filter referred to
3452 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3453 log only when "message" or "log_message" are set. However, if one of the
3454 conditions was an address verification, or a call to a nested ACL, the
3455 messages generated by the underlying test were being passed through. This
3456 no longer happens. The underlying message is available in $acl_verify_
3457 message for both "message" and "log_message" expansions, so it can be
3458 passed through if needed.
3460 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3461 new expansion $bh_ to give the encoded byte string without charset
3462 translation. Translation happens only if iconv() is available; HAVE_ICONV
3463 indicates this at build time. HEADERS_CHARSET gives the charset to
3464 translate to; headers_charset can change it in the configuration, and
3465 "headers charset" can change it in an individual filter file.
3467 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3468 that creates RFC 2047 encoded "words" labels them as that charset instead
3469 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3470 expansion operator; (ii) when Exim creates a From: line for a local
3471 message; (iii) when a header line is rewritten to include a "phrase" part.
3473 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3474 buggy, causing it to skip the first lines of messages whose message ID
3475 ended in 'D'. This would not have bitten before Exim release 4.14, because
3476 message IDs were unlikely to end in 'D' before then. The effect was to have
3477 incorrect size information for certain domains.
3479 45. #include "config.h" was missing at the start of the crypt16.c module. This
3480 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3483 46. If there was a timeout during a "random" callout check, Exim treated it as
3484 a failure of the random address, and carried on sending RSET and the real
3485 address. If the delay was just some slowness somewhere, the response to the
3486 original RCPT would be taken as a response to RSET and so on, causing
3487 mayhem of various kinds.
3489 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3490 when I implemented it. It didn't allow for the fact that some option values
3491 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3492 the right test for positive values.
3494 48. Domain names in DNS records are case-independent. Exim always looks them up
3495 in lower case. Some resolvers return domain names in exactly the case they
3496 appear in the zone file, that is, they may contain uppercase letters. Not
3497 all resolvers do this - some return always lower case. Exim was treating a
3498 change of case by a resolver as a change of domain, similar to a widening
3499 of a domain abbreviation. This triggered its re-routing code and so it was
3500 trying to route what was effectively the same domain again. This normally
3501 caused routing to fail (because the router wouldn't handle the domain
3502 twice). Now Exim checks for this case specially, and just changes the
3503 casing of the domain that it ultimately uses when it transmits the message
3506 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3509 50. If a filter generated a file delivery with a non-absolute name (possible if
3510 no home directory exists for the router), the forbid_file option was not
3513 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3514 the existing equality matching.
3516 52. Exim was using ints instead of ino_t variables in some places where it was
3517 dealing with inode numbers.
3519 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3520 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3521 called TMPDIR, and if it finds it is different, it changes its value.
3523 54. The smtp_printf() function is now made available to local_scan() so
3524 additional output lines can be written before returning. There is also an
3525 smtp_fflush() function to enable the detection of a dropped connection.
3526 The variables smtp_input and smtp_batched_input are exported to
3529 55. Changed the default runtime configuration: the message "Unknown user"
3530 has been removed from the ACL, and instead placed on the localuser router,
3531 using the cannot_route_message feature. This means that any verification
3532 failures that generate their own messages won't get overridden. Similarly,
3533 the "Unrouteable address" message that was in the ACL for unverifiable
3534 relay addresses has also been removed.
3536 56. Added hosts_avoid_esmtp to the smtp transport.
3538 57. The exicyclog script was not checking for the esoteric option
3539 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3540 will work only if exicyclog is run under the appropriate euid.
3542 58. Following a discussion on the list, the rules by which Exim recognises line
3543 endings on incoming messages have been changed. The -dropcr and drop_cr
3544 options are now no-ops, retained only for backwards compatibility. The
3545 following line terminators are recognized: LF CRLF CR. However, special
3546 processing applies to CR:
3548 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3549 nor a local message in the state where . is a terminator.
3551 (ii) If a bare CR is encountered in a header line, an extra space is added
3552 after the line terminator so as not to end the header. The reasoning
3553 behind this is that bare CRs in header lines are most likely either
3554 to be mistakes, or people trying to play silly games.
3556 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3557 was being incorrectly given as 18 bytes larger than it should have been.
3558 This is a VOB (very old bug).
3560 60. This may never have affected anything current, but just in case it has:
3561 When the local host is found other than at the start of a list of hosts,
3562 the local host, those with the same MX, and any that follow, are discarded.
