1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.284 2006/02/07 14:05:17 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.
93 PH/01 Two changes to the default runtime configuration:
95 (1) Move the checks for relay_from_hosts and authenticated clients from
96 after to before the (commented out) DNS black list checks.
98 (2) Add control=submission to the relay_from_hosts and authenticated
99 clients checks, on the grounds that messages accepted by these
100 statements are most likely to be submissions.
102 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
104 (1) Generate an error if the third argument for the ${prvs expansion is
107 (2) Treat a missing third argument of ${prvscheck as if it were an empty
110 (3) Reset the variables that are obtained from the first argument of
111 ${prvscheck and used in the second argument before leaving the code,
112 because their memory is reclaimed, so using them afterwards may do
115 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
116 one (it's much easier than Tom thought :-).
118 (5) Because of (4), we can now allow for the use of $prvscheck_result
119 inside the third argument.
121 PH/03 For some reason, the default setting of PATH when running a command from
122 a pipe transport was just "/usr/bin". I have changed it to
125 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
126 anything to be listed in the output from -bV.
128 PH/05 When a filter generated an autoreply, the entire To: header line was
129 quoted in the delivery log line, like this:
131 => >A.N.Other <ano@some.domain> <original@ddress> ...
133 This has been changed so that it extracts the operative address. There
134 may be more than one such address. If so, they are comma-separated, like
137 => >ano@some.domain,ona@other.domain <original@ddress> ...
139 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
140 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
141 not being looked up in the rDNS to get a host name, Exim was showing the
142 IP address twice in Received: lines, even though the IP addresses were
143 identical. For example:
145 Received: from [1.2.3.4] (helo=[1.2.3.4])
147 However, if the real host name was known, it was omitting the HELO data
148 if it matched the actual IP address. This has been tidied up so that it
149 doesn't show the same IP address twice.
151 PH/07 When both +timestamp and +memory debugging was on, the value given by
152 $tod_xxx expansions could be wrong, because the tod_stamp() function was
153 called by the debug printing, thereby overwriting the timestamp buffer.
154 Debugging no longer uses the tod_stamp() function when +timestamp is set.
156 PH/08 When the original message was included in an autoreply transport, it
157 always said "this is a copy of the message, including all the headers",
158 even if body_only or headers_only was set. It now gives an appropriate
161 PH/09 Applied a patch from the Sieve maintainer which:
163 o fixes some comments
164 o adds the (disabled) notify extension core
165 o adds some debug output for the result of if/elsif tests
166 o points to the current vacation draft in the documentation
167 and documents the missing references header update
171 o fixes a bug in processing the envelope test (when testing
172 multiple envelope elements, the last element determinted the
175 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
176 Electronic Mail") by including:
178 Auto-submitted: auto-generated
180 in the messages that it generates (bounce messages and others, such as
181 warnings). In the case of bounce messages for non-SMTP mesages, there was
182 also a typo: it was using "Auto_submitted" (underscore instead of
183 hyphen). Since every message generated by Exim is necessarily in response
184 to another message, thes have all been changed to:
186 Auto-Submitted: auto-replied
188 in accordance with these statements in the RFC:
190 The auto-replied keyword:
192 - SHOULD be used on messages sent in direct response to another
193 message by an automatic process,
195 - MUST NOT be used on manually-generated messages,
197 - MAY be used on Delivery Status Notifications (DSNs) and Message
198 Disposition Notifications (MDNs),
200 - MUST NOT be used on messages generated by automatic or periodic
201 processes, except for messages which are automatic responses to
204 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
205 to the default Received: header definition.
207 PH/12 Added log selector acl_warn_skipped (default on).
209 PH/13 After a successful wildlsearch lookup, discard the values of numeric
210 variables because (a) they are in the wrong storage pool and (b) even if
211 they were copied, it wouldn't work properly because of the caching.
213 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
214 checking when decoding. Apparently there are clients that generate
215 overlong encoded strings. Why am I not surprised?
217 PH/15 If the first argument of "${if match_address" was not empty, but did not
218 contain an "@" character, Exim crashed. Now it writes a panic log message
219 and treats the condition as false.
221 PH/16 In autoreply, treat an empty string for "once" the same as unset.
223 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
224 extension "envelope-auth". The code is finished and in agreement with
225 other implementations, but there is no documentation so far and in fact,
226 nobody wrote the draft yet. This extension is currently #undef'ed, thus
227 not changing the active code.
229 Print executed "if" and "elsif" statements when debugging is used. This
230 helps a great deal to understand what a filter does.
232 Document more things not specified clearly in RFC3028. I had all this
233 sorted out, when out of a sudden new issues came to my mind. Oops."
235 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
238 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
239 canonical form (as documented). However, after a host name lookup from
240 the IP address, check_host() was doing a simple string comparison with
241 addresses acquired from the DNS when checking that the found name did
242 have the original IP as one of its addresses. Since any found IPv6
243 addresses are likely to be in abbreviated form, the comparison could
244 fail. Luckily, there already exists a function for doing the comparison
245 by converting both addresses to binary, so now that is used instead of
248 PH/20 There was another similar case to PH/19, when a complete host name was
249 given in a host list; looking up its IP address could give an abbreviated
250 form, whereas the current host's name might or might not be abbreviated.
251 The same fix has been applied.
257 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
258 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
261 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
262 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
264 PH/03 Typo: missing ".o" in src/pcre/Makefile.
266 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
267 header line, restrict the check to what is listed in RFCs 2369 and 2929.
268 Also, for "Auto-Submitted", treat anything other than "no" as
269 non-personal, in accordance with RFC 3834. (Previously it treated
270 anything starting "auto-" as non-personal.)
272 TF/01 The control=submission/name=... option had a problem with syntax
273 errors if the name included a slash character. The /name= option
274 now slurps the rest of the string, so it can include any characters
275 but it must come last in the list of options (after /sender_retain
278 PH/05 Some modifications to the interface to the fake nameserver for the new
286 TK/01 Added the "success_on_redirect" address verification option. See
287 NewStuff for rationale and an example.
289 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
291 PH/02 Patch to exigrep to allow it to work on syslog lines.
293 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
294 fread() to skip over the body file's header line, because in Cygwin the
295 header line is locked and is inaccessible.
297 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
298 co-exist for some time) to make it clear that it is the Exim ID that is
299 referenced, not the Message-ID: header line.
301 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
302 string_format() function, because snprintf() does not exist on all
305 PH/06 The use of forbid_filter_existstest now also locks out the use of the
306 ${stat: expansion item.
308 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
309 protocol synchronization error", to keep the pedants happy.
311 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
312 well as for IRIX systems, when gcc is being used. See the host.c source
315 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
317 PH/10 Named domain lists were not working if used in a queue_smtp_domains
320 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
321 transport and to the smtp transport in LMTP mode.
323 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
325 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
326 run a filter in a subprocess. This could lead to confusion in subsequent
327 lookups in the parent process. There should also be a search_tidyup() at
328 the end of the subprocess.
330 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
331 only if the host matched helo_try_verify_hosts, which caused the
332 verification to occur when the EHLO/HELO command was issued. The ACL just
333 tested the remembered result. Now, if a previous verification attempt has
334 not happened, "verify = helo" does it there and then.
336 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
338 TK/03 Fix log output including CR from clamd.
340 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
341 exist provoked a memory error which could cause a segfault.
343 PH/15 Installed PCRE 6.2
345 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
347 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
348 of the problem. Specifically, suggested +O2 rather than +O1 for the
351 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
353 PH/20 If a delivery was routed to a non-standard port by means of an SRV
354 record, the port was not correctly logged when the outgoing_port log
355 selector was set (it logged the transort's default port).
357 PH/21 Added support for host-specific ports to manualroute, queryprogram,
358 fallback_hosts, and "hosts" in the smtp transport.
360 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
361 host errors such as "Connection refused".
363 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
364 authentication with radiusclient 0.4.9:
366 - Error returned from rc_read_config was caught wrongly
367 - Username/password not passed on to radius server due to wrong length.
369 The presumption is that some radiusclient API changes for 4.51/PH/17
370 were not taken care of correctly. The code is still untested by me (my
371 Linux distribution still has 0.3.2 of radiusclient), but it was
372 contributed by a Radius user.
374 PH/24 When doing a callout, the value of $domain wasn't set correctly when
375 expanding the "port" option of the smtp transport.
377 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
378 while reading a MIME header. Thanks to Tom Hughes for a patch.
380 PH/24 Include config.h inside local_scan.h so that configuration settings are
383 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
384 that in an ACL for RCPT (for example), you can examine exactly what was
387 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
388 commands, but it was not correctly comparing the address with the actual
389 client host address. Thus, it would show the EHLO address in Received:
390 header lines when this was not necessary.
392 PH/27 Added the % operator to ${eval:}.
394 PH/28 Exim tries to create and chdir to its spool directory when it starts;
395 it should be ignoring failures (because with -C, for example, it has lost
396 privilege). It wasn't ignoring creation failures other than "already
399 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
402 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
403 because an input line was too long, either on its own, or by virtue of
404 too many continuations, the temporary file was not being removed, and the
405 return code was incorrect.
407 PH/31 Missing "BOOL" in function definition in filtertest.c.
409 PH/32 Applied Sieve patches from the maintainer.
411 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
413 PH/33 Added "verify = not_blind".
415 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
416 Local/Makefile (with some defaults set). These are used in built scripts
417 such as exicyclog, but they have never been used in the exim_install
418 script (though there are many overriding facilities there). I have
419 arranged that the exim_install script now takes note of these two
422 PH/35 Installed configuration files for Dragonfly.
424 PH/36 When a locally submitted message by a trusted user did not contain a
425 From: header, and the sender address was obtained from -f or from an SMTP
426 MAIL command, and the trusted user did not use -F to supply a sender
427 name, $originator_name was incorrectly used when constructing a From:
428 header. Furthermore, $originator_name was used for submission mode
429 messages from external hosts without From: headers in a similar way,
430 which is clearly wrong.
432 PH/37 Added control=suppress_local_fixups.
434 PH/38 When log_selector = +received_sender was set, and the addition of the
435 sender made the log line's construction buffer exactly full, or one byte
436 less than full, an overflow happened when the terminating "\n" was
439 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
440 when the result of a list match is failure because a DNS lookup failed.
442 PH/40 RM_COMMAND is now used in the building process.
444 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
445 the "build-* directories that it finds.
447 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
448 address in a domain literal was a prefix of an interface address.
450 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
451 when verifying a sender address, unless rewrite_headers is false.
453 PH/44 Wrote a long comment about why errors_to addresses are verified as
454 recipients, not senders.
456 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
457 the ratelimit ACL was added.
459 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
461 PH/46 Added extra information about PostgreSQL errors to the error string.
463 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
464 avoiding the need to install special zones in a real server. This is
465 backwards compatible; if it can't find the fake resolver, it drops back.
466 Thus, both old and new test suites can be run.
468 TF/02 Added util/ratelimit.pl
470 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
471 clock is set back in time.
473 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
474 Candler <B.Candler@pobox.com>.
476 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
477 OK for addresses that are the result of redirections.
479 PH/48 A number of further additions for the benefit of the new test suite,
480 including a fake gethostbyname() that interfaces to the fake DNS resolver
483 TF/06 The fix for widen_domains has also been applied to qualify_single and
484 search_parents which are the other dnslookup options that can cause
487 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
490 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
492 TF/07 Exim produced the error message "an SRV record indicated no SMTP
493 service" if it encountered an MX record with an empty target hostname.
494 The message is now "an MX or SRV record indicated no SMTP service".
496 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
497 if the DNS of the sending site is misconfigured. This is quite a
498 common situation. This change restores the behaviour of treating a
499 helo verification defer as a failure.
501 PH/51 If self=fail was set on a router, the bounce message did not include the
502 actual error message.
508 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
510 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
511 command itself ended in error, the underlying message about the transport
512 filter timeout was being overwritten with the pipe command error. Now the
513 underlying error message should be appended to the second error message.
515 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
517 PH/02 Reduce the amount of output that "make" produces by default. Full output
518 can still be requested.
520 PH/03 The warning log line about a condition test deferring for a "warn" verb
521 was being output only once per connection, rather than after each
522 occurrence (because it was using the same function as for successful
523 "warn" verbs). This seems wrong, so I have changed it.
525 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
526 it should not have, which might have caused a crash in the right
527 circumstances, but probably never did.
529 PH/04 Installed a modified version of Tony Finch's patch to make submission
530 mode fix the return path as well as the Sender: header line, and to
531 add a /name= option so that you can make the user's friendly name appear
534 TF/03 Added the control = fakedefer ACL modifier.
536 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
537 Mark Lowes for thorough testing.
539 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
541 TK/03 Merged latest SRS patch from Miles Wilton.
543 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
544 with the definition in sysexits.h (which is #included earlier).
545 Fortunately, Exim does not actually use EX_OK. The code used to try to
546 preserve the sysexits.h value, by assumimg that macro definitions were
547 scanned for macro replacements. I have been disabused of this notion,
548 so now the code just undefines EX_OK before #including unistd.h.
550 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
551 in the smtp transport. When a block could not be written in a single
552 write() function, the timeout was being re-applied to each part-write.
553 This seems wrong - if the receiver was accepting one byte at a time it
554 would take for ever. The timeout is now adjusted when this happens. It
555 doesn't have to be particularly precise.
557 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
558 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
560 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
561 without a domain if the check to <postmaster@domain> fails.
563 SC/01 Eximstats: added -xls and the ability to specify output files
564 (patch written by Frank Heydlauf).
566 SC/02 Eximstats: use FileHandles for outputing results.
568 SC/03 Eximstats: allow any combination of xls, txt, and html output.
570 SC/04 Eximstats: fixed display of large numbers with -nvr option
572 SC/05 Eximstats: fixed merging of reports with empty tables.
574 SC/06 Eximstats: added the -include_original_destination flag
576 SC/07 Eximstats: removed tabs and trailing whitespace.
578 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
580 TK/06 MBOX spool code: Add real "From " MBOX separator line
581 so the .eml file is really in mbox format (even though
582 most programs do not really care). Patch from Alex Miller.
584 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
585 The latter is generated from $received_to and is only set if the
586 message has one envelope recipient. SA can use these headers,
587 obviously out-of-the-box. Patch from Alex Miller.
589 PH/08 The ${def test on a variable was returning false if the variable's
590 value was "0", contrary to what the specification has always said!
591 The result should be true unless the variable is empty.
593 PH/09 The syntax error of a character other than { following "${if
594 def:variable_name" (after optional whitespace) was not being diagnosed.
595 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
596 accidental colon was present, for example, could give incorrect results.
598 PH/10 Tidied the code in a number of places where the st_size field of a stat()
599 result is used (not including appendfile, where other changes are about
602 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
603 This involved changing a lot of size variables from int to off_t. It
604 should work with maildirs and everything.
606 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
607 spamd dying while we are connected to it.
609 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
612 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
613 function for BDB 4.3.
615 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
617 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
618 the sources for PCRE have been re-arranged and more files are now
621 PH/15 The code I had for printing potentially long long variables in PH/11
622 above was not the best (it lost precision). The length of off_t variables
623 is now inspected at build time, and an appropriate printing format (%ld
624 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
625 to be "long long int" or "long int". This is needed for the internal
626 formatting function string_vformat().
628 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
629 the configuration file to be ":syslog", then the script "guesses" where
630 the logs files are, rather than using the compiled in default. In our
631 case the guess is not the same as the compiled default, so the script
632 suddenly stopped working when I started to use syslog. The patch checks
633 to see if log_file_path is "". If so, it attempts to read it from exim
634 with no configuration file to get the compiled in version, before it
635 falls back to the previous guessing code."
637 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
638 implementing BATV in an Exim configuration. See NewStuff for the gory
641 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
642 Makefile that are specific to HP-UX.
644 PH/18 If the "use_postmaster" option was set for a recipient callout together
645 with the "random" option, the postmaster address was used as the MAIL
646 FROM address for the random test, but not for the subsequent recipient
647 test. It is now used for both.
649 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
650 patch removes a few documentation additions to RFC 3028, because the
651 latest draft now contains them. It adds the new en;ascii-case comparator
652 and a new error check for 8bit text in MIME parts. Comparator and
653 require names are now matched exactly. I enabled the subaddress
654 extension, but it is not well tested yet (read: it works for me)."
656 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
657 rework some of the code of TK/09 above to avoid the hardwired use of
658 "%lld" and "long long". Replaced the call to snprintf() with a call to
661 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
662 records point to non-existent hosts", "retry timeout exceeded", and
663 "retry time not reached for any host after a long failure period".
665 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
666 experimental DomainKeys support:
668 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
669 (2) On an error such as an illegally used "control", the wrong name for
670 the control was given.
672 These problems did NOT occur unless DomainKeys support was compiled.
674 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
676 PH/24 Added ${if match_ip condition.
678 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
679 databases so that it will be absolutely obvious if a crash occurs in the
680 DB library. This is a regular occurrence (often caused by mis-matched
683 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
684 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
685 picked up on a user's system that detects such things. There doesn't seem
686 to be a gcc warning option for this - only an attribute that has to be
687 put on the function's prototype. It seems that in Fedora Core 4 they have
688 set this on a number of new functions. No doubt there will be more in due
691 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
692 specify a transport. However, if an address that was verified by such a
693 router was the subject of a callout, Exim crashed because it tried to
694 read the rcpt_include_affixes from the non-existent transport. Now it
695 just assumes that the setting of that option is false. This bug was
696 introduced by 4.51/PH/31.
698 PH/28 Changed -d+all to exclude +memory, because that information is very
699 rarely of interest, but it makes the output a lot bigger. People tend to
700 do -d+all out of habit.
702 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
703 code in os-type was giving problems when libc.so lives in lib64, like on
706 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
707 aren't the modern standard, and it seems that some systems' include files
708 don't always have them. Exim was already checking for some of the newer
709 ones like T_AAAA, and defining it itself. I've added checks for all the
710 record types that Exim uses.
712 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
713 not automatically generating a new one, as it is supposed to. This
714 prevented TLS from working. If the file did exist, but contained invalid
715 data, a new version was generated, as expected. It was only the case of a
716 non-existent file that was broken.
718 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
719 with a change in libdomainkeys > 0.64.
721 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
722 from DNS. If the selector record carries the flag, it now has
723 precedence over the domain-wide flag.
725 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
727 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
728 the use of a port name that isn't defined in /etc/services) occurred, the
729 message was deferred as in a normal delivery, and thus remained on the
730 spool, instead of being failed because of the mua_wrapper setting. This
731 is now fixed, and I tidied up some of the mua_wrapper messages at the
734 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
735 the messages in a hash of arrays rather than using individual hashes.
736 This is a bit cleaner and results in dramatic memory savings, albeit
737 at a slight CPU cost.
739 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
740 as requested by Marc Sherman.
742 SC/10 Eximstats: added histograms for user specified patterns as requested
745 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
747 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
748 fopen() in the content-scanning modules that did not already have it.
754 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
755 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
757 TK/02 Fix ACL "control" statement not being available in MIME ACL.
759 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
761 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
762 to test Sieve filters that use "vacation".
764 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
765 that changes the way the GnuTLS parameters are stored in the cache file.
766 The new format can be generated externally. For backward compatibility,
767 if the data in the cache doesn't make sense, Exim assumes it has read an
768 old-format file, and it generates new data and writes a new file. This
769 means that you can't go back to an older release without removing the
772 PH/03 A redirect router that has both "unseen" and "one_time" set does not
773 work if there are any delivery delays because "one_time" forces the
774 parent to be marked "delivered", so its unseen clone is never tried
775 again. For this reason, Exim now forbids the simultaneous setting of
778 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
779 redirected to themselves ("homonym" addresses). Read the long ChangeLog
780 entry if you want to know the details. The fix, however, neglected to
781 consider the case when local delivery batching is involved. The test for
782 "previously delivered" was not happening when checking to see if an
783 address could be batched with a previous (undelivered) one; under
784 certain circumstances this could lead to multiple deliveries to the same
787 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
788 in its include files, and this causes problems building Exim.
790 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
791 header_syntax) but Exim was just ignoring anything given after a slash.
792 In particular, this caused confusion with an attempt to use "verify =
793 reverse_host_lookup/defer_ok". An error is now given when options are
794 supplied for verify items that do not have them. (Maybe reverse_host_
795 lookup should have a defer_ok option, but that's a different point.)
797 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
798 defined by RFC 821) to 2048, because there were problems with some AUTH
799 commands, and RFC 1869 says the size should be increased for extended
800 SMTP commands that take arguments.
802 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
805 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
806 "unknown" error; now it says that the functionality isn't in the binary.
808 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
809 an address' error message when a string expansion fails (syntax or
810 whatever). Otherwise the password may appear in the log. Following change
811 PH/42 below, there is no longer a chance of it appearing in a bounce
814 PH/11 Installed exipick version 20050225.0 from John Jetmore.
816 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
817 of its addresses was ever tried. (Bugzilla bug #2.)
819 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
820 the result incorrectly in the debug output. (It correctly added a newline
821 to what was transported.)
823 TF/01 Added $received_time.
825 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
826 commented out examples of how to interface to a virus scanner and to
827 SpamAssassin. Also added commented examples of av_scanner and
828 spamd_address settings.
830 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
831 and controls are allowed in which ACLs. There were a couple of minor
832 errors. Some of the entries in the conditions table (which is a table of
833 where they are NOT allowed) were getting very unwieldy; rewrote them as a
834 negation of where the condition IS allowed.
836 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
838 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
839 header file does not have a version number, so I've had to invent a new
840 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
841 API. The code is untested by me (my Linux distribution still has 0.3.2 of
842 radiusclient), but it was contributed by a Radius user.
844 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
845 files or directories, for OpenSSL.
847 PH/19 When an Exim process that is running as root has to create an Exim log
848 file, it does so in a subprocess that runs as exim:exim so as to get the
849 ownership right at creation (otherwise, other Exim processes might see
850 the file with the wrong ownership). There was no test for failure of this
851 fork() call, which would lead to the process getting stuck as it waited
852 for a non-existent subprocess. Forks do occasionally fail when resources
853 run out. I reviewed all the other calls to fork(); they all seem to check
856 PH/20 When checking for unexpected SMTP input at connect time (before writing
857 the banner), Exim was not dealing correctly with a non-positive return
858 from the read() function. If the client had disconnected by this time,
859 the result was a log entry for a synchronization error with an empty
860 string after "input=" when read() returned zero. If read() returned -1
861 (an event I could not check), uninitialized data bytes were printed.
862 There were reports of junk text (parts of files, etc) appearing after
865 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
867 PH/22 Added support for macro redefinition, and (re)definition in between
868 driver and ACL definitions.
870 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
871 forgetting to use the resulting value; it was using the unexpanded value.
873 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
874 hadn't been configured. The fix is from Juergen Kreileder, who
875 understands it better than I do:
877 "Here's what I see happening with three configured cyrus_sasl
878 authenticators configured (plain, login, cram-md5):
880 On startup auth_cyrus_sasl_init() gets called for each of these.
881 This means three calls to sasl_listmech() without a specified mech_list.
882 => SASL tests which mechs of all available mechs actually work
883 => three warnings about OTP not working
884 => the returned list contains: plain, login, cram-md5, digest-md5, ...
886 With the patch, sasl_listmech() also gets called three times. But now
887 SASL's mech_list option is set to the server_mech specified in the the
888 authenticator. Or in other words, the answer from sasl_listmech()
889 gets limited to just the mech you're testing for (which is different
891 => the return list contains just 'plain' or 'login', 'cram-md5' or
892 nothing depending on the value of ob->server_mech.
894 I've just tested the patch: Authentication still works fine,
895 unavailable mechs specified in the exim configuration are still
896 caught, and the auth.log warnings about OTP are gone."
898 PH/25 When debugging is enabled, the contents of the command line are added
899 to the debugging output, even when log_selector=+arguments is not
902 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
903 answer is "GNU", and only if the return is "GNU/something" is the answer
906 PH/27 $acl_verify_message is now set immediately after the failure of a
907 verification in an ACL, and so is available in subsequent modifiers. In
908 particular, the message can be preserved by coding like this:
910 warn !verify = sender
911 set acl_m0 = $acl_verify_message
913 Previously, $acl_verify_message was set only while expanding "message"
914 and "log_message" when a very denied access.
916 PH/28 Modified OS/os.c-Linux with
918 -#ifndef OS_LOAD_AVERAGE
919 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
921 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
922 nomenclature these days.)
924 PH/29 Installed patch from the Sieve maintainer that adds the options
925 sieve_useraddress and sieve_subaddress to the redirect router.
927 PH/30 In these circumstances:
928 . Two addresses routed to the same list of hosts;
929 . First host does not offer TLS;
930 . First host accepts first address;
931 . First host gives temporary error to second address;
932 . Second host offers TLS and a TLS session is established;
933 . Second host accepts second address.
934 Exim incorrectly logged both deliveries with the TLS parameters (cipher
935 and peerdn, if requested) that were in fact used only for the second
938 PH/31 When doing a callout as part of verifying an address, Exim was not paying
939 attention to any local part prefix or suffix that was matched by the
940 router that accepted the address. It now behaves in the same way as it
941 does for delivery: the affixes are removed from the local part unless
942 rcpt_include_affixes is set on the transport.
944 PH/32 Add the sender address, as F=<...>, to the log line when logging a
945 timeout during the DATA phase of an incoming message.
947 PH/33 Sieve envelope tests were broken for match types other than :is. I have
948 applied a patch sanctioned by the Sieve maintainer.
950 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
951 the uid or gid is negative. A case of a negative gid caused this to be
952 noticed. The fix allows for either to be negative.
954 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
955 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
958 PH/36 The change PH/12 above was broken. Fixed it.
960 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
961 the grounds that routing the same address twice would always produce the
962 same answer. This might have been true once, but it is certainly no
963 longer true now. Routing a child address may depend on the previous
964 routing that produced that child. Some complicated redirection strategies
965 went wrong when messages had multiple recipients, and made Exim's
966 behaviour dependent on the order in which the addresses were given.
968 I have moved the duplicate checking until after the routing is complete.
969 Exim scans the addresses that are assigned to local and remote
970 transports, and removes any duplicates. This means that more work will be
971 done, as duplicates will always all be routed, but duplicates are
972 presumably rare, so I don't expect this is of any significance.
974 For deliveries to pipes, files, and autoreplies, the duplicate checking
975 still happens during the routing process, since they are not going to be
978 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
979 It corrects a timeout issue with spamd. This is Ian's comment: "The
980 background is that sometimes spamd either never reads data from a
981 connection it has accepted, or it never writes response data. The exiscan
982 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
983 blindly assumes that writes won't block so it may never time out."
985 PH/39 Allow G after quota size as well as K and M.
987 PH/40 The value set for $authenticated_id in an authenticator may not contain
988 binary zeroes or newlines because the value is written to log lines and
989 to spool files. There was no check on this. Now the value is run through
990 the string_printing() function so that such characters are converted to
991 printable escape sequences.
993 PH/41 $message_linecount is a new variable that contains the total number of
994 lines in the message. Compare $body_linecount, which is the count for the
997 PH/42 Exim no longer gives details of delivery errors for specific addresses in
998 bounce and delay warning messages, except in certain special cases, which
1001 (a) An SMTP error message from a remote host;
1002 (b) A message specified in a :fail: redirection;
1003 (c) A message specified in a "fail" command in a system filter;
1004 (d) A message specified in a FAIL return from the queryprogram router;
1005 (e) A message specified by the cannot_route_message router option.
1007 In these cases only, Exim does include the error details in bounce and
1008 warning messages. There are also a few cases where bland messages such
1009 as "unrouteable address" or "local delivery error" are given.
1011 PH/43 $value is now also set for the "else" part of a ${run expansion.
1013 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1014 being worked on, but at least Exim now implements the latest version to
1017 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1018 process to complete was treated as a delivery failure, a timeout while
1019 writing the message to the pipe was logged, but erroneously treated as a
1020 successful delivery. Such timeouts include transport filter timeouts. For
1021 consistency with the overall process timeout, these timeouts are now
1022 treated as errors, giving rise to delivery failures by default. However,
1023 there is now a new Boolean option for the pipe transport called
1024 timeout_defer, which, if set TRUE, converts the failures into defers for
1025 both kinds of timeout. A transport filter timeout is now identified in
1028 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1029 systems where "make" and "gmake" are different, calling "gmake" at top
1030 level broke things. I've arranged for the value of $(MAKE) to be passed
1031 from the Makefile to this script so that it can call the same version of
1035 A note about Exim versions 4.44 and 4.50
1036 ----------------------------------------
1038 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1039 changes of various kinds. As a consequence, a big documentation update was
1040 needed. This delayed the release for rather longer than seemed good, especially
1041 in the light of a couple of (minor) security issues. Therefore, the changes
1042 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1043 release. So 4.44 and 4.50 are in effect two different branches that both start
1046 I have left the 4.50 change log unchanged; it contains all the changes since
1047 4.43. The change log for 4.44 is below; many of its items are identical to
1048 those for 4.50. This seems to be the most sensible way to preserve the
1049 historical information.
1055 1. Minor wording change to the doc/README.SIEVE file.
1057 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1058 computation of the current number of files was incorrect.
1060 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1061 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1064 4. Give more explanation in the error message when the command for a transport
1065 filter fails to execute.
1067 5. There are several places where Exim runs a non-Exim command in a
1068 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1069 was being done only for the command run by the queryprogram router. It is
1070 now done for all such subprocesses. The other cases are: ${run, transport
1071 filters, and the commands run by the lmtp and pipe transports.
1073 6. Added CONFIGURE_GROUP build-time option.
1075 7. Some older OS have a limit of 256 on the maximum number of file
1076 descriptors. Exim was using setrlimit() to set 1000 as a large value
1077 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1078 systems. I've change it so that if it can't get 1000, it tries for 256.
1080 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1081 was an oversight, and furthermore, ever since the addition of extra
1082 controls (e.g. 4.43/32), the checks on when to allow different forms of
1083 "control" were broken. There should now be diagnostics for all cases when a
1084 control that does not make sense is encountered.
1086 9. Added the /retain_sender option to "control=submission".
1088 10. $recipients is now available in the predata ACL (oversight).
1090 11. Tidy the search cache before the fork to do a delivery from a message
1091 received from the command line. Otherwise the child will trigger a lookup
1092 failure and thereby defer the delivery if it tries to use (for example) a
1093 cached ldap connection that the parent has called unbind on.
1095 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1096 of $address_data from the recipient verification was clobbered by the
1097 sender verification.
1099 13. The value of address_data from a sender verification is now available in
1100 $sender_address_data in subsequent conditions in the ACL statement.
1102 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1104 15. Added a new option "connect=<time>" to callout options, to set a different
1107 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1108 was its contents. (It was OK if the option was not defined at all.)
1110 17. A "Completed" log line is now written for messages that are removed from
1111 the spool by the -Mrm option.
1113 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1114 information about exactly what failed.
1116 19. Added -dd to debug only the daemon process.
1118 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1119 handles timeouts, both on the server side and network timeouts. Renamed the
1120 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1122 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1123 It is now set to "smtps".
1125 22. $host_address is now set to the target address during the checking of
1126 ignore_target_hosts.
1128 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1129 being passed; this would have caused $sender_host_name to have been used if
1130 matching the list had actually called for a host name (not very likely,
1131 since this list is usually IP addresses). A host name is now passed as
1134 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1135 code that specifies a non-restarting handler (typically sigaction() in
1136 modern systems) in an attempt to fix a rare and obscure crash bug.
1138 25. Narrowed the window for a race in the daemon that could cause it to ignore
1139 SIGCHLD signals. This is not a major problem, because they are used only to
1140 wake it up if nothing else does.
1142 26. A malformed maildirsize file could cause Exim to calculate negative values
1143 for the mailbox size or file count. Odd effects could occur as a result.
1144 The maildirsize information is now recalculated if the size or filecount
1147 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1148 support for a long time. Removed HAVE_SYS_VFS_H.
1150 28. Installed the latest version of exipick from John Jetmore.
1152 29. In an address list, if the pattern was not a regular expression, an empty
1153 subject address (from a bounce message) matched only if the pattern was an
1154 empty string. Non-empty patterns were not even tested. This was the wrong
1155 because it is perfectly reasonable to use an empty address as part of a
1156 database query. An empty address is now tested by patterns that are
1157 lookups. However, all the other forms of pattern expect the subject to
1158 contain a local part and a domain, and therefore, for them, an empty
1159 address still always fails if the pattern is not itself empty.
1161 30. Exim went into a mad DNS loop when attempting to do a callout where the
1162 host was specified on an smtp transport, and looking it up yielded more
1163 than one IP address.
1165 31. Re-factored the code for checking spool and log partition space into a
1166 function that finds that data and another that does the check. The former
1167 is then used to implement four new variables: $spool_space, $log_space,
1168 $spool_inodes, and $log_inodes.
1170 32. The RFC2047 encoding function was originally intended for short strings
1171 such as real names; it was not keeping to the 75-character limit for
1172 encoded words that the RFC imposes. It now respects the limit, and
1173 generates multiple encoded words if necessary. To be on the safe side, I
1174 have increased the buffer size for the ${rfc2047: expansion operator from
1177 33. It is now permitted to omit both strings after an "if" condition; if the
1178 condition is true, the result is "true". As before, when the second string
1179 is omitted, a false condition yields an empty string. This makes it less
1180 cumbersome to write custom ACL and router conditions.
1182 34. Failure to deliver a bounce message always caused it to be frozen, even if
1183 there was an errors_to setting on the router. The errors_to setting is now
1186 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1187 canonical form (fully expanded) before being placed in
1188 $sender_host_address.
1190 36. The table in the code that translates DNS record types into text (T_A to
1191 "A" for instance) was missing entries for NS and CNAME. It is just possible
1192 that this could have caused confusion if both these types were looked up
1193 for the same domain, because the text type is used as part of Exim's
1194 per-process caching. But the chance of anyone hitting this buglet seems
1197 37. The dnsdb lookup has been extended in a number of ways.
1199 (1) There is a new type, "zns", which walks up the domain tree until it
1200 finds some nameserver records. It should be used with care.
1202 (2) There is a new type, "mxh", which is like "mx" except that it returns
1203 just the host names, not the priorities.
1205 (3) It is now possible to give a list of domains (or IP addresses) to be
1206 looked up. The behaviour when one of the lookups defers can be
1207 controlled by a keyword.
1209 (4) It is now possible to specify the separator character for use when
1210 multiple records are returned.
1212 38. The dnslists ACL condition has been extended: it is now possible to supply
1213 a list of IP addresses and/or domains to be looked up in a particular DNS
1216 39. Added log_selector=+queue_time_overall.
1218 40. When running the queue in the test harness, wait just a tad after forking a
1219 delivery process, to get repeatability of debugging output.
1221 41. Include certificate and key file names in error message when GnuTLS fails
1222 to set them up, because the GnuTLS error message doesn't include the name
1223 of the failing file when there is a problem reading it.
1225 42. Allow both -bf and -bF in the same test run.
1227 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1229 44. The "Exiscan patch" is now merged into the mainline Exim source.
1231 45. Sometimes the final signoff response after QUIT could fail to get
1232 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1233 < 0 before doing a fflush(). This bug looks as though it goes back to the
1234 introduction of TLS in release 3.20, but "sometimes" must have been rare
1235 because the tests only now provoked it.
1237 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1238 (this can affect the format of dates).
1240 47. exim_tidydb, when checking for the continued existence of a message for
1241 which it has found a message-specific retry record, was not finding
1242 messages that were in split spool directories. Consequently, it was
1243 deleting retry records that should have stayed in existence.
1245 48. Steve fixed some bugs in eximstats.
1247 49. The SPA authentication driver was not abandoning authentication and moving
1248 on to the next authenticator when an expansion was forced to fail,
1249 contradicting the general specification for all authenticators. Instead it
1250 was generating a temporary error. It now behaves as specified.
1252 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1253 (the order specifies the preference for clients). The order is now AES256,
1254 AES128, 3DES, ARCFOUR128.
1256 51. Small patch to Sieve code - explicitly set From: when generating an
1259 52. Exim crashed if a remote delivery caused a very long error message to be
1260 recorded - for instance if somebody sent an entire SpamAssassin report back
1261 as a large number of 550 error lines. This bug was coincidentally fixed by
1262 increasing the size of one of Exim's internal buffers (big_buffer) that
1263 happened as part of the Exiscan merge. However, to be on the safe side, I
1264 have made the code more robust (and fixed the comments that describe what
1267 53. Now that there can be additional text after "Completed" in log lines (if
1268 the queue_time_overall log selector is set), a one-byte patch to exigrep
1269 was needed to allow it to recognize "Completed" as not the last thing in
1272 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1273 patch that reportedly fixes this has been added. I am not expert enough to
1274 create a test for it. This is what the patch creator wrote:
1276 "I found a little strange behaviour of ldap code when working with
1277 Windows 2003 AD Domain, where users was placed in more than one
1278 Organization Units. When I tried to give exim partial DN, the exit code
1279 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1280 But simultaneously result of request was absolutely normal ldap result,
1281 so I produce this patch..."
1283 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1284 REFERENCE, so I have modified the code to exclude the patch when that macro
1287 55. Some experimental protocols are using DNS PTR records for new purposes. The
1288 keys for these records are domain names, not reversed IP addresses. The
1289 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1290 leaves it alone. Component reversal etc. now happens only for IP addresses.
1293 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1295 57. Double the size of the debug message buffer (to 2048) so that more of very
1296 long debug lines gets shown.
1298 58. The exicyclog utility now does better if the number of log files to keep
1299 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1301 59. Two changes related to the smtp_active_hostname option:
1303 (1) $smtp_active_hostname is now available as a variable.
1304 (2) The default for smtp_banner uses $smtp_active_hostname instead
1305 of $primary_hostname.
1307 60. The host_aton() function is supposed to be passed a string that is known
1308 to be a valid IP address. However, in the case of IPv6 addresses, it was
1309 not checking this. This is a hostage to fortune. Exim now panics and dies
1310 if the condition is not met. A case was found where this could be provoked
1311 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1312 components; fortuitously, this particular loophole had already been fixed
1313 by change 4.50/55 above.
1315 If there are any other similar loopholes, the new check in host_aton()
1316 itself should stop them being exploited. The report I received stated that
1317 data on the command line could provoke the exploit when Exim was running as
1318 exim, but did not say which command line option was involved. All I could
1319 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1320 running as the user.
1323 61. There was a buffer overflow vulnerability in the SPA authentication code
1324 (which came originally from the Samba project). I have added a test to the
1325 spa_base64_to_bits() function which I hope fixes it.
1328 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1329 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1331 63. The daemon start-up calls getloadavg() while still root for those OS that
1332 need the first call to be done as root, but it missed one case: when
1333 deliver_queue_load_max is set with deliver_drop_privilege. This is
1334 necessary for the benefit of the queue runner, because there is no re-exec
1335 when deliver_drop_privilege is set.
1337 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1338 This has been fixed.
1340 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1341 was in use, was not putting the data itself into the right store pool;
1342 consequently, it could be overwritten for a subsequent message in the same
1343 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1346 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1348 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1349 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1350 the function were treating the return as a boolean value, which happened to
1351 work because 0=false and not-0=true, but is not correct code.
1353 68. The host_aton() function was not handling scoped IPv6 addresses (those
1354 with, for example, "%eth0" on the end) correctly.
1356 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1357 negated items (that is, ~something) in unsigned ints. Some compilers
1358 apparently mutter when there is no cast.
1360 70. If an address verification called from an ACL failed, and did not produce a
1361 user-specific message (i.e. there was only a "system" message), nothing was
1362 put in $acl_verify_message. In this situation, it now puts the system
1365 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1366 session; change 4.31/43 added the unwanted input to the log line - except
1367 that it did not do this in the start of session case. It now does.
1369 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1370 This is wrong and can cause the other end to generate a synchronization
1371 error if it is another Exim or anything else that does the synchronization
1372 check. A QUIT command is no longer sent after a timeout.
1374 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1375 during host lookups.
1377 74. The defer_ok option of callout verification was not working if it was used
1378 when verifying addresses in header lines, that is, for this case:
1380 verify = header_sender/callout=defer_ok
1382 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1383 those file descriptors could be used for SMTP connections. If anything
1384 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1385 could be sent to the SMTP client, causing chaos. The daemon now opens
1386 stdin, stdout, and stderr to /dev/null when it puts itself into the
1389 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1390 log by default. The user can override this with suitable Perl magic.
1392 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1393 the log message when discard triggers, was not working for the DATA ACL or
1394 for the non-SMTP ACL.
1396 78. Error message wording change in sieve.c.
1398 79. If smtp_accept_max_per_host was set, the number of connections could be
1399 restricted to fewer than expected, because the daemon was trying to set up
1400 a new connection before checking whether the processes handling previous
1401 connections had finished. The check for completed processes is now done
1402 earlier. On busy systems, this bug wouldn't be noticed because something
1403 else would have woken the daemon, and it would have reaped the completed
1406 80. If a message was submitted locally by a user whose login name contained one
1407 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1408 It caused a spool format error. I have fixed the spool reading code. A
1409 related problem was that the "from" clause in the Received: line became
1410 illegal because of the space(s). It is now covered by ${quote_local_part.
1412 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1415 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1417 83. Patch from Sieve maintainer to fix unterminated string problem in
1418 "vacation" handling.
1420 84. Some minor changes to the Linux configuration files to help with other
1421 OS variants using glibc.
1423 85. One more patch for Sieve to update vacation handling to latest spec.
1426 ----------------------------------------------------
1427 See the note above about the 4.44 and 4.50 releases.
1428 ----------------------------------------------------
1434 1. Change 4.43/35 introduced a bug that caused file counts to be
1435 incorrectly computed when quota_filecount was set in an appendfile
1438 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1439 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1442 3. Give more explanation in the error message when the command for a transport
1443 filter fails to execute.
1445 4. There are several places where Exim runs a non-Exim command in a
1446 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1447 was being done only for the command run by the queryprogram router. It is
1448 now done for all such subprocesses. The other cases are: ${run, transport
1449 filters, and the commands run by the lmtp and pipe transports.
1451 5. Some older OS have a limit of 256 on the maximum number of file
1452 descriptors. Exim was using setrlimit() to set 1000 as a large value
1453 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1454 systems. I've change it so that if it can't get 1000, it tries for 256.
1456 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1457 was an oversight, and furthermore, ever since the addition of extra
1458 controls (e.g. 4.43/32), the checks on when to allow different forms of
1459 "control" were broken. There should now be diagnostics for all cases when a
1460 control that does not make sense is encountered.
1462 7. $recipients is now available in the predata ACL (oversight).
1464 8. Tidy the search cache before the fork to do a delivery from a message
1465 received from the command line. Otherwise the child will trigger a lookup
1466 failure and thereby defer the delivery if it tries to use (for example) a
1467 cached ldap connection that the parent has called unbind on.
1469 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1470 of $address_data from the recipient verification was clobbered by the
1471 sender verification.
1473 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1474 was its contents. (It was OK if the option was not defined at all.)
1476 11. A "Completed" log line is now written for messages that are removed from
1477 the spool by the -Mrm option.
1479 12. $host_address is now set to the target address during the checking of
1480 ignore_target_hosts.
1482 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1483 being passed; this would have caused $sender_host_name to have been used if
1484 matching the list had actually called for a host name (not very likely,
1485 since this list is usually IP addresses). A host name is now passed as
1488 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1489 code that specifies a non-restarting handler (typically sigaction() in
1490 modern systems) in an attempt to fix a rare and obscure crash bug.
1492 15. Narrowed the window for a race in the daemon that could cause it to ignore
1493 SIGCHLD signals. This is not a major problem, because they are used only to
1494 wake it up if nothing else does.
1496 16. A malformed maildirsize file could cause Exim to calculate negative values
1497 for the mailbox size or file count. Odd effects could occur as a result.
1498 The maildirsize information is now recalculated if the size or filecount
1501 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1502 support for a long time. Removed HAVE_SYS_VFS_H.
1504 18. Updated exipick to current release from John Jetmore.
1506 19. Allow an empty sender to be matched against a lookup in an address list.
1507 Previously the only cases considered were a regular expression, or an
1510 20. Exim went into a mad DNS lookup loop when doing a callout where the
1511 host was specified on the transport, if the DNS lookup yielded more than
1514 21. The RFC2047 encoding function was originally intended for short strings
1515 such as real names; it was not keeping to the 75-character limit for
1516 encoded words that the RFC imposes. It now respects the limit, and
1517 generates multiple encoded words if necessary. To be on the safe side, I
1518 have increased the buffer size for the ${rfc2047: expansion operator from
1521 22. Failure to deliver a bounce message always caused it to be frozen, even if
1522 there was an errors_to setting on the router. The errors_to setting is now
1525 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1526 canonical form (fully expanded) before being placed in
1527 $sender_host_address.
1529 24. Updated eximstats to version 1.33
1531 25. Include certificate and key file names in error message when GnuTLS fails
1532 to set them up, because the GnuTLS error message doesn't include the name
1533 of the failing file when there is a problem reading it.
1535 26. Expand error message when OpenSSL has problems setting up cert/key files.
1538 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1539 (this can affect the format of dates).
1541 28. exim_tidydb, when checking for the continued existence of a message for
1542 which it has found a message-specific retry record, was not finding
1543 messages that were in split spool directories. Consequently, it was
1544 deleting retry records that should have stayed in existence.
1546 29. eximstats updated to version 1.35
1547 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1548 1.35 - bugfix such that pie charts by volume are generated correctly
1550 30. The SPA authentication driver was not abandoning authentication and moving
1551 on to the next authenticator when an expansion was forced to fail,
1552 contradicting the general specification for all authenticators. Instead it
1553 was generating a temporary error. It now behaves as specified.
1555 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1556 (the order specifies the preference for clients). The order is now AES256,
1557 AES128, 3DES, ARCFOUR128.
1559 31. Small patch to Sieve code - explicitly set From: when generating an
1562 32. Exim crashed if a remote delivery caused a very long error message to be
1563 recorded - for instance if somebody sent an entire SpamAssassin report back
1564 as a large number of 550 error lines. This bug was coincidentally fixed by
1565 increasing the size of one of Exim's internal buffers (big_buffer) that
1566 happened as part of the Exiscan merge. However, to be on the safe side, I
1567 have made the code more robust (and fixed the comments that describe what
1570 33. Some experimental protocols are using DNS PTR records for new purposes. The
1571 keys for these records are domain names, not reversed IP addresses. The
1572 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1573 leaves it alone. Component reversal etc. now happens only for IP addresses.
1576 34. The host_aton() function is supposed to be passed a string that is known
1577 to be a valid IP address. However, in the case of IPv6 addresses, it was
1578 not checking this. This is a hostage to fortune. Exim now panics and dies
1579 if the condition is not met. A case was found where this could be provoked
1580 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1581 components; fortuitously, this particular loophole had already been fixed
1582 by change 4.50/55 or 4.44/33 above.
1584 If there are any other similar loopholes, the new check in host_aton()
1585 itself should stop them being exploited. The report I received stated that
1586 data on the command line could provoke the exploit when Exim was running as
1587 exim, but did not say which command line option was involved. All I could
1588 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1589 running as the user.
1592 35. There was a buffer overflow vulnerability in the SPA authentication code
1593 (which came originally from the Samba project). I have added a test to the
1594 spa_base64_to_bits() function which I hope fixes it.
1597 36. The daemon start-up calls getloadavg() while still root for those OS that
1598 need the first call to be done as root, but it missed one case: when
1599 deliver_queue_load_max is set with deliver_drop_privilege. This is
1600 necessary for the benefit of the queue runner, because there is no re-exec
1601 when deliver_drop_privilege is set.
1603 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1604 was in use, was not putting the data itself into the right store pool;
1605 consequently, it could be overwritten for a subsequent message in the same
1606 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1609 38. Sometimes the final signoff response after QUIT could fail to get
1610 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1611 < 0 before doing a fflush(). This bug looks as though it goes back to the
1612 introduction of TLS in release 3.20, but "sometimes" must have been rare
1613 because the tests only now provoked it.
1619 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1620 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1621 Now it returns an integer. A number of calls were still expecting a T/F
1622 return. Fortuitously, in all cases, the tests worked in OK situations,
1623 which is the norm. However, things would have gone wrong on any write
1624 failures on the smtp file descriptor. This function is used when sending
1625 messages over SMTP and also when doing verify callouts.
1627 2. When Exim is called to do synchronous delivery of a locally submitted
1628 message (the -odf or -odi options), it no longer closes stderr before doing
1631 3. Implemented the mua_wrapper option.
1633 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1635 5. Implemented the functions header_remove(), header_testname(),
1636 header_add_at_position(), and receive_remove_recipient(), and exported them
1639 6. If an ACL "warn" statement specified the addition of headers, Exim already
1640 inserted X-ACL-Warn: at the start if there was no header name. However, it
1641 was not making this test for the second and subsequent header lines if
1642 there were newlines in the string. This meant that an invalid header could
1643 be inserted if Exim was badly configured.
1645 7. Allow an ACL "warn" statement to add header lines at the start or after all
1646 the Received: headers, as well as at the end.
1648 8. Added the rcpt_4xx retry error code.
1650 9. Added postmaster_mailfrom=xxx to callout verification option.
1652 10. Added mailfrom=xxxx to the callout verification option, for verify=
1655 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1656 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1658 12. Inserted some casts to stop certain compilers warning when using pointer
1659 differences as field lengths or precisions in printf-type calls (mostly
1660 affecting debugging statements).
1662 13. Added optional readline() support for -be (dynamically loaded).
1664 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1665 same clock tick as a message's arrival, so that its received time was the
1666 same as the "first fail" time on the retry record, and that message
1667 remained on the queue past the ultimate address timeout, every queue runner
1668 would try a delivery (because it was past the ultimate address timeout) but
1669 after another failure, the ultimate address timeout, which should have then
1670 bounced the address, did not kick in. This was a "< instead of <=" error;
1671 in most cases the first failure would have been in the next clock tick
1672 after the received time, and all would be well.
1674 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1675 being recognized when the domain list was tested by the match_domain
1676 condition in an expansion string.
1678 16. Added the ${str2b64: operator.
1680 17. Exim was always calling setrlimit() to set a large limit for the number of
1681 processes, without checking whether the existing limit was already
1682 adequate. (It did check for the limit on file descriptors.) Furthermore,
1683 errors from getrlimit() and setrlimit() were being ignored. Now they are
1684 logged to the main and panic logs, but Exim does carry on, to try to do its
1685 job under whatever limits there are.
1687 18. Imported PCRE 5.0.
1689 19. Trivial typo in log message " temporarily refused connection" (the leading
1692 20. If the log selector return_path_on_delivery was set and an address was
1693 redirected to /dev/null, the delivery process crashed because it assumed
1694 that a return path would always be set for a "successful" delivery. In this
1695 case, the whole delivery is bypassed as an optimization, and therefore no
1698 21. Internal re-arrangement: the function for sending a challenge and reading
1699 a response while authentication was assuming a zero-terminated challenge
1700 string. It's now changed to take a pointer and a length, to allow for
1701 binary data in such strings.
1703 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1705 23. Exim was not respecting finduser_retries when seeking the login of the
1706 uid under which it was called; it was always trying 10 times. (The default
1707 setting of finduser_retries is zero.) Also, it was sleeping after the final
1708 failure, which is pointless.
1710 24. Implemented tls_on_connect_ports.
1712 25. Implemented acl_smtp_predata.
1714 26. If the domain in control=submission is set empty, Exim assumes that the
1715 authenticated id is a complete email address when it generates From: or
1716 Sender: header lines.
1718 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1719 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1720 chown and chgrp in /bin and hostname in /usr/bin.
1722 28. Exim was keeping the "process log" file open after each use, just as it
1723 does for the main log. This opens the possibility of it remaining open for
1724 long periods when the USR1 signal hits a daemon. Occasional processlog
1725 errors were reported, that could have been caused by this. Anyway, it seems
1726 much more sensible not to leave this file open at all, so that is what now
1729 29. The long-running daemon process does not normally write to the log once it
1730 has entered its main loop, and it closes the log before doing so. This is
1731 so that log files can straightforwardly be renamed and moved. However,
1732 there are a couple of unusual error situations where the daemon does write
1733 log entries, and I had neglected to close the log afterwards.
1735 30. The text of an SMTP error response that was received during a remote
1736 delivery was being truncated at 512 bytes. This is too short for some of
1737 the long messages that one sometimes sees. I've increased the limit to
1740 31. It is now possible to make retry rules that apply only when a message has a
1741 specific sender, in particular, an empty sender.
1743 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1744 it possible to be selective about when SMTP synchronization is enforced.
1746 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1748 32. Implemented hosts_connection_nolog.
1750 33. Added an ACL for QUIT.
1752 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1755 35. Added mailbox_size and mailbox_filecount to appendfile.
1757 36. Added control = no_multiline_responses to ACLs.
1759 37. There was a bug in the logic of the code that waits for the clock to tick
1760 in the case where the clock went backwards by a substantial amount such
1761 that the microsecond fraction of "now" was more than the microsecond
1762 fraction of "then" (but the whole seconds number was less).
1764 38. Added support for the libradius Radius client library this is found on
1765 FreeBSD (previously only the radiusclient library was supported).
1771 1. When certain lookups returned multiple values in the form name=value, the
1772 quoting of the values was not always being done properly. Specifically:
1773 (a) If the value started with a double quote, but contained no whitespace,
1775 (b) If the value contained whitespace other than a space character (i.e.
1776 tabs or newlines or carriage returns) it was not quoted.
1777 This fix has been applied to the mysql and pgsql lookups by writing a
1778 separate quoting function and calling it from the lookup code. The fix
1779 should probably also be applied to nisplus, ibase and oracle lookups, but
1780 since I cannot test any of those, I have not disturbed their existing code.
1782 2. A hit in the callout cache for a specific address caused a log line with no
1783 reason for rejecting RCPT. Now it says "Previous (cached) callout
1784 verification failure".
1786 3. There was an off-by-one bug in the queryprogram router. An over-long
1787 return line was truncated at 256 instead of 255 characters, thereby
1788 overflowing its buffer with the terminating zero. As well as fixing this, I
1789 have increased the buffer size to 1024 (and made a note to document this).
1791 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1792 when Exim is waiting for an SMTP response from a remote server, Exim
1793 restarts its select() call on the socket, thereby resetting its timeout.
1794 This is not a problem when such interrupts are rare. Somebody set up a cron
1795 job to run exiwhat every 2 minutes, which is less than the normal select()
1796 timeout (5 or 10 minutes). This meant that the select() timeout never
1797 kicked in because it was always reset. I have fixed this by comparing the
1798 time when an interrupt arrives with the time at the start of the first call
1799 to select(). If more time than the timeout has elapsed, the interrupt is
1800 treated as a timeout.
1802 5. Some internal re-factoring in preparation for the addition of Sieve
1803 extensions (by MH). In particular, the "personal" test is moved to a
1804 separate function, and given an option for scanning Cc: and Bcc: (which is
1805 not set for Exim filters).
1807 6. When Exim created an email address using the login of the caller as the
1808 local part (e.g. when creating a From: or Sender: header line), it was not
1809 quoting the local part when it contained special characters such as @.
1811 7. Installed new OpenBSD configuration files.
1813 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1814 try to make them clearer.
1816 9. Callout options, other than the timeout value, were being ignored when
1817 verifying sender addresses in header lines. For example, when using
1819 verify = header_sender/callout=no_cache
1821 the cache was (incorrectly) being used.
1823 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1824 only the Cygwin environment.
1826 11. When return_path_on_delivery was set as a log selector, if different remote
1827 addresses in the same message used different return paths and parallel
1828 remote delivery occurred, the wrong values would sometimes be logged.
1829 (Whenever a remote delivery process finished, the return path value from
1830 the most recently started remote delivery process was logged.)
1832 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1833 lines when AUTH and/or TLS are in use. This is the "received protocol"
1834 field. Exim used to use "asmtp" for authenticated SMTP, without any
1835 indication (in the protocol name) for TLS use. Now it follows the RFC and
1836 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1837 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1838 names appear in log lines as well as in Received: header lines.
1840 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1841 extensions, and comparison tests, and to fix some bugs.
1843 14. Changes to the "personal" filter test:
1845 (1) The test was buggy in that it was just doing the equivalent of
1846 "contains" tests on header lines. For example, if a user's address was
1847 anne@some.where, the "personal" test would incorrectly be true for
1849 To: susanne@some.where
1851 This test is now done by extracting each address from the header in turn,
1852 and checking the entire address. Other tests that are part of "personal"
1853 are now done using regular expressions (for example, to check local parts
1854 of addresses in From: header lines).
1856 (2) The list of non-personal local parts in From: addresses has been
1857 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1858 taken from the Sieve specification recommendations.
1860 (3) If the message contains any header line starting with "List-" it is
1861 treated as non-personal.
1863 (4) The test for "circular" in the Subject: header line has been removed
1864 because it now seems ill-conceived.
1866 15. Minor typos in src/EDITME comments corrected.
1868 16. Installed latest exipick from John Jetmore.
1870 17. If headers_add on a router specified a text string that was too long for
1871 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1872 of string_sprintf() is now avoided.
1874 18. $message_body_size was not set (it was always zero) when running the DATA
1875 ACL and the local_scan() function.
1877 19. For the "mail" command in an Exim filter, no default was being set for
1878 the once_repeat time, causing a random time value to be used if "once" was
1879 specified. (If the value happened to be <= 0, no repeat happened.) The
1880 default is now 0s, meaning "never repeat". The "vacation" command was OK
1881 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1882 (I found it when inspecting the code).
1884 20. There is now an overall timeout for performing a callout verification. It
1885 defaults to 4 times the callout timeout, which applies to individual SMTP
1886 commands during the callout. The overall timeout applies when there is more
1887 than one host that can be tried. The timeout is checked before trying the
1888 next host. This prevents very long delays if there are a large number of
1889 hosts and all are timing out (e.g. when the network connections are timing
1890 out). The value of the overall timeout can be changed by specifying an
1891 additional sub-option for "callout", called "maxwait". For example:
1893 verify = sender/callout=5s,maxwait=20s
1895 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1896 to the end before writing, but this should make it even safer).
1898 22. Exim was forgetting that it had advertised PIPELINING for the second and
1899 subsequent messages on an SMTP connection. It was also not resetting its
1900 memory on STARTTLS and an internal HELO.
1902 23. When Exim logs an SMTP synchronization error within a session, it now
1903 records whether PIPELINING has been advertised or not.
1905 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1906 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1907 rather than long int.
1909 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1911 26. Added the never_mail option to autoreply.
1917 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1918 crash if the getsockname() call failed; this can happen if a connection is
1919 closed very soon after it is established. The problem was simply in the
1920 order in which certain operations were done, causing Exim to try to write
1921 to the SMTP stream before it had set up the file descriptor. The bug has
1922 been fixed by making things happen in the correct order.
1928 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1929 before the connection was closed, thus losing the rejection response.
1931 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1932 some early Solaris releases, but causes trouble in current releases where
1933 socklen_t is defined.
1935 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1938 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1941 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1942 that releasing the top of it at the end releases what was used for sub-
1943 expansions (unless the block got too big). However, discard this block if
1944 the first thing is a variable or header, so that we can use its block when
1945 it is dynamic (useful for very large $message_headers, for example).
1947 6. Lookups now cache *every* query, not just the most recent. A new, separate
1948 store pool is used for this. It can be recovered when all lookup caches are
1949 flushed. Lookups now release memory at the end of their result strings.
1950 This has involved some general refactoring of the lookup sources.
1952 7. Some code has been added to the store_xxx() functions to reduce the amount
1953 of flapping under certain conditions.
1955 8. log_incoming_interface used to affect only the <= reception log lines. Now
1956 it causes the local interface and port to be added to several more SMTP log
1957 lines, for example "SMTP connection from", and rejection lines.
1959 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1961 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1963 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1964 could be overwritten at the end of the current message (or the start of a
1965 new message if it was set in a HELO ACL). The value is now preserved for
1966 the duration of the SMTP connection.
1968 12. If a transport had a headers_rewrite setting, and a matching header line
1969 contained an unqualified address, that address was qualified, even if it
1970 did not match any rewriting rules. The underlying bug was that the values
1971 of the flags that permit the existence of unqualified sender and recipient
1972 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1973 non-local messages, and by -bnq for local messages) were not being
1974 preserved with the message after it was received.
1976 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1977 "next input=" as part of the text comprising the host identity instead of
1978 the correct text. The code was using the same buffer for two different
1979 strings. However, depending on which order the printing function evaluated
1980 its arguments, the bug did not always show up. Under Linux, for example, my
1981 test suite worked just fine.
1983 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1984 doesn't work with some older versions of Perl. It has been changed to "my",
1985 which in any case is probably the better facility to use.
1987 15. A really picky compiler found some instances of statements for creating
1988 error messages that either had too many or two few arguments for the format
1991 16. The size of the buffer for calls to the DNS resolver has been increased
1992 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1993 for addresses that have a lot of PTR records. This alleviates a problem; it
1994 does not fully solve it.
1996 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1997 buffer now truncates the list and logs the incident, which is the same
1998 action as happens when Exim is looking up a host name and its aliases.
1999 Previously in this situation something unpredictable would happen;
2000 sometimes it was "internal error: store_reset failed".
2002 18. If a server dropped the connection unexpectedly when an Exim client was
2003 using GnuTLS and trying to read a response, the client delivery process
2004 crashed while trying to generate an error log message.
2006 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2007 string, for example:
2009 warn message = H1: something\nH2: something
2011 the text was added as a single header line from Exim's point of view
2012 though it ended up OK in the delivered message. However, searching for the
2013 second and subsequent header lines using $h_h2: did not work. This has been
2014 fixed. Similarly, if a system filter added multiple headers in this way,
2015 the routers could not see them.
2017 20. Expanded the error message when iplsearch is called with an invalid key to
2018 suggest using net-iplsearch in a host list.
2020 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2021 ACLs are no longer actually imposed (and a message to that effect is
2024 22. If a "gecos" field in a passwd entry contained escaped characters, in
2025 particular, if it contained a \" sequence, Exim got it wrong when building
2026 a From: or a Sender: header from that name. A second bug also caused
2027 incorrect handling when an unquoted " was present following a character
2028 that needed quoting.
2030 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2031 was not being matched caselessly.
2033 24. Arranged for all hyphens in the exim.8 source to be escaped with
2036 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2037 properly. Recipient callout cache records were still being keyed to include
2038 the sender, even when use_sender was set false. This led to far more
2039 callouts that were necessary. The sender is no longer included in the key
2040 when use_sender is false.
2042 26. Added "control = submission" modifier to ACLs.
2044 27. Added the ${base62d: operator to decode base 62 numbers.
2046 28. dnsdb lookups can now access SRV records.
2048 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2049 the configuration file.
2051 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2052 (-v) mode. This makes the output for a verbose queue run more intelligible.
2054 31. Added a use_postmaster feature to recipient callouts.
2056 32. Added the $body_zerocount variable, containing the number of binary zero
2057 bytes in the message body.
2059 33. The time of last modification of the "new" subdirectory is now used as the
2060 "mailbox time last read" when there is a quota error for a maildir
2063 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2065 35. Added +ignore_unknown as a special item in host lists.
2067 36. Code for decoding IPv6 addresses in host lists is now included, even if
2068 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2069 address was recognized as an IP address, but was then not correctly decoded
2070 into binary, causing unexpected and incorrect effects when compared with
2077 1. Very minor rewording of debugging text in manualroute to say "list of
2078 hosts" instead of "hostlist".
2080 2. If verify=header_syntax was set, and a header line with an unqualified
2081 address (no domain) and a large number of spaces between the end of the
2082 name and the colon was received, the reception process suffered a buffer
2083 overflow, and (when I tested it) crashed. This was caused by some obsolete
2084 code that should have been removed. The fix is to remove it!
2086 3. When running in the test harness, delay a bit after writing a bounce
2087 message to get a bit more predictability in the log output.
2089 4. Added a call to search_tidyup() just before forking a reception process. In
2090 theory, someone could use a lookup in the expansion of smtp_accept_max_
2091 per_host which, without the tidyup, could leave open a database connection.
2093 5. Added the variables $recipient_data and $sender_data which get set from a
2094 lookup success in an ACL "recipients" or "senders" condition, or a router
2095 "senders" option, similar to $domain_data and $local_part_data.
2097 6. Moved the writing of debug_print from before to after the "senders" test
2100 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2101 problems for message scanning, either using a data ACL, or using
2102 local_scan() because the Received: header was not generated till after they
2103 were called (in order to set the time as the time of reception completion).
2104 I have revised the way this works. The header is now generated after the
2105 body is received, but before the ACL or local_scan() are called. After they
2106 are run, the timestamp in the header is updated.
2112 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2113 before starting a queue runner without re-exec. This happened only when
2114 deliver_drop_privilege was set or when the Exim user was set to root. The
2115 effect of the bug was that timeouts during subsequent deliveries caused
2116 crashes instead of being properly handled. The handler is now left at its
2117 default (and expected) setting.
2119 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2120 message, again when deliver_drop_privilege is set or Exim is run as root.
2121 The bug described in (1) was not present in this case, but the tidying up
2122 of the other signals was missing. I have made the two cases consistent.
2124 3. The ignore_target_hosts setting on a manualroute router was being ignored
2125 for hosts that were looked up using the /MX notation.
2127 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2130 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2131 operated on the sender address. After changing the $sender_address to <>
2132 for the sender address verify, Exim was re-instated it as the original
2133 (before rewriting) address, but remembering that it had rewritten it, so it
2134 wasn't rewriting it again. This bug also had the effect of breaking the
2135 sender address verification caching when the sender address was rewritten.
2137 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2138 This has been changed so that if the ip literal address matches
2139 ignore_target_hosts, the router declines.
2141 7. Added expansion conditions match_domain, match_address, and match_local_
2142 part (NOT match_host).
2144 8. The placeholder for the Received: header didn't have a length field set.
2146 9. Added code to Exim itself and to exim_lock to test for a specific race
2147 condition that could lead to file corruption when using MBX delivery. The
2148 issue is with the lockfile that is created in /tmp. If this file is removed
2149 after a process has opened it but before that process has acquired a lock,
2150 there is the potential for a second process to recreate the file and also
2151 acquire a lock. This could lead to two Exim processes writing to the file
2152 at the same time. The added code performs the same test as UW imapd; it
2153 checks after acquiring the lock that its file descriptor still refers to
2154 the same named file.
2156 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2157 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2160 11. Added the smtp_active_hostname option. If used, this will typically be made
2161 to depend on the incoming interface address. Because $interface_address is
2162 not set up until the daemon has forked a reception process, error responses
2163 that can happen earlier (such as "too many connections") no longer contain
2166 12. If an expansion in a condition on a "warn" statement fails because a lookup
2167 defers, the "warn" statement is abandoned, and the next ACL statement is
2168 processed. Previously this caused the whole ACL to be aborted.
2170 13. Added the iplsearch lookup type.
2172 14. Added ident_timeout as a log selector.
2174 15. Added tls_certificate_verified as a log selector.
2176 16. Added a global option tls_require_ciphers (compare the smtp transport
2177 option of the same name). This controls incoming TLS connections.
2179 17. I finally figured out how to make tls_require_ciphers do a similar thing
2180 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2181 before starting the TLS session.
2183 18. Tabs are now shown as \t in -bP output.
2185 19. If the log selector return_path_on_delivery was set, Exim crashed when
2186 bouncing a message because it had too many Received: header lines.
2188 20. If two routers both had headers_remove settings, and the first one included
2189 a superfluous trailing colon, the final name in the first list and the
2190 first name in the second list were incorrectly joined into one item (with a
2191 colon in the middle).
2197 1. Added -C and -D options to the exinext utility, mainly to make it easier
2198 to include in the automated testing, but these could be helpful when
2199 multiple configurations are in use.
2201 2. The exinext utility was not formatting the output nicely when there was
2202 an alternate port involved in the retry record key, nor when there was a
2203 message id as well (for retries that were specific to a specific message
2204 and a specific host). It was also confused by IPv6 addresses, because of
2205 the additional colons they contain. I have fixed the IPv4 problem, and
2206 patched it up to do a reasonable job for IPv6.
2208 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2209 delivery, the log line now contains "pipelined" if PIPELINING was used.
2211 4. An SMTP transport process used to panic and die if the bind() call to set
2212 an explicit outgoing interface failed. This has been changed; it is now
2213 treated in the same way as a connect() failure.
2215 5. A reference to $sender_host_name in the part of a conditional expansion
2216 that was being skipped was still causing a DNS lookup. This no longer
2219 6. The def: expansion condition was not recognizing references to header lines
2220 that used bh_ and bheader_.
2222 7. Added the _cache feature to named lists.
2224 8. The code for checking quota_filecount in the appendfile transport was
2225 allowing one more file than it should have been.
2227 9. For compatibility with Sendmail, the command line option
2235 and sets the incoming protocol and host name (for trusted callers). The
2236 host name and its colon can be omitted when only the protocol is to be set.
2237 Note the Exim already has two private options, -pd and -ps, that refer to
2238 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2239 "s", but I don't think that's a major issue.
2241 10. A number of refactoring changes to the code, none of which should affect
2244 (a) The number of logging options was getting close to filling up the
2245 32-bit word that was used as a bit map. I have split them into two classes:
2246 those that are passed in the argument to log_write(), and those that are
2247 only ever tested independently outside of that function. These are now in
2248 separate 32-bit words, so there is plenty of room for expansion again.
2249 There is no change in the user interface or the logging behaviour.
2251 (b) When building, for example, log lines, the code previously used a
2252 macro that called string_cat() twice, in order to add two strings. This is
2253 not really sufficiently general. Furthermore, there was one instance where
2254 it was actually wrong because one of the argument was used twice, and in
2255 one call a function was used. (As it happened, calling the function twice
2256 did not affect the overall behaviour.) The macro has been replaced by a
2257 function that can join an arbitrary number of extra strings onto a growing
2260 (c) The code for expansion conditions now uses a table and a binary chop
2261 instead of a serial search (which was left over from when there were very
2262 few conditions). Also, it now recognizes conditions like "pam" even when
2263 the relevant support is not compiled in: a suitably worded error message is
2264 given if an attempt is made to use such a condition.
2266 11. Added ${time_interval:xxxxx}.
2268 12. A bug was causing one of the ddress fields not to be passed back correctly
2269 from remote delivery subprocesses. The field in question was not being
2270 subsequently used, so this caused to problems in practice.
2272 13. Added new log selectors queue_time and deliver_time.
2274 14. Might have fixed a bug in maildirsizefile handling that threw up
2275 "unexpected character" debug warnings, and recalculated the data
2276 unnecessarily. In any case, I expanded the warning message to give more
2279 15. Added the message "Restricted characters in address" to the statements in
2280 the default ACL that block characters like @ and % in local parts.
2282 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2283 Three changes have been made:
2285 (a) There was a serious bug; a negative response to MAIL caused the whole
2286 recipient domain to be cached as invalid, thereby blocking all messages
2287 to all local parts at the same domain, from all senders. This bug has
2288 been fixed. The domain is no longer cached after a negative response to
2289 MAIL if the sender used is not empty.
2291 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2294 (c) A new callout option, "use_sender" has been added for people who want
2295 the modified behaviour.
2301 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2304 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2305 indeed breaks things for older releases.
2307 3. Added additional logging to the case where there is a problem reading data
2308 from a filter that is running in a subprocess using a pipe, in order to
2309 try to track down a specific problem.
2311 4. Testing facility fudge: when running in the test harness and attempting
2312 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2313 getting "No route to host". Convert this to a timeout.
2315 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2318 6. Some OS don't have socklen_t but use size_t instead. This affects the
2319 fifth argument of getsockopt() amongst other things. This is now
2320 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2321 can be set for individual OS. I have set it for SunOS5, OSF1, and
2322 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2323 some earlier ones do not.
2325 7. Change 4.30/15 was not doing the test caselessly.
2327 8. The standard form for an IPv6 address literal was being rejected by address
2328 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2329 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2330 this, as well as the form without the "IPv6" on the front (but only when
2331 address literals are enabled, of course).
2333 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2335 10. Exim crashed if a message with an empty sender address specified by -f
2336 encountered a router with an errors_to setting. This could be provoked only
2337 by a command such as
2341 where an empty string was supplied; "<>" did not hit this bug.
2343 11. Installed PCRE release 4.5.
2345 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2346 remained set. It is now erased.
2348 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2349 times from message ids (which are base 36 rather than the normal 62).
2351 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2352 were being counted as actual protocol errors, and logged if the log
2353 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2354 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2355 and DATA following a set of rejected RCPTs do not count as protocol errors.
2356 In other words, Exim assumes they were pipelined, though this may not
2357 actually be the case. Of course, in all cases the client gets an
2358 appropriate error code.
2360 15. If a lookup fails in an ACL condition, a message about the failure may
2361 be available; it is used if testing the ACL cannot continue, because most
2362 such messages specify what the cause of the deferral is. However, some
2363 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2364 that caused an old message to be retained and used if a later statement
2365 caused a defer, replacing the real cause of the deferral.
2367 16. If an IP address had so many PTR records that the DNS lookup buffer
2368 was not large enough to hold them, Exim could crash while trying to process
2369 the truncated data. It now detects and logs this case.
2371 17. Further to 4.21/58, another change has been made: if (and only if) the
2372 first line of a message (the first header line) ends with CRLF, a bare LF
2373 in a subsequent header line has a space inserted after it, so as not to
2374 terminate the header.
2376 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2377 unnecessarily, used atoi() instead of strtol(), and didn't check the
2378 termination when getting file sizes from file names by regex.
2380 19. Completely re-implemented the support for maildirsize files, in the light
2381 of a number of problems with the previous contributed implementation
2382 (4.30/29). In particular:
2384 . If the quota is zero, the maildirsize file is maintained, but no quota is
2387 . If the maildir directory does not exist, it is created before any attempt
2388 to write a maildirsize file.
2390 . The quota value in the file is just a cache; if the quota is changed in
2391 the transport, the new value overrides.
2393 . A regular expression is available for excluding directories from the
2396 20. The autoreply transport checks the characters in options that define the
2397 message's headers; it allows continued headers, but it was checking with
2398 isspace() after an embedded newline instead of explicitly looking for a
2401 21. If all the "regular" hosts to which an address was routed had passed their
2402 expiry times, and had not reached their retry times, the address was
2403 bounced, even if fallback hosts were defined. Now Exim should go on to try
2406 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2407 equivalent code in the SMTP transport. Some hosts send humungous responses
2408 to HELO/EHLO, more than 1024 it seems.
2410 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2411 gives compiler warnings in some environments. I've now done it "properly",
2414 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2415 (because of problems with the built-in one) was declared to return uschar *
2416 instead of char *, causing compiler failure.
2418 25. Fixed a file descriptor leak when processing alias/forward files.
2420 26. Fixed a minor format string issue in dbfn.c.
2422 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2424 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2425 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2428 29. When a sender address is verified, it is cached, to save repeating the test
2429 when there is more than one recipient in a message. However, when the
2430 verification involves a callout, it is possible for different callout
2431 options to be set for different recipients. It is too complicated to keep
2432 track of this in the cache, so now Exim always runs a verification when a
2433 callout is required, relying on the callout cache for the optimization.
2434 The overhead is duplication of the address routing, but this should not be
2437 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2438 to be verified with callout=postmaster, and the main callout worked but the
2439 postmaster check failed, the verification correctly failed. However, if a
2440 subsequent RCPT command asked for sender verification *without* the
2441 postmaster check, incorrect caching caused this verification also to fail,
2444 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2445 it was not caching the DNS options (qualify_single, search_parents) that
2446 were used when the lookup failed. A subsequent lookup with different
2447 options therefore always gave the same answer, though there were cases
2448 where it should not have. (Example: a "domains = !$mx_any" option on a
2449 dnslookup router: the "domains" option is always processed without any
2450 widening, but the router might have qualify_single set.) Now Exim uses the
2451 cached value only when the same options are set.
2453 32. Added John Jetmore's "exipick" utility to the distribution.
2455 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2456 than a timeout (e.g. a certificate is required, and is not provided), an
2457 Exim server now closes the connection immediately. Previously it waited for
2458 the client to close - but if the client is SSL, it seems that they each
2459 wait for each other, leading to a delay before one of them times out.
2461 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2462 maintained 0.8.x compatibility because I don't think many are using it, and
2463 it is clearly obsolete.
2465 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2468 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2469 client certificate was expired. A simple patch fixes this, though I don't
2470 understand the full logic of why the verify callback is called multiple
2473 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2474 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2475 which causes problems with some clients (such as the Certicom SSL Plus
2476 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2477 disables the coutermeasure allowing Eudora to connect."
2479 38. Exim was not checking that a write() to a log file succeeded. This could
2480 lead to Bad Things if a log got too big, in particular if it hit a file
2481 size limit. Exim now panics and dies if it cannot write to a log file, just
2482 as it does if it cannot open a log file.
2484 39. Modified OS/Makefile-Linux so that it now contains
2486 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2488 The two -D definitions ensure that Exim is compiled with large file
2489 support, which makes it possible to handle log files that are bigger than
2492 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2493 instance) a domain was checked against a named list that involved a lookup,
2494 causing $domain_data to be set, then another domain was checked against the
2495 same list, then the first domain was re-checked, the value of $domain_data
2496 after the final check could be wrong. In particular, if the second check
2497 failed, it could be set empty. This bug probably also applied to
2500 41. The strip_trailing_dot option was not being applied to the address given
2501 with the -f command-line option.
2503 42. The code for reading a message's header from the spool was incrementing
2504 $received_count, but never initializing it. This meant that the value was
2505 incorrect (doubled) while delivering a message in the same process in which
2506 it was received. In the most common configuration of Exim, this never
2507 happens - a fresh exec is done - but it can happen when
2508 deliver_drop_privilege is set.
2510 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2511 it now includes up to 150 characters of the unexpected data in the log
2514 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2515 and building data strings. The size of both of these buffers was 10 000
2516 bytes - far larger than anybody would *ever* want, thought I. Needless to
2517 say, somebody hit the limit. I have increased the maximum line length to
2518 20 000 and the maximum data length of concatenated lines to 100 000. I have
2519 also fixed two bugs, because there was no checking on these buffers. Tsk,
2520 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2521 buffer is too small.
2523 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2524 lsearch lookups. Now it does.
2526 46. When parsing a route_list item in a manualroute router, a fixed-length
2527 buffer was used for the list of hosts. I made this 1024 bytes long,
2528 thinking that nobody would ever have a list of hosts that long. Wrong.
2529 Somebody had a whole pile of complicated expansion conditions, and the
2530 string was silently truncated, leading to an expansion error. It turns out
2531 that it is easier to change to an unlimited length (owing to other changes
2532 that have happened since this code was originally written) than to build
2533 structure for giving a limitation error. The length of the item that
2534 expands into the list of hosts is now unlimited.
2536 47. The lsearch lookup could not handle data where the length of text line was
2537 more than 4095 characters. Such lines were truncated, leading to shortened
2538 data being returned. It should now handle lines of any length.
2540 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2541 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2544 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2545 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2546 no longer happens in comment lines. A list of replacements is now placed
2547 at the head of all of the source files, except those whose only change is
2548 to replace PERL_COMMAND in the very first #! line.
2550 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2551 messages on the queue, with a bottom-up merge sort, using code contributed
2552 by Michael Haardt. This should make operations like -bp somewhat faster on
2553 large queues. It won't affect queue runners, except when queue_run_in_order
2556 51. Installed eximstats 1.31 in the distribution.
2558 52. Added support for SRV lookups to the dnslookup router.
2560 53. If an ACL referred to $message_body or $message_body_end, the value was not
2561 reset for any messages that followed in the same SMTP session.
2563 54. The store-handling optimization for building very long strings was not
2564 differentiating between the different store pools. I don't think this
2565 actually made any difference in practice, but I've tidied it.
2567 55. While running the routers to verify a sender address, $sender_address
2568 was still set to the sender address. This is wrong, because when routing to
2569 send a bounce to the sender, it would be empty. Therefore, I have changed
2570 it so that, while verifying a sender address, $sender_address is set to <>.
2571 (There is no change to what happens when verifying a recipient address.)
2573 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2574 target A or AAAA records (if not already returned) without resetting the
2575 qualify_single or search_parents options of the DNS resolver. These are
2576 inappropriate in this case because the targets of MX and SRV records must
2577 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2578 target that, when qualified, matched something in the local domain. These
2579 two options are now turned off when doing these lookups.
2581 57. It seems that at least some releases of Reiserfs (which does not have the
2582 concept of a fixed number of inodes) returns zero and not -1 for the
2583 number of available inodes. This interacted badly with check_spool_inodes,
2584 which assumed that -1 was the "no such thing" setting. What I have done is
2585 to check that the total number of inodes is greater than zero before doing
2586 the test of how many are available.
2588 58. When a "warn" ACL statement has a log_message modifier, the message is
2589 remembered, and not repeated. This is to avoid a lot of repetition when a
2590 message has many recipients that cause the same warning to be written.
2591 Howewer, Exim was preserving the list of already written lines for an
2592 entire SMTP session, which doesn't seem right. The memory is now reset if a
2593 new message is started.
2595 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2596 debugging output unless log_rewrite was also set.
2598 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2599 of (int)(handle) when we know that handle contains (void *)(-1).
2601 61. The Exim daemon panic-logs an error return when it closes the incoming
2602 connection. However "connection reset by peer" seems to be common, and
2603 isn't really an error worthy of noting specially, so that particular error
2606 62. When Exim is trying to find all the local interfaces, it used to panic and
2607 die if the ioctl to get the interface flags failed. However, it seems that
2608 on at least one OS (Solaris 9) it is possible to have an interface that is
2609 included in the list of interfaces, but for which you get a failure error
2610 for this call. This happens when the interface is not "plumbed" into a
2611 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2612 failure of the "get flags" call assumes that the interface is down.
2614 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2615 makes life easier for people who are doing arithmetic on fields extracted
2616 from dates, where you often get leading zeros that should not be
2617 interpreted as octal.
2619 64. Added qualify_domain to the redirect router, to override the global
2622 65. If a pathologically long header line contained very many addresses (the
2623 report of this problem mentioned 10 000) and each of them was rewritten,
2624 Exim could use up a very large amount of memory. (It kept on making new
2625 copies of the header line as it rewrote, and never released the old ones.)
2626 At the expense of a bit more processing, the header rewriting function has
2627 been changed so that it no longer eats memory in this way.
2629 66. The generation of the Received: header has been moved from the time that a
2630 message starts to be received, to the time that it finishes. The timestamp
2631 in the Received: header should now be very close to that of the <= log
2632 line. There are two side-effects of this change:
2634 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2635 logged header lines no longer include the local Received: line, because
2636 it has not yet been created. The same applies to a copy of the message
2637 that is returned to a non-SMTP sender when a message is rejected.
2639 (b) When a filter file is tested using -bf, no additional Received: header
2640 is added to the test message. After some thought, I decided that this
2643 This change does not affect the value of $received_for. It is still set
2644 after address rewriting, but before local_scan() is called.
2646 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2648 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2649 gave an unhelpful panic error message, and a defer error. I have managed to
2650 change this behaviour so that it now rejects any supplied certificate,
2651 which seems right, as the list of acceptable certificates is empty.
2653 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2654 gave an unhelpful defer error. I have not managed to make this reject any
2655 supplied certificates, but the error message it gives is "no certificate
2656 supplied", which is not helpful.
2658 70. exigrep's output now also includes lines that are not associated with any
2659 message, but which match the given pattern. Implemented by a patch from
2660 Martin Sluka, which also tidied up the Perl a bit.
2662 71. Recipient callout verification, like sender verification, was using <> in
2663 the MAIL FROM command. This isn't really the right thing, since the actual
2664 sender may affect whether the remote host accepts the recipient or not. I
2665 have changed it to use the actual sender in the callout; this means that
2666 the cache record is now keyed on a recipient/sender pair, not just the
2667 recipient address. There doesn't seem to be a real danger of callout loops,
2668 since a callout by the remote host to check the sender would use <>.
2669 [SEE ABOVE: changed after hitting problems.]
2671 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2672 temporary errors. However, in the case of such a code being given after
2673 the end of a data transmission (i.e. after ".") Exim was failing to write
2674 a retry record for the message. (Yes, there was some broken host that was
2675 actually sending 8xx at this point.)
2677 73. An unknown lookup type in a host list could cause Exim to panic-die when
2678 the list was checked. (An example that provoked this was putting <; in the
2679 middle of a list instead of at the start.) If this happened during a DATA
2680 ACL check, a -D file could be left lying around. This kind of configuration
2681 error no longer causes Exim to die; instead it causes a defer errror. The
2682 incident is still logged to the main and panic logs.
2684 74. Buglet left over from Exim 3 conversion. The message "too many messages
2685 in one connection" was written to the rejectlog but not the mainlog, except
2686 when address rewriting (yes!) was being logged.
2688 75. Added write_rejectlog option.
2690 76. When a system filter was run not as root (that is, when system_filter_user
2691 was set), the values of the $n variables were not being returned to the
2692 main process; thus, they were not subsequently available in the $sn
2695 77. Added +return_path_on_delivery log selector.
2697 78. A connection timeout was being treated differently from recipients deferred
2698 when testing hosts_max_try with a message that was older than the host's
2699 retry timeout. (The host should not be counted, thus allowing all hosts to
2700 be tried at least once before bouncing.) This may have been the cause of an
2701 occasionally reported bug whereby a message would remain on the queue
2702 longer than the retry timeout, but would be bounced if a delivery was
2703 forced. I say "may" because I never totally pinned down the problem;
2704 setting up timeout/retry tests is difficult. See also the next item.
2706 79. The ultimate address timeout was not being applied to errors that involved
2707 a combination of host plus message (for example, a timeout on a MAIL
2708 command). When an address resolved to a number of possible hosts, and they
2709 were not all tried for each delivery (e.g. because of hosts_max_try), a
2710 message could remain on the queue longer than the retry timeout.
2712 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2715 81. Fixed an obscure SMTP outgoing bug which required at least the following
2716 conditions: (a) there was another message waiting for the same server;
2717 (b) the server returned 5xx to all RCPT commands in the first message so
2718 that the message was not completed; (c) the server dropped the connection
2719 or gave a negative response to the RSET that Exim sends to abort the
2720 transaction. The observed case was a dropped connection after DATA that had
2721 been sent in pipelining mode. That is, the server had advertised PIPELINING
2722 but was not implementing it correctly. The effect of the bug was incorrect
2723 behaviour, such as trying another host, and this could lead to a crash.
2729 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2730 and daemon.c were passed as pointers to ints; they should have been
2731 pointers to socklen_t variables (which are typically unsigned ints).
2733 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2736 3. Fixed a really odd bug that affected only the testing scheme; patching a
2737 certain fixed string in the binary changed the value of another string that
2738 happened to be identical to the end of the original first string.
2740 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2741 name", it returns that address as the IP address. On some operating
2742 systems (e.g. Solaris), it also passes back the IP address string as the
2743 "host name". However, on others (e.g. Linux), it passes back an empty
2744 string. Exim wasn't checking for this, and was changing the host name to an
2745 empty string, assuming it had been canonicized.
2747 5. Although rare, it is permitted to have more than one PTR record for a given
2748 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2749 all the names associated with an address, because they do in Solaris.
2750 However, it seems that they do not in Linux for data that comes from the
2751 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2752 I found this out when I moved to a new Linux workstation and tried to run
2753 the Exim test suite.
2755 To get round this problem I have changed the code so that it now does its
2756 own call to the DNS to look up PTR records when searching for a host name.
2757 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2758 addresses that are only in /etc/hosts are still found.
2760 This behaviour is, however, controlled by an option called host_lookup_
2761 order, which defaults to "bydns:byaddr". If people want to use the other
2762 order, or indeed, just use one or the other means of lookup, they can
2763 specify it in this variable.
2765 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2766 some operating systems, this comes back from gethostbyaddr() as an empty
2767 string, and this is what Exim used to test for. However, it seems that in
2768 other systems, "." is yielded. Exim now tests for this case too.
2770 7. The values of check_spool_space and check_log_space are now held internally
2771 as a number of kilobytes instead of an absolute number of bytes. If a
2772 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2773 kilobyte. This means that much larger values can be stored.
2775 8. Exim monitor: an attempt to get the action menu when not actually pointing
2776 at a message produces an empty menu entitled "No message selected". This
2777 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2778 no entries in it ("Shell widget menu has zero width and/or height"). So I
2779 have added a single, blank menu entry in this case.
2781 9. Added ${quote_local_part.
2783 10. MIME decoding is now applied to the contents of Subject: header lines when
2786 11. Now that a reference to $sender_host_address automatically causes a reverse
2787 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2788 host lookup before query-style lookups in lists that might use this
2789 variable. This has therefore been abolished, and the "net-" prefix is no
2790 longer necessary for query-style lookups.
2792 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2793 have been a typo for LFLAGS, so it has been changed.
2795 13. The install script calls Exim with "-C /dev/null" in order to find the
2796 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2797 to be output. Howeve, since Exim outputs its version number before the
2798 error, it didn't break the script. It just looked ugly. I fixed this by
2799 always allowing "-C /dev/null" if the caller is root.
2801 14. Ignore overlarge ACL variable number when reading spool file - insurance
2802 against a later release with more variables having written the file.
2804 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2805 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2806 this, as well as the form without the "IPv6" on the front.
2808 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2809 OS/Makefile-Darwin file.
2811 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2812 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2814 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2815 variable. However, some OS use ints for inodes. I've added cast to long int
2816 to get rid of the compiler warning.
2818 19. I had forgotten to lock out "/../" in configuration file names when
2819 ALT_CONFIG_PREFIX was set.
2821 20. Routers used for verification do not need to specify transports. However,
2822 if such a router generated a host list, and callout was configured, Exim
2823 crashed, because it could not find a port number from the (non-existent)
2824 transport. It now assumes port 25 in this circumstance.
2826 21. Added the -t option to exigrep.
2828 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2829 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2830 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2831 them is set, LOOKUP_LSEARCH is forced.
2833 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2835 24. Added sender and host information to the "rejected by local_scan()" log
2836 line; previously there was no indication of these.
2838 25. Added .include_if_exists.
2840 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2841 Linux. It turns out that not all file systems support this. Apparently some
2842 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2843 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2844 not supported on the file descriptor, is now ignored when Exim is trying to
2845 sync a directory. This applies only to Linux.
2847 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2849 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2850 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2853 29. Added support for maildirsize files from supplied patch (modified a bit).
2855 30. The use of :fail: followed by an empty string could lead Exim to respond to
2856 sender verification failures with (e.g.):
2858 550 Verification failed for <xxx>
2859 550 Sender verify failed
2861 where the first response line was missing the '-' that indicates it is not
2862 the final line of the response.
2864 31. The loop for finding the name of the user that called Exim had a hardwired
2865 limit of 10; it now uses the value of finduser_retries, which is used for
2866 all other user lookups.
2868 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2871 33. Exim was neglecting to zero errno before one call of strtol() when
2872 expanding a string and expecting an integer value. On some systems this
2873 resulted in spurious "integer overflow" errors. Also, it was casting the
2874 result into an int without checking.
2876 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2877 did not work. The code looks as if it has *never* worked, though it appears
2878 to have been documented since at least releast 1.62. I have made it work.
2880 35. The "timeout_DNS" error in retry rules, also documented since at least
2881 1.62, also never worked. As it isn't clear exactly what this means, and
2882 clearly it isn't a major issue, I have abolished the feature by treating it
2883 as "timeout", and writing a warning to the main and panic logs.
2885 36. The display of retry rules for -brt wasn't always showing the error code
2888 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2889 timeout_connect_A, timeout_connect_MX.
2891 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2892 to the empty sender.
2894 39. The daemon was not analysing the content of -oX till after it had closed
2895 stderr and disconnected from the controlling terminal. This meant that any
2896 syntax errors were only noted on the panic log, and the return code from
2897 the command was 0. By re-arranging the code a little, I've made the
2898 decoding happen first, so such errors now appear on stderr, and the return
2899 code is 1. However, the actual setting up of the sockets still happens in
2900 the disconnected process, so errors there are still only recorded on the
2903 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2904 connections (as happens on some IP stacks) was logged at start up time as
2905 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2906 it from "IPv6 and IPv4", which means that two separate sockets are being
2909 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2910 says whether AF_INET or AF_INET6 was passed as an argument.
2912 42. Exiwhat output was messed up when time zones were included in log
2915 43. Exiwhat now gives more information about the daemon's listening ports,
2916 and whether -tls-on-connect was used.
2918 44. The "port" option of the smtp transport is now expanded.
2920 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2921 silently ignored. Now an error message is written to the main and panic
2924 46. There's a new ACL modifier called "logwrite" which writes to a log file
2925 as soon as it is encountered.
2927 47. Added $local_user_uid and $local_user_gid at routing time.
2929 48. Exim crashed when trying to verify a sender address that was being
2932 49. Exim was recognizing only a space character after ".include". It now also
2933 recognizes a tab character.
2935 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2936 extracting the relevant information from the specification. The man page no
2937 longer contains scrambled data for the -d option, and I've added a section
2938 at the front about calling Exim under different names.
2940 51. Added "extra_headers" argument to the "mail" command in filter files.
2942 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2945 53. Installed eximstats 1.29.
2947 54. Added transport_filter_timeout as a generic transport option.
2949 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2950 Cc: header lines. This was required by RFC 822, but it not required by RFC
2953 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2954 incoming messages that did not have them. Now it does so only if the
2955 message originates locally, that is, if there is no associated remote host
2956 address. When Resent- header lines are present, this applies to the Resent-
2957 lines rather than the non-Resent- lines.
2959 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2960 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2962 58. Messages for configuration errors now include the name of the main
2963 configuration file - useful now that there may be more than one file in a
2964 list (.included file names were always shown).
2966 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2967 for those rare installations that do not start the daemon as root or run it
2968 setuid root. I've cut out the call to initgroups() if the daemon is not
2971 60. The Exim user and group can now be bound into the binary as text strings
2972 that are looked up at the start of Exim's processing.
2974 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2976 62. Added $mailstore_basename variable.
2978 63. Installed patch to sieve.c from Michael Haardt.
2980 64. When Exim failed to open the panic log after failing to open the main log,
2981 the original message it was trying to log was written to stderr and debug
2982 output, but if they were not available (the usual case in production), it
2983 was lost. Now it is written to syslog before the two lines that record the
2984 failures to open the logs.
2986 65. Users' Exim filters run in subprocesses under the user's uid. It is
2987 possible for a "deliver" command or an alias in a "personal" command to
2988 provoke an address rewrite. If logging of address rewriting is configured,
2989 this fails because the process is not running as root or exim. There may be
2990 a better way of dealing with this, but for the moment (because 4.30 needs
2991 to be released), I have disabled address rewrite logging when running a
2992 filter in a non-root, non-exim process.
2998 1. The buildconfig auxiliary program wasn't quoting the value set for
2999 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3000 not defined. This bug was masked in 4.22 by the effect that was fixed in
3003 2. Some messages that were rejected after a message id was allocated were
3004 shown as "incomplete" by exigrep. It no longer does this for messages that
3005 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3007 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3008 have allow_domain_literals set, the ID did not get logged in the <= line.
3009 Domain literals are now always recognized in Message-ID: header lines.
3011 4. The first argument for a ${extract expansion item is the key name or field
3012 number. Leading and trailing spaces in this item were not being ignored,
3013 causing some misleading effects.
3015 5. When deliver_drop_privilege was set, single queue runner processes started
3016 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3017 same command in the process it spins off) were not dropping privilege.
3019 6. When the daemon running as "exim" started a queue runner, it always
3020 re-executed Exim in the spun-off process. This is a waste of effort when
3021 deliver_drop_privilege is set. The new process now just calls the
3022 queue-runner function directly.
3028 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3031 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3032 ignored. Though the use of -oP was forcing the writing of a pid file, it
3033 was always written to the default place.
3035 3. If the message "no IP address found for host xxxx" is generated during
3036 incoming verification, it is now followed by identification of the incoming
3037 connection (so you can more easily find what provoked it).
3039 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3041 5. Added some features to "harden" Exim a bit more against certain attacks:
3043 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3044 be put in Local/Makefile. This is like the never_users runtime option,
3045 but it cannot be overridden. The default setting is "root".
3047 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3048 prefix string with which any file named in a -C command line option
3051 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3052 is retained for -C and -D only if the caller of Exim is root. Without
3053 it, the exim user may also use -C and -D and retain privilege.
3055 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3056 command line option is disabled.
3058 6. Macro names set by the -D option must start with an upper case letter, just
3059 like macro names defined in the configuration file.
3061 7. Added "dereference=" facility to LDAP.
3063 8. Two instances of the typo "uknown" in the source files are fixed.
3065 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3066 the Configure-Makefile script screwed up while processing it.
3068 10. Incorporated PCRE 4.4.
3070 11. The SMTP synchronization check was not operating right at the start of an
3071 SMTP session. For example, it could not catch a HELO sent before the client
3072 waited for the greeting. There is now a check for outstanding input at the
3073 point when the greeting is written. Because of the duplex, asynchronous
3074 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3075 way, but not yet received, when the check is performed.
3077 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3078 on TCP/IP sockets, because this apparently causes some broken clients to
3081 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3082 unchanged) from the Cygwin maintainer.
3084 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3085 was supported instead of testing for mbx. Effectively a typo.
3087 15. The spa authenticator server code was not checking that the input it
3088 received was valid base64.
3090 16. The debug output line for the "set" modifier in ACLs was not showing the
3091 name of the variable that was being set.
3093 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3095 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3096 lookup on its own. Something else was needed to trigger the lookup. For
3097 example, a match in host_lookup or the need for a host name in a host list.
3098 Now, if $sender_host_name is referenced and the host name has not yet been
3099 looked up, a lookup is performed. If the lookup fails, the variable remains
3100 empty, and $host_lookup_failed is set to "1".
3102 19. Added "eqi" as a case-independent comparison operator.
3104 20. The saslauthd authentication condition could segfault if neither service
3105 nor realm was specified.
3107 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3108 error message that was logged was misleading, and incoming SMTP
3109 connections were dropped. The message is now more accurate, and temporary
3110 errors are given to SMTP connections.
3112 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3113 (except RSET) with a 421 error, until QUIT is received. However, it was
3114 failing to send a response to QUIT.
3116 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3117 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3118 ACL is now run after the helo_try_verify_hosts code.
3120 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3121 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3122 case-independent, so other case variants are also recognized). Apparently
3123 some systems use these upper case variants.
3125 25. If more than two messages were waiting for the same host, and a transport
3126 filter was specified for the transport, Exim sent two messages over the
3127 same TCP/IP connection, and then failed with "socket operation on non-
3128 socket" when it tried to send the third.
3130 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3132 27. The extern definition of crypt16() in expand.c was not being excluded when
3133 the OS had its own crypt16() function.
3135 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3136 as a preferred synonym for return_size_limit, both as an option and as an
3139 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3141 30. Changed the default configuration ACL to relax the local part checking rule
3142 for addresses that are not in any local domains. For these addresses,
3143 slashes and pipe symbols are allowed within local parts, but the sequence
3144 /../ is explicitly forbidden.
3146 31. SPA server authentication was not clearing the challenge buffer before
3149 32. log_message in a "warn" ACL statement was writing to the reject log as
3150 well as to the main log, which contradicts the documentation and doesn't
3151 seem right (because no rejection is happening). So I have stopped it.
3153 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3154 However, I am unable to do any testing of this.
3156 34. Fixed an infelicity in the appendfile transport. When checking directories
3157 for a mailbox, to see if any needed to be created, it was accidentally
3158 using path names with one or more superfluous leading slashes; tracing
3159 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3161 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3162 contents are added to the log line that is written for every discarded
3163 recipient. (Previously a log_message setting was ignored.)
3165 36. The ${quote: operator now quotes the string if it is empty.
3167 37. The install script runs exim in order to find its version number. If for
3168 some reason other than non-existence or emptiness, which it checks, it
3169 could not run './exim', it was installing it with an empty version number,
3170 i.e. as "exim-". This error state is now caught, and the installation is
3173 38. An argument was missing from the function that creates an error message
3174 when Exim fails to connect to the socket for saslauthd authentication.
3175 This could cause Exim to crash, or give a corrupted message.
3177 39. Added isip, isip4, and isip6 to ${if conditions.
3179 40. The ACL variables $acl_xx are now saved with the message, and can be
3180 accessed later in routers, transports, and filters.
3182 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3183 strings in the file are not string-expanded.
3185 42. If a MAIL command specified a SIZE value that was too large to fit into an
3186 int variable, the check against message_size_limit failed. Such values are
3187 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3188 day this will have to be increased, but I don't think I want to be around
3189 when emails are that large.
3196 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3197 iconv() is not standard in FreeBSD.
3199 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3200 IPv6 enabled. The observed symptom was a segmentation fault on return from
3201 the function os_common_find_running_interfaces() in src/os.c.
3203 3. In the check_special_case() function in daemon.c I had used "errno" as an
3204 argument name, which causes warnings on some systems. This was basically a
3205 typo, since it was named "eno" in the comments!
3207 4. The code that waits for the clock to tick (at a resolution of some fraction
3208 of a second) so as to ensure message-id uniqueness was always waiting for
3209 at least one whole tick, when it could have waited for less. [This is
3210 almost certainly not relevant at current processor speeds, where it is
3211 unlikely to ever wait at all. But we try to future-proof.]
3213 5. The function that sleeps for a time interval that includes fractions of a
3214 second contained a race. It did not block SIGALRM between setting the
3215 timer, and suspending (a couple of lines later). If the interval was short
3216 and the sigsuspend() was delayed until after it had expired, the suspension
3217 never ended. On busy systems this could lead to processes getting stuck for
3220 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3221 process, before it forks any delivery processes. The open lookup caching
3222 mechanism meant that the open file or database connection was passed into
3223 the delivery process. The problem was that delivery processes always tidy
3224 up cached lookup data. This could cause a problem for the next delivery
3225 process started by the queue runner, because the external queue runner
3226 process does not know about the closure. So the next delivery process
3227 still has data in the lookup cache. In the case of a file lookup, there was
3228 no problem because closing a file descriptor in a subprocess doesn't affect
3229 the parent. However, if the lookup was caching a connection to a database,
3230 the connection was closed, and the second delivery process was likely to
3231 see errors such as "PGSQL: query failed: server closed the connection
3232 unexpectedly". The problem has been fixed by closing all cached lookups
3233 in a queue runner before running a delivery process.
3235 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3236 seem to have the "const" qualifier which it has on other OS. I've
3239 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3240 *of the same type* (client or server) with the same public name that an
3241 error should be diagnosed.
3243 9. When Exim looked up a host name for an IP address, but failed to find the
3244 original IP address when looking up the host name (a safety check), it
3245 output the message "<ip address> does not match any IP for NULL", which was
3246 confusing, to say the least. The bug was that the host name should have
3247 appeared instead of "NULL".
3249 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3250 user that is built into the binary, and the -C or -D options is used, root
3251 privilege is dropped before the configuration file is read. In addition,
3252 logging is switched to stderr instead of the normal log files. If the
3253 configuration then re-defines the Exim user, the unprivileged environment
3254 is probably not what is expected, so Exim logs a panic warning message (but
3257 However, if deliver_drop_privilege is set, the unprivileged state may well
3258 be exactly what is intended, so the warning has been cut out in that case,
3259 and Exim is allowed to try to write to its normal log files.
3265 1. smtp_return_error_details was not giving details for temporary sender
3266 or receiver verification errors.
3268 2. Diagnose a configuration error if two authenticators have the same public
3271 3. Exim used not to create the message log file for a message until the first
3272 delivery attempt. This could be confusing when incoming messages were held
3273 for policy or load reasons. The message log file is now created at the time
3274 the message is received, and an initial "Received" line is written to it.
3276 4. The automatically generated man page for command line options had a minor
3277 bug that caused no ill effects; however, a more serious problem was that
3278 the procedure for building the man page automatically didn't always
3279 operate. Consequently, release 4.20 contains an out-of-date version. This
3280 shouldn't happen again.
3282 5. When building Exim with embedded Perl support, the script that builds the
3283 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3284 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3286 6. The freeze_tell option was not being used for messages that were frozen on
3287 arrival, either by an ACL or by local_scan().
3289 7. Added the smtp_incomplete_transaction log selector.
3291 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3292 was accepting AUTH without a new EHLO.
3294 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3295 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3298 10. Logging of TCP/IP connections (when configured) now happens in the main
3299 daemon process instead of the child process, so that the TCP/IP connection
3300 count is more accurate (but it can never be perfect).
3302 11. The use of "drop" in a nested ACL was not being handled correctly in the
3303 outer ACL. Now, if condition failure induced by the nested "drop" causes
3304 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3305 or "require"), the connection is dropped.
3307 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3308 that yield "discard" can now be used with an "accept" or a "discard" verb,
3309 but an error is generated for any others (because I can't see a useful way
3310 to define what should happen).
3312 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3313 are now processed in the same way as lines in the Exim configuration file.
3314 In particular, continuation lines are supported.
3316 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3318 15. Added -ti meaning -t -i.
3320 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3321 domains, and warn by logging if others are found.
3323 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3324 returned from GIFCONF when Exim is trying to find the list of interfaces on
3325 a host. The code in os.c has been modified to copy each ifreq to an aligned
3326 structure in all cases.
3328 Also, in some cases, the returned ifreq's were being copied to a 'struct
3329 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3330 means the last couple of bytes of an IPv6 address could be chopped if the
3331 ifreq contained only a normal sockaddr (14 bytes storage).
3333 18. Named domain lists were not supported in the hosts_treat_as_local option.
3334 An entry such as +xxxx was not recognized, and was treated as a literal
3337 19. Ensure that header lines added by a DATA ACL are included in the reject log
3338 if the ACL subsequently rejects the message.
3340 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3341 MD5 (which is deprecated).
3343 21. When testing a filter file using -bf, Exim was writing a message when it
3344 took the sender from a "From " line in the message, but it was not doing so
3345 when it took $return_path from a Return-Path: header line. It now does.
3347 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3348 with a valid header line field name (a series of printing characters
3349 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3351 23. Changed "disc" in the source to "disk" to conform to the documentation and
3352 the book and for uniformity.
3354 24. Ignore Sendmail's -Ooption=value command line item.
3356 25. When execve() failed while trying to run a command in a pipe transport,
3357 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3358 could be confused with a return value of 69 from the command itself. This
3359 has been changed to 127, the value the shell returns if it is asked to run
3360 a non-existent command. The wording for the related log line suggests a
3361 non-existent command as the problem.
3363 26. If received_header_text expands to an empty string, do not add a Received:
3364 header line to the message. (Well, it adds a token one on the spool, but
3365 marks it "old" so that it doesn't get used or transmitted.)
3367 27. Installed eximstats 1.28 (addition of -nt option).
3369 28. There was no check for failure on the call to getsockname() in the daemon
3370 code. This can fail if there is a shortage of resources on the system, with
3371 ENOMEM, for example. A temporary error is now given on failure.
3373 29. Contrary to the C standard, it seems that in some environments, the
3374 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3375 program. Exim now does this explicitly; it affects the formatting of
3376 timestamps using strftime().
3378 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3379 complaints. I've now initialized all the variables, to avoid this.
3381 32. Header lines added by a system filter were not being "seen" during
3382 transport-time rewrites.
3384 33. The info_callback() function passed to OpenSSL is set up with type void
3385 (*)(SSL *, int, int), as described somewhere. However, when calling the
3386 function (actually a macro) that sets it up, the type void(*)() is
3387 expected. I've put in a cast to prevent warnings from picky compilers.
3389 34. If a DNS black list lookup found a CNAME record, but there were no A
3390 records associated with the domain it pointed at, Exim crashed.
3392 35. If a DNS black list lookup returned more than one A record, Exim ignored
3393 all but the first. It now scans all returned addresses if a particular IP
3394 value is being sought. In this situation, the contents of the
3395 $dnslist_value variable are a list of all the addresses, separated by a
3398 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3399 to accept a host name and all its aliases if the forward lookup for any of
3400 them yielded the IP address of the incoming connection. Now it accepts only
3401 those names whose forward lookup yields the correct IP address. Any other
3402 names are discarded. This closes a loophole whereby a rogue DNS
3403 administrator could create reverse DNS records to break through a
3404 wildcarded host restriction in an ACL.
3406 37. If a user filter or a system filter that ran in a subprocess used any of
3407 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3408 the wrong values were passed to the pipe command ($thisaddress had the
3409 value of $0, $0 had the value of $1, etc). This bug was introduced by
3410 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3412 38. Improved the line breaking for long SMTP error messages from ACLs.
3413 Previously, if there was no break point between 40 and 75 characters, Exim
3414 left the rest of the message alone. Two changes have been made: (a) I've
3415 reduced the minimum length to 35 characters; (b) if it can't find a break
3416 point between 35 and 75 characters, it looks ahead and uses the first one
3417 that it finds. This may give the occasional overlong line, but at least the
3418 remaining text gets split now.
3420 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3421 file descriptors might be low, and that setting 1000 would always raise it.
3422 It turns out that in some environments, the limit is already over 1000 and
3423 that lowering it causes trouble. So now Exim takes care not to decrease it.
3425 40. When delivering a message, the value of $return_path is set to $sender_
3426 address at the start of routing (routers may change the value). By an
3427 oversight, this default was not being set up when an address was tested by
3428 -bt or -bv, which affected the outcome if any router or filter referred to
3431 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3432 log only when "message" or "log_message" are set. However, if one of the
3433 conditions was an address verification, or a call to a nested ACL, the
3434 messages generated by the underlying test were being passed through. This
3435 no longer happens. The underlying message is available in $acl_verify_
3436 message for both "message" and "log_message" expansions, so it can be
3437 passed through if needed.
3439 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3440 new expansion $bh_ to give the encoded byte string without charset
3441 translation. Translation happens only if iconv() is available; HAVE_ICONV
3442 indicates this at build time. HEADERS_CHARSET gives the charset to
3443 translate to; headers_charset can change it in the configuration, and
3444 "headers charset" can change it in an individual filter file.
3446 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3447 that creates RFC 2047 encoded "words" labels them as that charset instead
3448 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3449 expansion operator; (ii) when Exim creates a From: line for a local
3450 message; (iii) when a header line is rewritten to include a "phrase" part.
3452 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3453 buggy, causing it to skip the first lines of messages whose message ID
3454 ended in 'D'. This would not have bitten before Exim release 4.14, because
3455 message IDs were unlikely to end in 'D' before then. The effect was to have
3456 incorrect size information for certain domains.
3458 45. #include "config.h" was missing at the start of the crypt16.c module. This
3459 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3462 46. If there was a timeout during a "random" callout check, Exim treated it as
3463 a failure of the random address, and carried on sending RSET and the real
3464 address. If the delay was just some slowness somewhere, the response to the
3465 original RCPT would be taken as a response to RSET and so on, causing
3466 mayhem of various kinds.
3468 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3469 when I implemented it. It didn't allow for the fact that some option values
3470 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3471 the right test for positive values.
3473 48. Domain names in DNS records are case-independent. Exim always looks them up
3474 in lower case. Some resolvers return domain names in exactly the case they
3475 appear in the zone file, that is, they may contain uppercase letters. Not
3476 all resolvers do this - some return always lower case. Exim was treating a
3477 change of case by a resolver as a change of domain, similar to a widening
3478 of a domain abbreviation. This triggered its re-routing code and so it was
3479 trying to route what was effectively the same domain again. This normally
3480 caused routing to fail (because the router wouldn't handle the domain
3481 twice). Now Exim checks for this case specially, and just changes the
3482 casing of the domain that it ultimately uses when it transmits the message
3485 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3488 50. If a filter generated a file delivery with a non-absolute name (possible if
3489 no home directory exists for the router), the forbid_file option was not
3492 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3493 the existing equality matching.
3495 52. Exim was using ints instead of ino_t variables in some places where it was
3496 dealing with inode numbers.
3498 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3499 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3500 called TMPDIR, and if it finds it is different, it changes its value.
3502 54. The smtp_printf() function is now made available to local_scan() so
3503 additional output lines can be written before returning. There is also an
3504 smtp_fflush() function to enable the detection of a dropped connection.
3505 The variables smtp_input and smtp_batched_input are exported to
3508 55. Changed the default runtime configuration: the message "Unknown user"
3509 has been removed from the ACL, and instead placed on the localuser router,
3510 using the cannot_route_message feature. This means that any verification
3511 failures that generate their own messages won't get overridden. Similarly,
3512 the "Unrouteable address" message that was in the ACL for unverifiable
3513 relay addresses has also been removed.
3515 56. Added hosts_avoid_esmtp to the smtp transport.
3517 57. The exicyclog script was not checking for the esoteric option
3518 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3519 will work only if exicyclog is run under the appropriate euid.
3521 58. Following a discussion on the list, the rules by which Exim recognises line
3522 endings on incoming messages have been changed. The -dropcr and drop_cr
3523 options are now no-ops, retained only for backwards compatibility. The
3524 following line terminators are recognized: LF CRLF CR. However, special
3525 processing applies to CR:
3527 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3528 nor a local message in the state where . is a terminator.
3530 (ii) If a bare CR is encountered in a header line, an extra space is added
3531 after the line terminator so as not to end the header. The reasoning
3532 behind this is that bare CRs in header lines are most likely either
3533 to be mistakes, or people trying to play silly games.
3535 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3536 was being incorrectly given as 18 bytes larger than it should have been.
3537 This is a VOB (very old bug).
3539 60. This may never have affected anything current, but just in case it has:
3540 When the local host is found other than at the start of a list of hosts,
3541 the local host, those with the same MX, and any that follow, are discarded.
3542 When the list in question was part of a longer list of hosts, the following
3543 hosts (not currently being processed) were also being discarded. This no
3544 longer happens. I'm not sure if this situation could ever has previously
3547 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3550 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3551 header. This is something that is recommended in a new Internet Draft, and
3552 is something that is documented as being done by Sendmail. There are two
3553 possible values. For messages generated by the autoreply transport, Exim
3556 Auto-Submitted: auto-replied
3558 whereas for all other generated messages (e.g. bounces) it adds
3560 Auto-Submitted: auto-generated
3562 63. The "personal" condition in filters now includes a test for the
3563 Auto-Submitted: header. If it contains the string "auto-" the message it
3564 not considered personal.
3566 64. Added rcpt_include_affixes as a generic transport option.
3568 65. Added queue_only_override (default true).
3570 66. Added the syslog_duplication option.
3572 67. If what should have been the first header line of a message consisted of
3573 a space followed by a colon, Exim was mis-interpreting it as a header line.
3574 It isn't of course - it is syntactically invalid and should therefore be
3575 treated as the start of the message body. The misbehaviour could have
3576 caused a number of strange effects, including loss of data in subsequent
3577 header lines, and spool format errors.
3579 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3580 client host had authenticated. This control can now be exercised by an ACL
3581 for more flexibility.
3583 69. By default, callouts do not happen when testing with -bh. There is now a
3584 variant, -bhc, which does actually run the callout code, including
3585 consulting and updating the callout cache.
3587 70. Added support for saslauthd authentication, courtesy of Alexander
3590 71. If statvfs() failed on the spool or log directories while checking their
3591 size for availability, Exim confusingly gave the error "space shortage".
3592 Furthermore, in debugging mode it crashed with a floating point exception.
3593 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3594 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3595 serious problem, Exim now writes to the main and panic logs when this
3596 happens, with details of the failure. It then refuses to accept the
3597 incoming message, giving the message "spool directory problem" or "log
3598 directory problem" with a 421 code for SMTP messages.
3600 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3601 0, 1, and 2 exist, because some OS complain for execs without them (see
3602 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3603 descriptors. However, the code omitted to check that the open succeeded,
3604 causing mysterious errors if for some reason the permissions on /dev/null
3605 got screwed. Now Exim writes a message to the main and panic logs, and
3606 bombs out if it can't open /dev/null.
3608 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3609 interact so that it is all more flexible. It is supposed to remain
3610 backwards compatible. Also added extra_local_interfaces.
3612 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3613 to bomb out with an assertion failure - to the client this appears as a
3614 connection drop. This problem occurs in the part of the code that was taken
3615 from the Samba project. Fortunately, the assertion is in a very simple
3616 function, so I have fixed this by reproducing the function inline in the
3617 one place where it is called, and arranging for authentication to fail
3618 instead of killing the process with assert().
3620 75. The SPA client code was not working when the server requested OEM rather
3621 than Unicode encoding.
3623 76. Added code to make require_files with a specific uid setting more usable in
3624 the case where statting the file as root fails - usually a non-root-mounted
3625 NFS file system. When this happens and the failure is EACCES, Exim now
3626 forks a subprocess and does the per-uid checking as the relevant uid.
3628 77. Added process_log_path.
3630 78. If log_file_path was not explicitly set, a setting of check_log_space or
3631 check_log_inodes was ignored.
3633 79. If a space check for the spool or log partitions fails, the incident is now
3634 logged. Of course, in the latter case the data may get lost...
3636 80. Added the %p formatting code to string_format() so that it can be used to
3637 print addresses in debug_print(). Adjusted all the address printing in the
3638 debugging in store.c to use %p rather than %d.
3640 81. There was a concern that a line of code in smtp_in.c could overflow a
3641 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3642 initially expressed, the concern was not well-founded, because trailing
3643 spaces are removed early. However, if the trailing spaces were followed by
3644 a NULL, they did not get removed, so the overflow was possible. Two fixes
3647 (a) I re-wrote the offending code in a cleaner fashion.
3648 (b) If an incoming SMTP command contains a NULL character, it is rejected
3651 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3652 runs initgroups(), so that if the Exim user is in any additional groups,
3653 they will be used during message reception.
3659 The change log for 4.20 and earlier releases has been archived.