3563 When the list in question was part of a longer list of hosts, the following
3564 hosts (not currently being processed) were also being discarded. This no
3565 longer happens. I'm not sure if this situation could ever has previously
3568 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3571 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3572 header. This is something that is recommended in a new Internet Draft, and
3573 is something that is documented as being done by Sendmail. There are two
3574 possible values. For messages generated by the autoreply transport, Exim
3577 Auto-Submitted: auto-replied
3579 whereas for all other generated messages (e.g. bounces) it adds
3581 Auto-Submitted: auto-generated
3583 63. The "personal" condition in filters now includes a test for the
3584 Auto-Submitted: header. If it contains the string "auto-" the message it
3585 not considered personal.
3587 64. Added rcpt_include_affixes as a generic transport option.
3589 65. Added queue_only_override (default true).
3591 66. Added the syslog_duplication option.
3593 67. If what should have been the first header line of a message consisted of
3594 a space followed by a colon, Exim was mis-interpreting it as a header line.
3595 It isn't of course - it is syntactically invalid and should therefore be
3596 treated as the start of the message body. The misbehaviour could have
3597 caused a number of strange effects, including loss of data in subsequent
3598 header lines, and spool format errors.
3600 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3601 client host had authenticated. This control can now be exercised by an ACL
3602 for more flexibility.
3604 69. By default, callouts do not happen when testing with -bh. There is now a
3605 variant, -bhc, which does actually run the callout code, including
3606 consulting and updating the callout cache.
3608 70. Added support for saslauthd authentication, courtesy of Alexander
3611 71. If statvfs() failed on the spool or log directories while checking their
3612 size for availability, Exim confusingly gave the error "space shortage".
3613 Furthermore, in debugging mode it crashed with a floating point exception.
3614 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3615 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3616 serious problem, Exim now writes to the main and panic logs when this
3617 happens, with details of the failure. It then refuses to accept the
3618 incoming message, giving the message "spool directory problem" or "log
3619 directory problem" with a 421 code for SMTP messages.
3621 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3622 0, 1, and 2 exist, because some OS complain for execs without them (see
3623 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3624 descriptors. However, the code omitted to check that the open succeeded,
3625 causing mysterious errors if for some reason the permissions on /dev/null
3626 got screwed. Now Exim writes a message to the main and panic logs, and
3627 bombs out if it can't open /dev/null.
3629 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3630 interact so that it is all more flexible. It is supposed to remain
3631 backwards compatible. Also added extra_local_interfaces.
3633 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3634 to bomb out with an assertion failure - to the client this appears as a
3635 connection drop. This problem occurs in the part of the code that was taken
3636 from the Samba project. Fortunately, the assertion is in a very simple
3637 function, so I have fixed this by reproducing the function inline in the
3638 one place where it is called, and arranging for authentication to fail
3639 instead of killing the process with assert().
3641 75. The SPA client code was not working when the server requested OEM rather
3642 than Unicode encoding.
3644 76. Added code to make require_files with a specific uid setting more usable in
3645 the case where statting the file as root fails - usually a non-root-mounted
3646 NFS file system. When this happens and the failure is EACCES, Exim now
3647 forks a subprocess and does the per-uid checking as the relevant uid.
3649 77. Added process_log_path.
3651 78. If log_file_path was not explicitly set, a setting of check_log_space or
3652 check_log_inodes was ignored.
3654 79. If a space check for the spool or log partitions fails, the incident is now
3655 logged. Of course, in the latter case the data may get lost...
3657 80. Added the %p formatting code to string_format() so that it can be used to
3658 print addresses in debug_print(). Adjusted all the address printing in the
3659 debugging in store.c to use %p rather than %d.
3661 81. There was a concern that a line of code in smtp_in.c could overflow a
3662 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3663 initially expressed, the concern was not well-founded, because trailing
3664 spaces are removed early. However, if the trailing spaces were followed by
3665 a NULL, they did not get removed, so the overflow was possible. Two fixes
3668 (a) I re-wrote the offending code in a cleaner fashion.
3669 (b) If an incoming SMTP command contains a NULL character, it is rejected
3672 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3673 runs initgroups(), so that if the Exim user is in any additional groups,
3674 they will be used during message reception.
3680 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob