1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.262 2005/11/15 10:08:24 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 Two changes to the default runtime configuration:
11 (1) Move the checks for relay_from_hosts and authenticated clients from
12 after to before the (commented out) DNS black list checks.
14 (2) Add control=submission to the relay_from_hosts and authenticated
15 clients checks, on the grounds that messages accepted by these
16 statements are most likely to be submissions.
18 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
20 (1) Generate an error if the third argument for the ${prvs expansion is
23 (2) Treat a missing third argument of ${prvscheck as if it were an empty
26 (3) Reset the variables that are obtained from the first argument of
27 ${prvscheck and used in the second argument before leaving the code,
28 because their memory is reclaimed, so using them afterwards may do
31 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
32 one (it's much easier than Tom thought :-).
34 (5) Because of (4), we can now allow for the use of $prvscheck_result
35 inside the third argument.
37 PH/03 For some reason, the default setting of PATH when running a command from
38 a pipe transport was just "/usr/bin". I have changed it to
41 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
42 anything to be listed in the output from -bV.
44 PH/05 When a filter generated an autoreply, the entire To: header line was
45 quoted in the delivery log line, like this:
47 => >A.N.Other <ano@some.domain> <original@ddress> ...
49 This has been changed so that it extracts the operative address. There
50 may be more than one such address. If so, they are comma-separated, like
53 => >ano@some.domain,ona@other.domain <original@ddress> ...
55 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
56 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
57 not being looked up in the rDNS to get a host name, Exim was showing the
58 IP address twice in Received: lines, even though the IP addresses were
59 identical. For example:
61 Received: from [1.2.3.4] (helo=[1.2.3.4])
63 However, if the real host name was known, it was omitting the HELO data
64 if it matched the actual IP address. This has been tidied up so that it
65 doesn't show the same IP address twice.
67 PH/07 When both +timestamp and +memory debugging was on, the value given by
68 $tod_xxx expansions could be wrong, because the tod_stamp() function was
69 called by the debug printing, thereby overwriting the timestamp buffer.
70 Debugging no longer uses the tod_stamp() function when +timestamp is set.
72 PH/08 When the original message was included in an autoreply transport, it
73 always said "this is a copy of the message, including all the headers",
74 even if body_only or headers_only was set. It now gives an appropriate
77 PH/09 Applied a patch from the Sieve maintainer which:
80 o adds the (disabled) notify extension core
81 o adds some debug output for the result of if/elsif tests
82 o points to the current vacation draft in the documentation
83 and documents the missing references header update
87 o fixes a bug in processing the envelope test (when testing
88 multiple envelope elements, the last element determinted the
91 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
92 Electronic Mail") by including:
94 Auto-submitted: auto-generated
96 in the messages that it generates (bounce messages and others, such as
97 warnings). In the case of bounce messages for non-SMTP mesages, there was
98 also a typo: it was using "Auto_submitted" (underscore instead of
99 hyphen). Since every message generated by Exim is necessarily in response
100 to another message, thes have all been changed to:
102 Auto-Submitted: auto-replied
104 in accordance with these statements in the RFC:
106 The auto-replied keyword:
108 - SHOULD be used on messages sent in direct response to another
109 message by an automatic process,
111 - MUST NOT be used on manually-generated messages,
113 - MAY be used on Delivery Status Notifications (DSNs) and Message
114 Disposition Notifications (MDNs),
116 - MUST NOT be used on messages generated by automatic or periodic
117 processes, except for messages which are automatic responses to
120 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
121 to the default Received: header definition.
123 PH/12 Added log selector acl_warn_skipped (default on).
125 PH/13 After a successful wildlsearch lookup, discard the values of numeric
126 variables because (a) they are in the wrong storage pool and (b) even if
127 they were copied, it wouldn't work properly because of the caching.
129 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
130 checking when decoding. Apparently there are clients that generate
131 overlong encoded strings. Why am I not surprised?
137 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
138 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
141 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
142 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
144 PH/03 Typo: missing ".o" in src/pcre/Makefile.
146 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
147 header line, restrict the check to what is listed in RFCs 2369 and 2929.
148 Also, for "Auto-Submitted", treat anything other than "no" as
149 non-personal, in accordance with RFC 3834. (Previously it treated
150 anything starting "auto-" as non-personal.)
152 TF/01 The control=submission/name=... option had a problem with syntax
153 errors if the name included a slash character. The /name= option
154 now slurps the rest of the string, so it can include any characters
155 but it must come last in the list of options (after /sender_retain
158 PH/05 Some modifications to the interface to the fake nameserver for the new
161 PH/06 Add ${if def:sender_address {(envelope-from <$sender_address>)\n\t}}
162 into the default Received: header string.
169 TK/01 Added the "success_on_redirect" address verification option. See
170 NewStuff for rationale and an example.
172 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
174 PH/02 Patch to exigrep to allow it to work on syslog lines.
176 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
177 fread() to skip over the body file's header line, because in Cygwin the
178 header line is locked and is inaccessible.
180 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
181 co-exist for some time) to make it clear that it is the Exim ID that is
182 referenced, not the Message-ID: header line.
184 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
185 string_format() function, because snprintf() does not exist on all
188 PH/06 The use of forbid_filter_existstest now also locks out the use of the
189 ${stat: expansion item.
191 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
192 protocol synchronization error", to keep the pedants happy.
194 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
195 well as for IRIX systems, when gcc is being used. See the host.c source
198 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
200 PH/10 Named domain lists were not working if used in a queue_smtp_domains
203 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
204 transport and to the smtp transport in LMTP mode.
206 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
208 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
209 run a filter in a subprocess. This could lead to confusion in subsequent
210 lookups in the parent process. There should also be a search_tidyup() at
211 the end of the subprocess.
213 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
214 only if the host matched helo_try_verify_hosts, which caused the
215 verification to occur when the EHLO/HELO command was issued. The ACL just
216 tested the remembered result. Now, if a previous verification attempt has
217 not happened, "verify = helo" does it there and then.
219 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
221 TK/03 Fix log output including CR from clamd.
223 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
224 exist provoked a memory error which could cause a segfault.
226 PH/15 Installed PCRE 6.2
228 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
230 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
231 of the problem. Specifically, suggested +O2 rather than +O1 for the
234 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
236 PH/20 If a delivery was routed to a non-standard port by means of an SRV
237 record, the port was not correctly logged when the outgoing_port log
238 selector was set (it logged the transort's default port).
240 PH/21 Added support for host-specific ports to manualroute, queryprogram,
241 fallback_hosts, and "hosts" in the smtp transport.
243 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
244 host errors such as "Connection refused".
246 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
247 authentication with radiusclient 0.4.9:
249 - Error returned from rc_read_config was caught wrongly
250 - Username/password not passed on to radius server due to wrong length.
252 The presumption is that some radiusclient API changes for 4.51/PH/17
253 were not taken care of correctly. The code is still untested by me (my
254 Linux distribution still has 0.3.2 of radiusclient), but it was
255 contributed by a Radius user.
257 PH/24 When doing a callout, the value of $domain wasn't set correctly when
258 expanding the "port" option of the smtp transport.
260 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
261 while reading a MIME header. Thanks to Tom Hughes for a patch.
263 PH/24 Include config.h inside local_scan.h so that configuration settings are
266 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
267 that in an ACL for RCPT (for example), you can examine exactly what was
270 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
271 commands, but it was not correctly comparing the address with the actual
272 client host address. Thus, it would show the EHLO address in Received:
273 header lines when this was not necessary.
275 PH/27 Added the % operator to ${eval:}.
277 PH/28 Exim tries to create and chdir to its spool directory when it starts;
278 it should be ignoring failures (because with -C, for example, it has lost
279 privilege). It wasn't ignoring creation failures other than "already
282 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
285 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
286 because an input line was too long, either on its own, or by virtue of
287 too many continuations, the temporary file was not being removed, and the
288 return code was incorrect.
290 PH/31 Missing "BOOL" in function definition in filtertest.c.
292 PH/32 Applied Sieve patches from the maintainer.
294 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
296 PH/33 Added "verify = not_blind".
298 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
299 Local/Makefile (with some defaults set). These are used in built scripts
300 such as exicyclog, but they have never been used in the exim_install
301 script (though there are many overriding facilities there). I have
302 arranged that the exim_install script now takes note of these two
305 PH/35 Installed configuration files for Dragonfly.
307 PH/36 When a locally submitted message by a trusted user did not contain a
308 From: header, and the sender address was obtained from -f or from an SMTP
309 MAIL command, and the trusted user did not use -F to supply a sender
310 name, $originator_name was incorrectly used when constructing a From:
311 header. Furthermore, $originator_name was used for submission mode
312 messages from external hosts without From: headers in a similar way,
313 which is clearly wrong.
315 PH/37 Added control=suppress_local_fixups.
317 PH/38 When log_selector = +received_sender was set, and the addition of the
318 sender made the log line's construction buffer exactly full, or one byte
319 less than full, an overflow happened when the terminating "\n" was
322 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
323 when the result of a list match is failure because a DNS lookup failed.
325 PH/40 RM_COMMAND is now used in the building process.
327 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
328 the "build-* directories that it finds.
330 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
331 address in a domain literal was a prefix of an interface address.
333 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
334 when verifying a sender address, unless rewrite_headers is false.
336 PH/44 Wrote a long comment about why errors_to addresses are verified as
337 recipients, not senders.
339 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
340 the ratelimit ACL was added.
342 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
344 PH/46 Added extra information about PostgreSQL errors to the error string.
346 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
347 avoiding the need to install special zones in a real server. This is
348 backwards compatible; if it can't find the fake resolver, it drops back.
349 Thus, both old and new test suites can be run.
351 TF/02 Added util/ratelimit.pl
353 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
354 clock is set back in time.
356 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
357 Candler <B.Candler@pobox.com>.
359 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
360 OK for addresses that are the result of redirections.
362 PH/48 A number of further additions for the benefit of the new test suite,
363 including a fake gethostbyname() that interfaces to the fake DNS resolver
366 TF/06 The fix for widen_domains has also been applied to qualify_single and
367 search_parents which are the other dnslookup options that can cause
370 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
373 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
375 TF/07 Exim produced the error message "an SRV record indicated no SMTP
376 service" if it encountered an MX record with an empty target hostname.
377 The message is now "an MX or SRV record indicated no SMTP service".
379 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
380 if the DNS of the sending site is misconfigured. This is quite a
381 common situation. This change restores the behaviour of treating a
382 helo verification defer as a failure.
384 PH/51 If self=fail was set on a router, the bounce message did not include the
385 actual error message.
391 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
393 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
394 command itself ended in error, the underlying message about the transport
395 filter timeout was being overwritten with the pipe command error. Now the
396 underlying error message should be appended to the second error message.
398 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
400 PH/02 Reduce the amount of output that "make" produces by default. Full output
401 can still be requested.
403 PH/03 The warning log line about a condition test deferring for a "warn" verb
404 was being output only once per connection, rather than after each
405 occurrence (because it was using the same function as for successful
406 "warn" verbs). This seems wrong, so I have changed it.
408 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
409 it should not have, which might have caused a crash in the right
410 circumstances, but probably never did.
412 PH/04 Installed a modified version of Tony Finch's patch to make submission
413 mode fix the return path as well as the Sender: header line, and to
414 add a /name= option so that you can make the user's friendly name appear
417 TF/03 Added the control = fakedefer ACL modifier.
419 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
420 Mark Lowes for thorough testing.
422 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
424 TK/03 Merged latest SRS patch from Miles Wilton.
426 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
427 with the definition in sysexits.h (which is #included earlier).
428 Fortunately, Exim does not actually use EX_OK. The code used to try to
429 preserve the sysexits.h value, by assumimg that macro definitions were
430 scanned for macro replacements. I have been disabused of this notion,
431 so now the code just undefines EX_OK before #including unistd.h.
433 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
434 in the smtp transport. When a block could not be written in a single
435 write() function, the timeout was being re-applied to each part-write.
436 This seems wrong - if the receiver was accepting one byte at a time it
437 would take for ever. The timeout is now adjusted when this happens. It
438 doesn't have to be particularly precise.
440 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
441 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
443 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
444 without a domain if the check to <postmaster@domain> fails.
446 SC/01 Eximstats: added -xls and the ability to specify output files
447 (patch written by Frank Heydlauf).
449 SC/02 Eximstats: use FileHandles for outputing results.
451 SC/03 Eximstats: allow any combination of xls, txt, and html output.
453 SC/04 Eximstats: fixed display of large numbers with -nvr option
455 SC/05 Eximstats: fixed merging of reports with empty tables.
457 SC/06 Eximstats: added the -include_original_destination flag
459 SC/07 Eximstats: removed tabs and trailing whitespace.
461 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
463 TK/06 MBOX spool code: Add real "From " MBOX separator line
464 so the .eml file is really in mbox format (even though
465 most programs do not really care). Patch from Alex Miller.
467 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
468 The latter is generated from $received_to and is only set if the
469 message has one envelope recipient. SA can use these headers,
470 obviously out-of-the-box. Patch from Alex Miller.
472 PH/08 The ${def test on a variable was returning false if the variable's
473 value was "0", contrary to what the specification has always said!
474 The result should be true unless the variable is empty.
476 PH/09 The syntax error of a character other than { following "${if
477 def:variable_name" (after optional whitespace) was not being diagnosed.
478 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
479 accidental colon was present, for example, could give incorrect results.
481 PH/10 Tidied the code in a number of places where the st_size field of a stat()
482 result is used (not including appendfile, where other changes are about
485 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
486 This involved changing a lot of size variables from int to off_t. It
487 should work with maildirs and everything.
489 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
490 spamd dying while we are connected to it.
492 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
495 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
496 function for BDB 4.3.
498 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
500 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
501 the sources for PCRE have been re-arranged and more files are now
504 PH/15 The code I had for printing potentially long long variables in PH/11
505 above was not the best (it lost precision). The length of off_t variables
506 is now inspected at build time, and an appropriate printing format (%ld
507 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
508 to be "long long int" or "long int". This is needed for the internal
509 formatting function string_vformat().
511 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
512 the configuration file to be ":syslog", then the script "guesses" where
513 the logs files are, rather than using the compiled in default. In our
514 case the guess is not the same as the compiled default, so the script
515 suddenly stopped working when I started to use syslog. The patch checks
516 to see if log_file_path is "". If so, it attempts to read it from exim
517 with no configuration file to get the compiled in version, before it
518 falls back to the previous guessing code."
520 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
521 implementing BATV in an Exim configuration. See NewStuff for the gory
524 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
525 Makefile that are specific to HP-UX.
527 PH/18 If the "use_postmaster" option was set for a recipient callout together
528 with the "random" option, the postmaster address was used as the MAIL
529 FROM address for the random test, but not for the subsequent recipient
530 test. It is now used for both.
532 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
533 patch removes a few documentation additions to RFC 3028, because the
534 latest draft now contains them. It adds the new en;ascii-case comparator
535 and a new error check for 8bit text in MIME parts. Comparator and
536 require names are now matched exactly. I enabled the subaddress
537 extension, but it is not well tested yet (read: it works for me)."
539 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
540 rework some of the code of TK/09 above to avoid the hardwired use of
541 "%lld" and "long long". Replaced the call to snprintf() with a call to
544 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
545 records point to non-existent hosts", "retry timeout exceeded", and
546 "retry time not reached for any host after a long failure period".
548 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
549 experimental DomainKeys support:
551 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
552 (2) On an error such as an illegally used "control", the wrong name for
553 the control was given.
555 These problems did NOT occur unless DomainKeys support was compiled.
557 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
559 PH/24 Added ${if match_ip condition.
561 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
562 databases so that it will be absolutely obvious if a crash occurs in the
563 DB library. This is a regular occurrence (often caused by mis-matched
566 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
567 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
568 picked up on a user's system that detects such things. There doesn't seem
569 to be a gcc warning option for this - only an attribute that has to be
570 put on the function's prototype. It seems that in Fedora Core 4 they have
571 set this on a number of new functions. No doubt there will be more in due
574 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
575 specify a transport. However, if an address that was verified by such a
576 router was the subject of a callout, Exim crashed because it tried to
577 read the rcpt_include_affixes from the non-existent transport. Now it
578 just assumes that the setting of that option is false. This bug was
579 introduced by 4.51/PH/31.
581 PH/28 Changed -d+all to exclude +memory, because that information is very
582 rarely of interest, but it makes the output a lot bigger. People tend to
583 do -d+all out of habit.
585 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
586 code in os-type was giving problems when libc.so lives in lib64, like on
589 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
590 aren't the modern standard, and it seems that some systems' include files
591 don't always have them. Exim was already checking for some of the newer
592 ones like T_AAAA, and defining it itself. I've added checks for all the
593 record types that Exim uses.
595 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
596 not automatically generating a new one, as it is supposed to. This
597 prevented TLS from working. If the file did exist, but contained invalid
598 data, a new version was generated, as expected. It was only the case of a
599 non-existent file that was broken.
601 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
602 with a change in libdomainkeys > 0.64.
604 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
605 from DNS. If the selector record carries the flag, it now has
606 precedence over the domain-wide flag.
608 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
610 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
611 the use of a port name that isn't defined in /etc/services) occurred, the
612 message was deferred as in a normal delivery, and thus remained on the
613 spool, instead of being failed because of the mua_wrapper setting. This
614 is now fixed, and I tidied up some of the mua_wrapper messages at the
617 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
618 the messages in a hash of arrays rather than using individual hashes.
619 This is a bit cleaner and results in dramatic memory savings, albeit
620 at a slight CPU cost.
622 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
623 as requested by Marc Sherman.
625 SC/10 Eximstats: added histograms for user specified patterns as requested
628 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
630 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
631 fopen() in the content-scanning modules that did not already have it.
637 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
638 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
640 TK/02 Fix ACL "control" statement not being available in MIME ACL.
642 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
644 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
645 to test Sieve filters that use "vacation".
647 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
648 that changes the way the GnuTLS parameters are stored in the cache file.
649 The new format can be generated externally. For backward compatibility,
650 if the data in the cache doesn't make sense, Exim assumes it has read an
651 old-format file, and it generates new data and writes a new file. This
652 means that you can't go back to an older release without removing the
655 PH/03 A redirect router that has both "unseen" and "one_time" set does not
656 work if there are any delivery delays because "one_time" forces the
657 parent to be marked "delivered", so its unseen clone is never tried
658 again. For this reason, Exim now forbids the simultaneous setting of
661 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
662 redirected to themselves ("homonym" addresses). Read the long ChangeLog
663 entry if you want to know the details. The fix, however, neglected to
664 consider the case when local delivery batching is involved. The test for
665 "previously delivered" was not happening when checking to see if an
666 address could be batched with a previous (undelivered) one; under
667 certain circumstances this could lead to multiple deliveries to the same
670 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
671 in its include files, and this causes problems building Exim.
673 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
674 header_syntax) but Exim was just ignoring anything given after a slash.
675 In particular, this caused confusion with an attempt to use "verify =
676 reverse_host_lookup/defer_ok". An error is now given when options are
677 supplied for verify items that do not have them. (Maybe reverse_host_
678 lookup should have a defer_ok option, but that's a different point.)
680 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
681 defined by RFC 821) to 2048, because there were problems with some AUTH
682 commands, and RFC 1869 says the size should be increased for extended
683 SMTP commands that take arguments.
685 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
688 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
689 "unknown" error; now it says that the functionality isn't in the binary.
691 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
692 an address' error message when a string expansion fails (syntax or
693 whatever). Otherwise the password may appear in the log. Following change
694 PH/42 below, there is no longer a chance of it appearing in a bounce
697 PH/11 Installed exipick version 20050225.0 from John Jetmore.
699 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
700 of its addresses was ever tried. (Bugzilla bug #2.)
702 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
703 the result incorrectly in the debug output. (It correctly added a newline
704 to what was transported.)
706 TF/01 Added $received_time.
708 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
709 commented out examples of how to interface to a virus scanner and to
710 SpamAssassin. Also added commented examples of av_scanner and
711 spamd_address settings.
713 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
714 and controls are allowed in which ACLs. There were a couple of minor
715 errors. Some of the entries in the conditions table (which is a table of
716 where they are NOT allowed) were getting very unwieldy; rewrote them as a
717 negation of where the condition IS allowed.
719 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
721 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
722 header file does not have a version number, so I've had to invent a new
723 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
724 API. The code is untested by me (my Linux distribution still has 0.3.2 of
725 radiusclient), but it was contributed by a Radius user.
727 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
728 files or directories, for OpenSSL.
730 PH/19 When an Exim process that is running as root has to create an Exim log
731 file, it does so in a subprocess that runs as exim:exim so as to get the
732 ownership right at creation (otherwise, other Exim processes might see
733 the file with the wrong ownership). There was no test for failure of this
734 fork() call, which would lead to the process getting stuck as it waited
735 for a non-existent subprocess. Forks do occasionally fail when resources
736 run out. I reviewed all the other calls to fork(); they all seem to check
739 PH/20 When checking for unexpected SMTP input at connect time (before writing
740 the banner), Exim was not dealing correctly with a non-positive return
741 from the read() function. If the client had disconnected by this time,
742 the result was a log entry for a synchronization error with an empty
743 string after "input=" when read() returned zero. If read() returned -1
744 (an event I could not check), uninitialized data bytes were printed.
745 There were reports of junk text (parts of files, etc) appearing after
748 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
750 PH/22 Added support for macro redefinition, and (re)definition in between
751 driver and ACL definitions.
753 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
754 forgetting to use the resulting value; it was using the unexpanded value.
756 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
757 hadn't been configured. The fix is from Juergen Kreileder, who
758 understands it better than I do:
760 "Here's what I see happening with three configured cyrus_sasl
761 authenticators configured (plain, login, cram-md5):
763 On startup auth_cyrus_sasl_init() gets called for each of these.
764 This means three calls to sasl_listmech() without a specified mech_list.
765 => SASL tests which mechs of all available mechs actually work
766 => three warnings about OTP not working
767 => the returned list contains: plain, login, cram-md5, digest-md5, ...
769 With the patch, sasl_listmech() also gets called three times. But now
770 SASL's mech_list option is set to the server_mech specified in the the
771 authenticator. Or in other words, the answer from sasl_listmech()
772 gets limited to just the mech you're testing for (which is different
774 => the return list contains just 'plain' or 'login', 'cram-md5' or
775 nothing depending on the value of ob->server_mech.
777 I've just tested the patch: Authentication still works fine,
778 unavailable mechs specified in the exim configuration are still
779 caught, and the auth.log warnings about OTP are gone."
781 PH/25 When debugging is enabled, the contents of the command line are added
782 to the debugging output, even when log_selector=+arguments is not
785 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
786 answer is "GNU", and only if the return is "GNU/something" is the answer
789 PH/27 $acl_verify_message is now set immediately after the failure of a
790 verification in an ACL, and so is available in subsequent modifiers. In
791 particular, the message can be preserved by coding like this:
793 warn !verify = sender
794 set acl_m0 = $acl_verify_message
796 Previously, $acl_verify_message was set only while expanding "message"
797 and "log_message" when a very denied access.
799 PH/28 Modified OS/os.c-Linux with
801 -#ifndef OS_LOAD_AVERAGE
802 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
804 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
805 nomenclature these days.)
807 PH/29 Installed patch from the Sieve maintainer that adds the options
808 sieve_useraddress and sieve_subaddress to the redirect router.
810 PH/30 In these circumstances:
811 . Two addresses routed to the same list of hosts;
812 . First host does not offer TLS;
813 . First host accepts first address;
814 . First host gives temporary error to second address;
815 . Second host offers TLS and a TLS session is established;
816 . Second host accepts second address.
817 Exim incorrectly logged both deliveries with the TLS parameters (cipher
818 and peerdn, if requested) that were in fact used only for the second
821 PH/31 When doing a callout as part of verifying an address, Exim was not paying
822 attention to any local part prefix or suffix that was matched by the
823 router that accepted the address. It now behaves in the same way as it
824 does for delivery: the affixes are removed from the local part unless
825 rcpt_include_affixes is set on the transport.
827 PH/32 Add the sender address, as F=<...>, to the log line when logging a
828 timeout during the DATA phase of an incoming message.
830 PH/33 Sieve envelope tests were broken for match types other than :is. I have
831 applied a patch sanctioned by the Sieve maintainer.
833 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
834 the uid or gid is negative. A case of a negative gid caused this to be
835 noticed. The fix allows for either to be negative.
837 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
838 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
841 PH/36 The change PH/12 above was broken. Fixed it.
843 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
844 the grounds that routing the same address twice would always produce the
845 same answer. This might have been true once, but it is certainly no
846 longer true now. Routing a child address may depend on the previous
847 routing that produced that child. Some complicated redirection strategies
848 went wrong when messages had multiple recipients, and made Exim's
849 behaviour dependent on the order in which the addresses were given.
851 I have moved the duplicate checking until after the routing is complete.
852 Exim scans the addresses that are assigned to local and remote
853 transports, and removes any duplicates. This means that more work will be
854 done, as duplicates will always all be routed, but duplicates are
855 presumably rare, so I don't expect this is of any significance.
857 For deliveries to pipes, files, and autoreplies, the duplicate checking
858 still happens during the routing process, since they are not going to be
861 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
862 It corrects a timeout issue with spamd. This is Ian's comment: "The
863 background is that sometimes spamd either never reads data from a
864 connection it has accepted, or it never writes response data. The exiscan
865 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
866 blindly assumes that writes won't block so it may never time out."
868 PH/39 Allow G after quota size as well as K and M.
870 PH/40 The value set for $authenticated_id in an authenticator may not contain
871 binary zeroes or newlines because the value is written to log lines and
872 to spool files. There was no check on this. Now the value is run through
873 the string_printing() function so that such characters are converted to
874 printable escape sequences.
876 PH/41 $message_linecount is a new variable that contains the total number of
877 lines in the message. Compare $body_linecount, which is the count for the
880 PH/42 Exim no longer gives details of delivery errors for specific addresses in
881 bounce and delay warning messages, except in certain special cases, which
884 (a) An SMTP error message from a remote host;
885 (b) A message specified in a :fail: redirection;
886 (c) A message specified in a "fail" command in a system filter;
887 (d) A message specified in a FAIL return from the queryprogram router;
888 (e) A message specified by the cannot_route_message router option.
890 In these cases only, Exim does include the error details in bounce and
891 warning messages. There are also a few cases where bland messages such
892 as "unrouteable address" or "local delivery error" are given.
894 PH/43 $value is now also set for the "else" part of a ${run expansion.
896 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
897 being worked on, but at least Exim now implements the latest version to
900 PH/45 In a pipe transport, although a timeout while waiting for the pipe
901 process to complete was treated as a delivery failure, a timeout while
902 writing the message to the pipe was logged, but erroneously treated as a
903 successful delivery. Such timeouts include transport filter timeouts. For
904 consistency with the overall process timeout, these timeouts are now
905 treated as errors, giving rise to delivery failures by default. However,
906 there is now a new Boolean option for the pipe transport called
907 timeout_defer, which, if set TRUE, converts the failures into defers for
908 both kinds of timeout. A transport filter timeout is now identified in
911 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
912 systems where "make" and "gmake" are different, calling "gmake" at top
913 level broke things. I've arranged for the value of $(MAKE) to be passed
914 from the Makefile to this script so that it can call the same version of
918 A note about Exim versions 4.44 and 4.50
919 ----------------------------------------
921 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
922 changes of various kinds. As a consequence, a big documentation update was
923 needed. This delayed the release for rather longer than seemed good, especially
924 in the light of a couple of (minor) security issues. Therefore, the changes
925 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
926 release. So 4.44 and 4.50 are in effect two different branches that both start
929 I have left the 4.50 change log unchanged; it contains all the changes since
930 4.43. The change log for 4.44 is below; many of its items are identical to
931 those for 4.50. This seems to be the most sensible way to preserve the
932 historical information.
938 1. Minor wording change to the doc/README.SIEVE file.
940 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
941 computation of the current number of files was incorrect.
943 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
944 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
947 4. Give more explanation in the error message when the command for a transport
948 filter fails to execute.
950 5. There are several places where Exim runs a non-Exim command in a
951 subprocess. The SIGUSR1 signal should be disabled for these processes. This
952 was being done only for the command run by the queryprogram router. It is
953 now done for all such subprocesses. The other cases are: ${run, transport
954 filters, and the commands run by the lmtp and pipe transports.
956 6. Added CONFIGURE_GROUP build-time option.
958 7. Some older OS have a limit of 256 on the maximum number of file
959 descriptors. Exim was using setrlimit() to set 1000 as a large value
960 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
961 systems. I've change it so that if it can't get 1000, it tries for 256.
963 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
964 was an oversight, and furthermore, ever since the addition of extra
965 controls (e.g. 4.43/32), the checks on when to allow different forms of
966 "control" were broken. There should now be diagnostics for all cases when a
967 control that does not make sense is encountered.
969 9. Added the /retain_sender option to "control=submission".
971 10. $recipients is now available in the predata ACL (oversight).
973 11. Tidy the search cache before the fork to do a delivery from a message
974 received from the command line. Otherwise the child will trigger a lookup
975 failure and thereby defer the delivery if it tries to use (for example) a
976 cached ldap connection that the parent has called unbind on.
978 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
979 of $address_data from the recipient verification was clobbered by the
982 13. The value of address_data from a sender verification is now available in
983 $sender_address_data in subsequent conditions in the ACL statement.
985 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
987 15. Added a new option "connect=<time>" to callout options, to set a different
990 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
991 was its contents. (It was OK if the option was not defined at all.)
993 17. A "Completed" log line is now written for messages that are removed from
994 the spool by the -Mrm option.
996 18. New variables $sender_verify_failure and $recipient_verify_failure contain
997 information about exactly what failed.
999 19. Added -dd to debug only the daemon process.
1001 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1002 handles timeouts, both on the server side and network timeouts. Renamed the
1003 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1005 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1006 It is now set to "smtps".
1008 22. $host_address is now set to the target address during the checking of
1009 ignore_target_hosts.
1011 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1012 being passed; this would have caused $sender_host_name to have been used if
1013 matching the list had actually called for a host name (not very likely,
1014 since this list is usually IP addresses). A host name is now passed as
1017 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1018 code that specifies a non-restarting handler (typically sigaction() in
1019 modern systems) in an attempt to fix a rare and obscure crash bug.
1021 25. Narrowed the window for a race in the daemon that could cause it to ignore
1022 SIGCHLD signals. This is not a major problem, because they are used only to
1023 wake it up if nothing else does.
1025 26. A malformed maildirsize file could cause Exim to calculate negative values
1026 for the mailbox size or file count. Odd effects could occur as a result.
1027 The maildirsize information is now recalculated if the size or filecount
1030 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1031 support for a long time. Removed HAVE_SYS_VFS_H.
1033 28. Installed the latest version of exipick from John Jetmore.
1035 29. In an address list, if the pattern was not a regular expression, an empty
1036 subject address (from a bounce message) matched only if the pattern was an
1037 empty string. Non-empty patterns were not even tested. This was the wrong
1038 because it is perfectly reasonable to use an empty address as part of a
1039 database query. An empty address is now tested by patterns that are
1040 lookups. However, all the other forms of pattern expect the subject to
1041 contain a local part and a domain, and therefore, for them, an empty
1042 address still always fails if the pattern is not itself empty.
1044 30. Exim went into a mad DNS loop when attempting to do a callout where the
1045 host was specified on an smtp transport, and looking it up yielded more
1046 than one IP address.
1048 31. Re-factored the code for checking spool and log partition space into a
1049 function that finds that data and another that does the check. The former
1050 is then used to implement four new variables: $spool_space, $log_space,
1051 $spool_inodes, and $log_inodes.
1053 32. The RFC2047 encoding function was originally intended for short strings
1054 such as real names; it was not keeping to the 75-character limit for
1055 encoded words that the RFC imposes. It now respects the limit, and
1056 generates multiple encoded words if necessary. To be on the safe side, I
1057 have increased the buffer size for the ${rfc2047: expansion operator from
1060 33. It is now permitted to omit both strings after an "if" condition; if the
1061 condition is true, the result is "true". As before, when the second string
1062 is omitted, a false condition yields an empty string. This makes it less
1063 cumbersome to write custom ACL and router conditions.
1065 34. Failure to deliver a bounce message always caused it to be frozen, even if
1066 there was an errors_to setting on the router. The errors_to setting is now
1069 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1070 canonical form (fully expanded) before being placed in
1071 $sender_host_address.
1073 36. The table in the code that translates DNS record types into text (T_A to
1074 "A" for instance) was missing entries for NS and CNAME. It is just possible
1075 that this could have caused confusion if both these types were looked up
1076 for the same domain, because the text type is used as part of Exim's
1077 per-process caching. But the chance of anyone hitting this buglet seems
1080 37. The dnsdb lookup has been extended in a number of ways.
1082 (1) There is a new type, "zns", which walks up the domain tree until it
1083 finds some nameserver records. It should be used with care.
1085 (2) There is a new type, "mxh", which is like "mx" except that it returns
1086 just the host names, not the priorities.
1088 (3) It is now possible to give a list of domains (or IP addresses) to be
1089 looked up. The behaviour when one of the lookups defers can be
1090 controlled by a keyword.
1092 (4) It is now possible to specify the separator character for use when
1093 multiple records are returned.
1095 38. The dnslists ACL condition has been extended: it is now possible to supply
1096 a list of IP addresses and/or domains to be looked up in a particular DNS
1099 39. Added log_selector=+queue_time_overall.
1101 40. When running the queue in the test harness, wait just a tad after forking a
1102 delivery process, to get repeatability of debugging output.
1104 41. Include certificate and key file names in error message when GnuTLS fails
1105 to set them up, because the GnuTLS error message doesn't include the name
1106 of the failing file when there is a problem reading it.
1108 42. Allow both -bf and -bF in the same test run.
1110 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1112 44. The "Exiscan patch" is now merged into the mainline Exim source.
1114 45. Sometimes the final signoff response after QUIT could fail to get
1115 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1116 < 0 before doing a fflush(). This bug looks as though it goes back to the
1117 introduction of TLS in release 3.20, but "sometimes" must have been rare
1118 because the tests only now provoked it.
1120 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1121 (this can affect the format of dates).
1123 47. exim_tidydb, when checking for the continued existence of a message for
1124 which it has found a message-specific retry record, was not finding
1125 messages that were in split spool directories. Consequently, it was
1126 deleting retry records that should have stayed in existence.
1128 48. Steve fixed some bugs in eximstats.
1130 49. The SPA authentication driver was not abandoning authentication and moving
1131 on to the next authenticator when an expansion was forced to fail,
1132 contradicting the general specification for all authenticators. Instead it
1133 was generating a temporary error. It now behaves as specified.
1135 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1136 (the order specifies the preference for clients). The order is now AES256,
1137 AES128, 3DES, ARCFOUR128.
1139 51. Small patch to Sieve code - explicitly set From: when generating an
1142 52. Exim crashed if a remote delivery caused a very long error message to be
1143 recorded - for instance if somebody sent an entire SpamAssassin report back
1144 as a large number of 550 error lines. This bug was coincidentally fixed by
1145 increasing the size of one of Exim's internal buffers (big_buffer) that
1146 happened as part of the Exiscan merge. However, to be on the safe side, I
1147 have made the code more robust (and fixed the comments that describe what
1150 53. Now that there can be additional text after "Completed" in log lines (if
1151 the queue_time_overall log selector is set), a one-byte patch to exigrep
1152 was needed to allow it to recognize "Completed" as not the last thing in
1155 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1156 patch that reportedly fixes this has been added. I am not expert enough to
1157 create a test for it. This is what the patch creator wrote:
1159 "I found a little strange behaviour of ldap code when working with
1160 Windows 2003 AD Domain, where users was placed in more than one
1161 Organization Units. When I tried to give exim partial DN, the exit code
1162 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1163 But simultaneously result of request was absolutely normal ldap result,
1164 so I produce this patch..."
1166 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1167 REFERENCE, so I have modified the code to exclude the patch when that macro
1170 55. Some experimental protocols are using DNS PTR records for new purposes. The
1171 keys for these records are domain names, not reversed IP addresses. The
1172 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1173 leaves it alone. Component reversal etc. now happens only for IP addresses.
1176 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1178 57. Double the size of the debug message buffer (to 2048) so that more of very
1179 long debug lines gets shown.
1181 58. The exicyclog utility now does better if the number of log files to keep
1182 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1184 59. Two changes related to the smtp_active_hostname option:
1186 (1) $smtp_active_hostname is now available as a variable.
1187 (2) The default for smtp_banner uses $smtp_active_hostname instead
1188 of $primary_hostname.
1190 60. The host_aton() function is supposed to be passed a string that is known
1191 to be a valid IP address. However, in the case of IPv6 addresses, it was
1192 not checking this. This is a hostage to fortune. Exim now panics and dies
1193 if the condition is not met. A case was found where this could be provoked
1194 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1195 components; fortuitously, this particular loophole had already been fixed
1196 by change 4.50/55 above.
1198 If there are any other similar loopholes, the new check in host_aton()
1199 itself should stop them being exploited. The report I received stated that
1200 data on the command line could provoke the exploit when Exim was running as
1201 exim, but did not say which command line option was involved. All I could
1202 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1203 running as the user.
1206 61. There was a buffer overflow vulnerability in the SPA authentication code
1207 (which came originally from the Samba project). I have added a test to the
1208 spa_base64_to_bits() function which I hope fixes it.
1211 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1212 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1214 63. The daemon start-up calls getloadavg() while still root for those OS that
1215 need the first call to be done as root, but it missed one case: when
1216 deliver_queue_load_max is set with deliver_drop_privilege. This is
1217 necessary for the benefit of the queue runner, because there is no re-exec
1218 when deliver_drop_privilege is set.
1220 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1221 This has been fixed.
1223 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1224 was in use, was not putting the data itself into the right store pool;
1225 consequently, it could be overwritten for a subsequent message in the same
1226 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1229 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1231 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1232 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1233 the function were treating the return as a boolean value, which happened to
1234 work because 0=false and not-0=true, but is not correct code.
1236 68. The host_aton() function was not handling scoped IPv6 addresses (those
1237 with, for example, "%eth0" on the end) correctly.
1239 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1240 negated items (that is, ~something) in unsigned ints. Some compilers
1241 apparently mutter when there is no cast.
1243 70. If an address verification called from an ACL failed, and did not produce a
1244 user-specific message (i.e. there was only a "system" message), nothing was
1245 put in $acl_verify_message. In this situation, it now puts the system
1248 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1249 session; change 4.31/43 added the unwanted input to the log line - except
1250 that it did not do this in the start of session case. It now does.
1252 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1253 This is wrong and can cause the other end to generate a synchronization
1254 error if it is another Exim or anything else that does the synchronization
1255 check. A QUIT command is no longer sent after a timeout.
1257 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1258 during host lookups.
1260 74. The defer_ok option of callout verification was not working if it was used
1261 when verifying addresses in header lines, that is, for this case:
1263 verify = header_sender/callout=defer_ok
1265 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1266 those file descriptors could be used for SMTP connections. If anything
1267 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1268 could be sent to the SMTP client, causing chaos. The daemon now opens
1269 stdin, stdout, and stderr to /dev/null when it puts itself into the
1272 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1273 log by default. The user can override this with suitable Perl magic.
1275 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1276 the log message when discard triggers, was not working for the DATA ACL or
1277 for the non-SMTP ACL.
1279 78. Error message wording change in sieve.c.
1281 79. If smtp_accept_max_per_host was set, the number of connections could be
1282 restricted to fewer than expected, because the daemon was trying to set up
1283 a new connection before checking whether the processes handling previous
1284 connections had finished. The check for completed processes is now done
1285 earlier. On busy systems, this bug wouldn't be noticed because something
1286 else would have woken the daemon, and it would have reaped the completed
1289 80. If a message was submitted locally by a user whose login name contained one
1290 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1291 It caused a spool format error. I have fixed the spool reading code. A
1292 related problem was that the "from" clause in the Received: line became
1293 illegal because of the space(s). It is now covered by ${quote_local_part.
1295 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1298 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1300 83. Patch from Sieve maintainer to fix unterminated string problem in
1301 "vacation" handling.
1303 84. Some minor changes to the Linux configuration files to help with other
1304 OS variants using glibc.
1306 85. One more patch for Sieve to update vacation handling to latest spec.
1309 ----------------------------------------------------
1310 See the note above about the 4.44 and 4.50 releases.
1311 ----------------------------------------------------
1317 1. Change 4.43/35 introduced a bug that caused file counts to be
1318 incorrectly computed when quota_filecount was set in an appendfile
1321 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1322 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1325 3. Give more explanation in the error message when the command for a transport
1326 filter fails to execute.
1328 4. There are several places where Exim runs a non-Exim command in a
1329 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1330 was being done only for the command run by the queryprogram router. It is
1331 now done for all such subprocesses. The other cases are: ${run, transport
1332 filters, and the commands run by the lmtp and pipe transports.
1334 5. Some older OS have a limit of 256 on the maximum number of file
1335 descriptors. Exim was using setrlimit() to set 1000 as a large value
1336 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1337 systems. I've change it so that if it can't get 1000, it tries for 256.
1339 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1340 was an oversight, and furthermore, ever since the addition of extra
1341 controls (e.g. 4.43/32), the checks on when to allow different forms of
1342 "control" were broken. There should now be diagnostics for all cases when a
1343 control that does not make sense is encountered.
1345 7. $recipients is now available in the predata ACL (oversight).
1347 8. Tidy the search cache before the fork to do a delivery from a message
1348 received from the command line. Otherwise the child will trigger a lookup
1349 failure and thereby defer the delivery if it tries to use (for example) a
1350 cached ldap connection that the parent has called unbind on.
1352 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1353 of $address_data from the recipient verification was clobbered by the
1354 sender verification.
1356 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1357 was its contents. (It was OK if the option was not defined at all.)
1359 11. A "Completed" log line is now written for messages that are removed from
1360 the spool by the -Mrm option.
1362 12. $host_address is now set to the target address during the checking of
1363 ignore_target_hosts.
1365 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1366 being passed; this would have caused $sender_host_name to have been used if
1367 matching the list had actually called for a host name (not very likely,
1368 since this list is usually IP addresses). A host name is now passed as
1371 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1372 code that specifies a non-restarting handler (typically sigaction() in
1373 modern systems) in an attempt to fix a rare and obscure crash bug.
1375 15. Narrowed the window for a race in the daemon that could cause it to ignore
1376 SIGCHLD signals. This is not a major problem, because they are used only to
1377 wake it up if nothing else does.
1379 16. A malformed maildirsize file could cause Exim to calculate negative values
1380 for the mailbox size or file count. Odd effects could occur as a result.
1381 The maildirsize information is now recalculated if the size or filecount
1384 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1385 support for a long time. Removed HAVE_SYS_VFS_H.
1387 18. Updated exipick to current release from John Jetmore.
1389 19. Allow an empty sender to be matched against a lookup in an address list.
1390 Previously the only cases considered were a regular expression, or an
1393 20. Exim went into a mad DNS lookup loop when doing a callout where the
1394 host was specified on the transport, if the DNS lookup yielded more than
1397 21. The RFC2047 encoding function was originally intended for short strings
1398 such as real names; it was not keeping to the 75-character limit for
1399 encoded words that the RFC imposes. It now respects the limit, and
1400 generates multiple encoded words if necessary. To be on the safe side, I
1401 have increased the buffer size for the ${rfc2047: expansion operator from
1404 22. Failure to deliver a bounce message always caused it to be frozen, even if
1405 there was an errors_to setting on the router. The errors_to setting is now
1408 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1409 canonical form (fully expanded) before being placed in
1410 $sender_host_address.
1412 24. Updated eximstats to version 1.33
1414 25. Include certificate and key file names in error message when GnuTLS fails
1415 to set them up, because the GnuTLS error message doesn't include the name
1416 of the failing file when there is a problem reading it.
1418 26. Expand error message when OpenSSL has problems setting up cert/key files.
1421 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1422 (this can affect the format of dates).
1424 28. exim_tidydb, when checking for the continued existence of a message for
1425 which it has found a message-specific retry record, was not finding
1426 messages that were in split spool directories. Consequently, it was
1427 deleting retry records that should have stayed in existence.
1429 29. eximstats updated to version 1.35
1430 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1431 1.35 - bugfix such that pie charts by volume are generated correctly
1433 30. The SPA authentication driver was not abandoning authentication and moving
1434 on to the next authenticator when an expansion was forced to fail,
1435 contradicting the general specification for all authenticators. Instead it
1436 was generating a temporary error. It now behaves as specified.
1438 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1439 (the order specifies the preference for clients). The order is now AES256,
1440 AES128, 3DES, ARCFOUR128.
1442 31. Small patch to Sieve code - explicitly set From: when generating an
1445 32. Exim crashed if a remote delivery caused a very long error message to be
1446 recorded - for instance if somebody sent an entire SpamAssassin report back
1447 as a large number of 550 error lines. This bug was coincidentally fixed by
1448 increasing the size of one of Exim's internal buffers (big_buffer) that
1449 happened as part of the Exiscan merge. However, to be on the safe side, I
1450 have made the code more robust (and fixed the comments that describe what
1453 33. Some experimental protocols are using DNS PTR records for new purposes. The
1454 keys for these records are domain names, not reversed IP addresses. The
1455 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1456 leaves it alone. Component reversal etc. now happens only for IP addresses.
1459 34. The host_aton() function is supposed to be passed a string that is known
1460 to be a valid IP address. However, in the case of IPv6 addresses, it was
1461 not checking this. This is a hostage to fortune. Exim now panics and dies
1462 if the condition is not met. A case was found where this could be provoked
1463 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1464 components; fortuitously, this particular loophole had already been fixed
1465 by change 4.50/55 or 4.44/33 above.
1467 If there are any other similar loopholes, the new check in host_aton()
1468 itself should stop them being exploited. The report I received stated that
1469 data on the command line could provoke the exploit when Exim was running as
1470 exim, but did not say which command line option was involved. All I could
1471 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1472 running as the user.
1475 35. There was a buffer overflow vulnerability in the SPA authentication code
1476 (which came originally from the Samba project). I have added a test to the
1477 spa_base64_to_bits() function which I hope fixes it.
1480 36. The daemon start-up calls getloadavg() while still root for those OS that
1481 need the first call to be done as root, but it missed one case: when
1482 deliver_queue_load_max is set with deliver_drop_privilege. This is
1483 necessary for the benefit of the queue runner, because there is no re-exec
1484 when deliver_drop_privilege is set.
1486 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1487 was in use, was not putting the data itself into the right store pool;
1488 consequently, it could be overwritten for a subsequent message in the same
1489 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1492 38. Sometimes the final signoff response after QUIT could fail to get
1493 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1494 < 0 before doing a fflush(). This bug looks as though it goes back to the
1495 introduction of TLS in release 3.20, but "sometimes" must have been rare
1496 because the tests only now provoked it.
1502 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1503 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1504 Now it returns an integer. A number of calls were still expecting a T/F
1505 return. Fortuitously, in all cases, the tests worked in OK situations,
1506 which is the norm. However, things would have gone wrong on any write
1507 failures on the smtp file descriptor. This function is used when sending
1508 messages over SMTP and also when doing verify callouts.
1510 2. When Exim is called to do synchronous delivery of a locally submitted
1511 message (the -odf or -odi options), it no longer closes stderr before doing
1514 3. Implemented the mua_wrapper option.
1516 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1518 5. Implemented the functions header_remove(), header_testname(),
1519 header_add_at_position(), and receive_remove_recipient(), and exported them
1522 6. If an ACL "warn" statement specified the addition of headers, Exim already
1523 inserted X-ACL-Warn: at the start if there was no header name. However, it
1524 was not making this test for the second and subsequent header lines if
1525 there were newlines in the string. This meant that an invalid header could
1526 be inserted if Exim was badly configured.
1528 7. Allow an ACL "warn" statement to add header lines at the start or after all
1529 the Received: headers, as well as at the end.
1531 8. Added the rcpt_4xx retry error code.
1533 9. Added postmaster_mailfrom=xxx to callout verification option.
1535 10. Added mailfrom=xxxx to the callout verification option, for verify=
1538 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1539 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1541 12. Inserted some casts to stop certain compilers warning when using pointer
1542 differences as field lengths or precisions in printf-type calls (mostly
1543 affecting debugging statements).
1545 13. Added optional readline() support for -be (dynamically loaded).
1547 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1548 same clock tick as a message's arrival, so that its received time was the
1549 same as the "first fail" time on the retry record, and that message
1550 remained on the queue past the ultimate address timeout, every queue runner
1551 would try a delivery (because it was past the ultimate address timeout) but
1552 after another failure, the ultimate address timeout, which should have then
1553 bounced the address, did not kick in. This was a "< instead of <=" error;
1554 in most cases the first failure would have been in the next clock tick
1555 after the received time, and all would be well.
1557 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1558 being recognized when the domain list was tested by the match_domain
1559 condition in an expansion string.
1561 16. Added the ${str2b64: operator.
1563 17. Exim was always calling setrlimit() to set a large limit for the number of
1564 processes, without checking whether the existing limit was already
1565 adequate. (It did check for the limit on file descriptors.) Furthermore,
1566 errors from getrlimit() and setrlimit() were being ignored. Now they are
1567 logged to the main and panic logs, but Exim does carry on, to try to do its
1568 job under whatever limits there are.
1570 18. Imported PCRE 5.0.
1572 19. Trivial typo in log message " temporarily refused connection" (the leading
1575 20. If the log selector return_path_on_delivery was set and an address was
1576 redirected to /dev/null, the delivery process crashed because it assumed
1577 that a return path would always be set for a "successful" delivery. In this
1578 case, the whole delivery is bypassed as an optimization, and therefore no
1581 21. Internal re-arrangement: the function for sending a challenge and reading
1582 a response while authentication was assuming a zero-terminated challenge
1583 string. It's now changed to take a pointer and a length, to allow for
1584 binary data in such strings.
1586 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1588 23. Exim was not respecting finduser_retries when seeking the login of the
1589 uid under which it was called; it was always trying 10 times. (The default
1590 setting of finduser_retries is zero.) Also, it was sleeping after the final
1591 failure, which is pointless.
1593 24. Implemented tls_on_connect_ports.
1595 25. Implemented acl_smtp_predata.
1597 26. If the domain in control=submission is set empty, Exim assumes that the
1598 authenticated id is a complete email address when it generates From: or
1599 Sender: header lines.
1601 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1602 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1603 chown and chgrp in /bin and hostname in /usr/bin.
1605 28. Exim was keeping the "process log" file open after each use, just as it
1606 does for the main log. This opens the possibility of it remaining open for
1607 long periods when the USR1 signal hits a daemon. Occasional processlog
1608 errors were reported, that could have been caused by this. Anyway, it seems
1609 much more sensible not to leave this file open at all, so that is what now
1612 29. The long-running daemon process does not normally write to the log once it
1613 has entered its main loop, and it closes the log before doing so. This is
1614 so that log files can straightforwardly be renamed and moved. However,
1615 there are a couple of unusual error situations where the daemon does write
1616 log entries, and I had neglected to close the log afterwards.
1618 30. The text of an SMTP error response that was received during a remote
1619 delivery was being truncated at 512 bytes. This is too short for some of
1620 the long messages that one sometimes sees. I've increased the limit to
1623 31. It is now possible to make retry rules that apply only when a message has a
1624 specific sender, in particular, an empty sender.
1626 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1627 it possible to be selective about when SMTP synchronization is enforced.
1629 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1631 32. Implemented hosts_connection_nolog.
1633 33. Added an ACL for QUIT.
1635 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1638 35. Added mailbox_size and mailbox_filecount to appendfile.
1640 36. Added control = no_multiline_responses to ACLs.
1642 37. There was a bug in the logic of the code that waits for the clock to tick
1643 in the case where the clock went backwards by a substantial amount such
1644 that the microsecond fraction of "now" was more than the microsecond
1645 fraction of "then" (but the whole seconds number was less).
1647 38. Added support for the libradius Radius client library this is found on
1648 FreeBSD (previously only the radiusclient library was supported).
1654 1. When certain lookups returned multiple values in the form name=value, the
1655 quoting of the values was not always being done properly. Specifically:
1656 (a) If the value started with a double quote, but contained no whitespace,
1658 (b) If the value contained whitespace other than a space character (i.e.
1659 tabs or newlines or carriage returns) it was not quoted.
1660 This fix has been applied to the mysql and pgsql lookups by writing a
1661 separate quoting function and calling it from the lookup code. The fix
1662 should probably also be applied to nisplus, ibase and oracle lookups, but
1663 since I cannot test any of those, I have not disturbed their existing code.
1665 2. A hit in the callout cache for a specific address caused a log line with no
1666 reason for rejecting RCPT. Now it says "Previous (cached) callout
1667 verification failure".
1669 3. There was an off-by-one bug in the queryprogram router. An over-long
1670 return line was truncated at 256 instead of 255 characters, thereby
1671 overflowing its buffer with the terminating zero. As well as fixing this, I
1672 have increased the buffer size to 1024 (and made a note to document this).
1674 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1675 when Exim is waiting for an SMTP response from a remote server, Exim
1676 restarts its select() call on the socket, thereby resetting its timeout.
1677 This is not a problem when such interrupts are rare. Somebody set up a cron
1678 job to run exiwhat every 2 minutes, which is less than the normal select()
1679 timeout (5 or 10 minutes). This meant that the select() timeout never
1680 kicked in because it was always reset. I have fixed this by comparing the
1681 time when an interrupt arrives with the time at the start of the first call
1682 to select(). If more time than the timeout has elapsed, the interrupt is
1683 treated as a timeout.
1685 5. Some internal re-factoring in preparation for the addition of Sieve
1686 extensions (by MH). In particular, the "personal" test is moved to a
1687 separate function, and given an option for scanning Cc: and Bcc: (which is
1688 not set for Exim filters).
1690 6. When Exim created an email address using the login of the caller as the
1691 local part (e.g. when creating a From: or Sender: header line), it was not
1692 quoting the local part when it contained special characters such as @.
1694 7. Installed new OpenBSD configuration files.
1696 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1697 try to make them clearer.
1699 9. Callout options, other than the timeout value, were being ignored when
1700 verifying sender addresses in header lines. For example, when using
1702 verify = header_sender/callout=no_cache
1704 the cache was (incorrectly) being used.
1706 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1707 only the Cygwin environment.
1709 11. When return_path_on_delivery was set as a log selector, if different remote
1710 addresses in the same message used different return paths and parallel
1711 remote delivery occurred, the wrong values would sometimes be logged.
1712 (Whenever a remote delivery process finished, the return path value from
1713 the most recently started remote delivery process was logged.)
1715 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1716 lines when AUTH and/or TLS are in use. This is the "received protocol"
1717 field. Exim used to use "asmtp" for authenticated SMTP, without any
1718 indication (in the protocol name) for TLS use. Now it follows the RFC and
1719 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1720 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1721 names appear in log lines as well as in Received: header lines.
1723 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1724 extensions, and comparison tests, and to fix some bugs.
1726 14. Changes to the "personal" filter test:
1728 (1) The test was buggy in that it was just doing the equivalent of
1729 "contains" tests on header lines. For example, if a user's address was
1730 anne@some.where, the "personal" test would incorrectly be true for
1732 To: susanne@some.where
1734 This test is now done by extracting each address from the header in turn,
1735 and checking the entire address. Other tests that are part of "personal"
1736 are now done using regular expressions (for example, to check local parts
1737 of addresses in From: header lines).
1739 (2) The list of non-personal local parts in From: addresses has been
1740 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1741 taken from the Sieve specification recommendations.
1743 (3) If the message contains any header line starting with "List-" it is
1744 treated as non-personal.
1746 (4) The test for "circular" in the Subject: header line has been removed
1747 because it now seems ill-conceived.
1749 15. Minor typos in src/EDITME comments corrected.
1751 16. Installed latest exipick from John Jetmore.
1753 17. If headers_add on a router specified a text string that was too long for
1754 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1755 of string_sprintf() is now avoided.
1757 18. $message_body_size was not set (it was always zero) when running the DATA
1758 ACL and the local_scan() function.
1760 19. For the "mail" command in an Exim filter, no default was being set for
1761 the once_repeat time, causing a random time value to be used if "once" was
1762 specified. (If the value happened to be <= 0, no repeat happened.) The
1763 default is now 0s, meaning "never repeat". The "vacation" command was OK
1764 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1765 (I found it when inspecting the code).
1767 20. There is now an overall timeout for performing a callout verification. It
1768 defaults to 4 times the callout timeout, which applies to individual SMTP
1769 commands during the callout. The overall timeout applies when there is more
1770 than one host that can be tried. The timeout is checked before trying the
1771 next host. This prevents very long delays if there are a large number of
1772 hosts and all are timing out (e.g. when the network connections are timing
1773 out). The value of the overall timeout can be changed by specifying an
1774 additional sub-option for "callout", called "maxwait". For example:
1776 verify = sender/callout=5s,maxwait=20s
1778 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1779 to the end before writing, but this should make it even safer).
1781 22. Exim was forgetting that it had advertised PIPELINING for the second and
1782 subsequent messages on an SMTP connection. It was also not resetting its
1783 memory on STARTTLS and an internal HELO.
1785 23. When Exim logs an SMTP synchronization error within a session, it now
1786 records whether PIPELINING has been advertised or not.
1788 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1789 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1790 rather than long int.
1792 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1794 26. Added the never_mail option to autoreply.
1800 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1801 crash if the getsockname() call failed; this can happen if a connection is
1802 closed very soon after it is established. The problem was simply in the
1803 order in which certain operations were done, causing Exim to try to write
1804 to the SMTP stream before it had set up the file descriptor. The bug has
1805 been fixed by making things happen in the correct order.
1811 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1812 before the connection was closed, thus losing the rejection response.
1814 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1815 some early Solaris releases, but causes trouble in current releases where
1816 socklen_t is defined.
1818 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1821 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1824 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1825 that releasing the top of it at the end releases what was used for sub-
1826 expansions (unless the block got too big). However, discard this block if
1827 the first thing is a variable or header, so that we can use its block when
1828 it is dynamic (useful for very large $message_headers, for example).
1830 6. Lookups now cache *every* query, not just the most recent. A new, separate
1831 store pool is used for this. It can be recovered when all lookup caches are
1832 flushed. Lookups now release memory at the end of their result strings.
1833 This has involved some general refactoring of the lookup sources.
1835 7. Some code has been added to the store_xxx() functions to reduce the amount
1836 of flapping under certain conditions.
1838 8. log_incoming_interface used to affect only the <= reception log lines. Now
1839 it causes the local interface and port to be added to several more SMTP log
1840 lines, for example "SMTP connection from", and rejection lines.
1842 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1844 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1846 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1847 could be overwritten at the end of the current message (or the start of a
1848 new message if it was set in a HELO ACL). The value is now preserved for
1849 the duration of the SMTP connection.
1851 12. If a transport had a headers_rewrite setting, and a matching header line
1852 contained an unqualified address, that address was qualified, even if it
1853 did not match any rewriting rules. The underlying bug was that the values
1854 of the flags that permit the existence of unqualified sender and recipient
1855 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1856 non-local messages, and by -bnq for local messages) were not being
1857 preserved with the message after it was received.
1859 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1860 "next input=" as part of the text comprising the host identity instead of
1861 the correct text. The code was using the same buffer for two different
1862 strings. However, depending on which order the printing function evaluated
1863 its arguments, the bug did not always show up. Under Linux, for example, my
1864 test suite worked just fine.
1866 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1867 doesn't work with some older versions of Perl. It has been changed to "my",
1868 which in any case is probably the better facility to use.
1870 15. A really picky compiler found some instances of statements for creating
1871 error messages that either had too many or two few arguments for the format
1874 16. The size of the buffer for calls to the DNS resolver has been increased
1875 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1876 for addresses that have a lot of PTR records. This alleviates a problem; it
1877 does not fully solve it.
1879 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1880 buffer now truncates the list and logs the incident, which is the same
1881 action as happens when Exim is looking up a host name and its aliases.
1882 Previously in this situation something unpredictable would happen;
1883 sometimes it was "internal error: store_reset failed".
1885 18. If a server dropped the connection unexpectedly when an Exim client was
1886 using GnuTLS and trying to read a response, the client delivery process
1887 crashed while trying to generate an error log message.
1889 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1890 string, for example:
1892 warn message = H1: something\nH2: something
1894 the text was added as a single header line from Exim's point of view
1895 though it ended up OK in the delivered message. However, searching for the
1896 second and subsequent header lines using $h_h2: did not work. This has been
1897 fixed. Similarly, if a system filter added multiple headers in this way,
1898 the routers could not see them.
1900 20. Expanded the error message when iplsearch is called with an invalid key to
1901 suggest using net-iplsearch in a host list.
1903 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1904 ACLs are no longer actually imposed (and a message to that effect is
1907 22. If a "gecos" field in a passwd entry contained escaped characters, in
1908 particular, if it contained a \" sequence, Exim got it wrong when building
1909 a From: or a Sender: header from that name. A second bug also caused
1910 incorrect handling when an unquoted " was present following a character
1911 that needed quoting.
1913 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1914 was not being matched caselessly.
1916 24. Arranged for all hyphens in the exim.8 source to be escaped with
1919 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1920 properly. Recipient callout cache records were still being keyed to include
1921 the sender, even when use_sender was set false. This led to far more
1922 callouts that were necessary. The sender is no longer included in the key
1923 when use_sender is false.
1925 26. Added "control = submission" modifier to ACLs.
1927 27. Added the ${base62d: operator to decode base 62 numbers.
1929 28. dnsdb lookups can now access SRV records.
1931 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1932 the configuration file.
1934 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1935 (-v) mode. This makes the output for a verbose queue run more intelligible.
1937 31. Added a use_postmaster feature to recipient callouts.
1939 32. Added the $body_zerocount variable, containing the number of binary zero
1940 bytes in the message body.
1942 33. The time of last modification of the "new" subdirectory is now used as the
1943 "mailbox time last read" when there is a quota error for a maildir
1946 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1948 35. Added +ignore_unknown as a special item in host lists.
1950 36. Code for decoding IPv6 addresses in host lists is now included, even if
1951 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1952 address was recognized as an IP address, but was then not correctly decoded
1953 into binary, causing unexpected and incorrect effects when compared with
1960 1. Very minor rewording of debugging text in manualroute to say "list of
1961 hosts" instead of "hostlist".
1963 2. If verify=header_syntax was set, and a header line with an unqualified
1964 address (no domain) and a large number of spaces between the end of the
1965 name and the colon was received, the reception process suffered a buffer
1966 overflow, and (when I tested it) crashed. This was caused by some obsolete
1967 code that should have been removed. The fix is to remove it!
1969 3. When running in the test harness, delay a bit after writing a bounce
1970 message to get a bit more predictability in the log output.
1972 4. Added a call to search_tidyup() just before forking a reception process. In
1973 theory, someone could use a lookup in the expansion of smtp_accept_max_
1974 per_host which, without the tidyup, could leave open a database connection.
1976 5. Added the variables $recipient_data and $sender_data which get set from a
1977 lookup success in an ACL "recipients" or "senders" condition, or a router
1978 "senders" option, similar to $domain_data and $local_part_data.
1980 6. Moved the writing of debug_print from before to after the "senders" test
1983 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1984 problems for message scanning, either using a data ACL, or using
1985 local_scan() because the Received: header was not generated till after they
1986 were called (in order to set the time as the time of reception completion).
1987 I have revised the way this works. The header is now generated after the
1988 body is received, but before the ACL or local_scan() are called. After they
1989 are run, the timestamp in the header is updated.
1995 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1996 before starting a queue runner without re-exec. This happened only when
1997 deliver_drop_privilege was set or when the Exim user was set to root. The
1998 effect of the bug was that timeouts during subsequent deliveries caused
1999 crashes instead of being properly handled. The handler is now left at its
2000 default (and expected) setting.
2002 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2003 message, again when deliver_drop_privilege is set or Exim is run as root.
2004 The bug described in (1) was not present in this case, but the tidying up
2005 of the other signals was missing. I have made the two cases consistent.
2007 3. The ignore_target_hosts setting on a manualroute router was being ignored
2008 for hosts that were looked up using the /MX notation.
2010 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2013 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2014 operated on the sender address. After changing the $sender_address to <>
2015 for the sender address verify, Exim was re-instated it as the original
2016 (before rewriting) address, but remembering that it had rewritten it, so it
2017 wasn't rewriting it again. This bug also had the effect of breaking the
2018 sender address verification caching when the sender address was rewritten.
2020 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2021 This has been changed so that if the ip literal address matches
2022 ignore_target_hosts, the router declines.
2024 7. Added expansion conditions match_domain, match_address, and match_local_
2025 part (NOT match_host).
2027 8. The placeholder for the Received: header didn't have a length field set.
2029 9. Added code to Exim itself and to exim_lock to test for a specific race
2030 condition that could lead to file corruption when using MBX delivery. The
2031 issue is with the lockfile that is created in /tmp. If this file is removed
2032 after a process has opened it but before that process has acquired a lock,
2033 there is the potential for a second process to recreate the file and also
2034 acquire a lock. This could lead to two Exim processes writing to the file
2035 at the same time. The added code performs the same test as UW imapd; it
2036 checks after acquiring the lock that its file descriptor still refers to
2037 the same named file.
2039 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2040 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2043 11. Added the smtp_active_hostname option. If used, this will typically be made
2044 to depend on the incoming interface address. Because $interface_address is
2045 not set up until the daemon has forked a reception process, error responses
2046 that can happen earlier (such as "too many connections") no longer contain
2049 12. If an expansion in a condition on a "warn" statement fails because a lookup
2050 defers, the "warn" statement is abandoned, and the next ACL statement is
2051 processed. Previously this caused the whole ACL to be aborted.
2053 13. Added the iplsearch lookup type.
2055 14. Added ident_timeout as a log selector.
2057 15. Added tls_certificate_verified as a log selector.
2059 16. Added a global option tls_require_ciphers (compare the smtp transport
2060 option of the same name). This controls incoming TLS connections.
2062 17. I finally figured out how to make tls_require_ciphers do a similar thing
2063 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2064 before starting the TLS session.
2066 18. Tabs are now shown as \t in -bP output.
2068 19. If the log selector return_path_on_delivery was set, Exim crashed when
2069 bouncing a message because it had too many Received: header lines.
2071 20. If two routers both had headers_remove settings, and the first one included
2072 a superfluous trailing colon, the final name in the first list and the
2073 first name in the second list were incorrectly joined into one item (with a
2074 colon in the middle).
2080 1. Added -C and -D options to the exinext utility, mainly to make it easier
2081 to include in the automated testing, but these could be helpful when
2082 multiple configurations are in use.
2084 2. The exinext utility was not formatting the output nicely when there was
2085 an alternate port involved in the retry record key, nor when there was a
2086 message id as well (for retries that were specific to a specific message
2087 and a specific host). It was also confused by IPv6 addresses, because of
2088 the additional colons they contain. I have fixed the IPv4 problem, and
2089 patched it up to do a reasonable job for IPv6.
2091 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2092 delivery, the log line now contains "pipelined" if PIPELINING was used.
2094 4. An SMTP transport process used to panic and die if the bind() call to set
2095 an explicit outgoing interface failed. This has been changed; it is now
2096 treated in the same way as a connect() failure.
2098 5. A reference to $sender_host_name in the part of a conditional expansion
2099 that was being skipped was still causing a DNS lookup. This no longer
2102 6. The def: expansion condition was not recognizing references to header lines
2103 that used bh_ and bheader_.
2105 7. Added the _cache feature to named lists.
2107 8. The code for checking quota_filecount in the appendfile transport was
2108 allowing one more file than it should have been.
2110 9. For compatibility with Sendmail, the command line option
2118 and sets the incoming protocol and host name (for trusted callers). The
2119 host name and its colon can be omitted when only the protocol is to be set.
2120 Note the Exim already has two private options, -pd and -ps, that refer to
2121 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2122 "s", but I don't think that's a major issue.
2124 10. A number of refactoring changes to the code, none of which should affect
2127 (a) The number of logging options was getting close to filling up the
2128 32-bit word that was used as a bit map. I have split them into two classes:
2129 those that are passed in the argument to log_write(), and those that are
2130 only ever tested independently outside of that function. These are now in
2131 separate 32-bit words, so there is plenty of room for expansion again.
2132 There is no change in the user interface or the logging behaviour.
2134 (b) When building, for example, log lines, the code previously used a
2135 macro that called string_cat() twice, in order to add two strings. This is
2136 not really sufficiently general. Furthermore, there was one instance where
2137 it was actually wrong because one of the argument was used twice, and in
2138 one call a function was used. (As it happened, calling the function twice
2139 did not affect the overall behaviour.) The macro has been replaced by a
2140 function that can join an arbitrary number of extra strings onto a growing
2143 (c) The code for expansion conditions now uses a table and a binary chop
2144 instead of a serial search (which was left over from when there were very
2145 few conditions). Also, it now recognizes conditions like "pam" even when
2146 the relevant support is not compiled in: a suitably worded error message is
2147 given if an attempt is made to use such a condition.
2149 11. Added ${time_interval:xxxxx}.
2151 12. A bug was causing one of the ddress fields not to be passed back correctly
2152 from remote delivery subprocesses. The field in question was not being
2153 subsequently used, so this caused to problems in practice.
2155 13. Added new log selectors queue_time and deliver_time.
2157 14. Might have fixed a bug in maildirsizefile handling that threw up
2158 "unexpected character" debug warnings, and recalculated the data
2159 unnecessarily. In any case, I expanded the warning message to give more
2162 15. Added the message "Restricted characters in address" to the statements in
2163 the default ACL that block characters like @ and % in local parts.
2165 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2166 Three changes have been made:
2168 (a) There was a serious bug; a negative response to MAIL caused the whole
2169 recipient domain to be cached as invalid, thereby blocking all messages
2170 to all local parts at the same domain, from all senders. This bug has
2171 been fixed. The domain is no longer cached after a negative response to
2172 MAIL if the sender used is not empty.
2174 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2177 (c) A new callout option, "use_sender" has been added for people who want
2178 the modified behaviour.
2184 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2187 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2188 indeed breaks things for older releases.
2190 3. Added additional logging to the case where there is a problem reading data
2191 from a filter that is running in a subprocess using a pipe, in order to
2192 try to track down a specific problem.
2194 4. Testing facility fudge: when running in the test harness and attempting
2195 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2196 getting "No route to host". Convert this to a timeout.
2198 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2201 6. Some OS don't have socklen_t but use size_t instead. This affects the
2202 fifth argument of getsockopt() amongst other things. This is now
2203 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2204 can be set for individual OS. I have set it for SunOS5, OSF1, and
2205 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2206 some earlier ones do not.
2208 7. Change 4.30/15 was not doing the test caselessly.
2210 8. The standard form for an IPv6 address literal was being rejected by address
2211 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2212 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2213 this, as well as the form without the "IPv6" on the front (but only when
2214 address literals are enabled, of course).
2216 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2218 10. Exim crashed if a message with an empty sender address specified by -f
2219 encountered a router with an errors_to setting. This could be provoked only
2220 by a command such as
2224 where an empty string was supplied; "<>" did not hit this bug.
2226 11. Installed PCRE release 4.5.
2228 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2229 remained set. It is now erased.
2231 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2232 times from message ids (which are base 36 rather than the normal 62).
2234 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2235 were being counted as actual protocol errors, and logged if the log
2236 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2237 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2238 and DATA following a set of rejected RCPTs do not count as protocol errors.
2239 In other words, Exim assumes they were pipelined, though this may not
2240 actually be the case. Of course, in all cases the client gets an
2241 appropriate error code.
2243 15. If a lookup fails in an ACL condition, a message about the failure may
2244 be available; it is used if testing the ACL cannot continue, because most
2245 such messages specify what the cause of the deferral is. However, some
2246 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2247 that caused an old message to be retained and used if a later statement
2248 caused a defer, replacing the real cause of the deferral.
2250 16. If an IP address had so many PTR records that the DNS lookup buffer
2251 was not large enough to hold them, Exim could crash while trying to process
2252 the truncated data. It now detects and logs this case.
2254 17. Further to 4.21/58, another change has been made: if (and only if) the
2255 first line of a message (the first header line) ends with CRLF, a bare LF
2256 in a subsequent header line has a space inserted after it, so as not to
2257 terminate the header.
2259 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2260 unnecessarily, used atoi() instead of strtol(), and didn't check the
2261 termination when getting file sizes from file names by regex.
2263 19. Completely re-implemented the support for maildirsize files, in the light
2264 of a number of problems with the previous contributed implementation
2265 (4.30/29). In particular:
2267 . If the quota is zero, the maildirsize file is maintained, but no quota is
2270 . If the maildir directory does not exist, it is created before any attempt
2271 to write a maildirsize file.
2273 . The quota value in the file is just a cache; if the quota is changed in
2274 the transport, the new value overrides.
2276 . A regular expression is available for excluding directories from the
2279 20. The autoreply transport checks the characters in options that define the
2280 message's headers; it allows continued headers, but it was checking with
2281 isspace() after an embedded newline instead of explicitly looking for a
2284 21. If all the "regular" hosts to which an address was routed had passed their
2285 expiry times, and had not reached their retry times, the address was
2286 bounced, even if fallback hosts were defined. Now Exim should go on to try
2289 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2290 equivalent code in the SMTP transport. Some hosts send humungous responses
2291 to HELO/EHLO, more than 1024 it seems.
2293 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2294 gives compiler warnings in some environments. I've now done it "properly",
2297 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2298 (because of problems with the built-in one) was declared to return uschar *
2299 instead of char *, causing compiler failure.
2301 25. Fixed a file descriptor leak when processing alias/forward files.
2303 26. Fixed a minor format string issue in dbfn.c.
2305 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2307 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2308 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2311 29. When a sender address is verified, it is cached, to save repeating the test
2312 when there is more than one recipient in a message. However, when the
2313 verification involves a callout, it is possible for different callout
2314 options to be set for different recipients. It is too complicated to keep
2315 track of this in the cache, so now Exim always runs a verification when a
2316 callout is required, relying on the callout cache for the optimization.
2317 The overhead is duplication of the address routing, but this should not be
2320 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2321 to be verified with callout=postmaster, and the main callout worked but the
2322 postmaster check failed, the verification correctly failed. However, if a
2323 subsequent RCPT command asked for sender verification *without* the
2324 postmaster check, incorrect caching caused this verification also to fail,
2327 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2328 it was not caching the DNS options (qualify_single, search_parents) that
2329 were used when the lookup failed. A subsequent lookup with different
2330 options therefore always gave the same answer, though there were cases
2331 where it should not have. (Example: a "domains = !$mx_any" option on a
2332 dnslookup router: the "domains" option is always processed without any
2333 widening, but the router might have qualify_single set.) Now Exim uses the
2334 cached value only when the same options are set.
2336 32. Added John Jetmore's "exipick" utility to the distribution.
2338 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2339 than a timeout (e.g. a certificate is required, and is not provided), an
2340 Exim server now closes the connection immediately. Previously it waited for
2341 the client to close - but if the client is SSL, it seems that they each
2342 wait for each other, leading to a delay before one of them times out.
2344 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2345 maintained 0.8.x compatibility because I don't think many are using it, and
2346 it is clearly obsolete.
2348 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2351 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2352 client certificate was expired. A simple patch fixes this, though I don't
2353 understand the full logic of why the verify callback is called multiple
2356 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2357 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2358 which causes problems with some clients (such as the Certicom SSL Plus
2359 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2360 disables the coutermeasure allowing Eudora to connect."
2362 38. Exim was not checking that a write() to a log file succeeded. This could
2363 lead to Bad Things if a log got too big, in particular if it hit a file
2364 size limit. Exim now panics and dies if it cannot write to a log file, just
2365 as it does if it cannot open a log file.
2367 39. Modified OS/Makefile-Linux so that it now contains
2369 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2371 The two -D definitions ensure that Exim is compiled with large file
2372 support, which makes it possible to handle log files that are bigger than
2375 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2376 instance) a domain was checked against a named list that involved a lookup,
2377 causing $domain_data to be set, then another domain was checked against the
2378 same list, then the first domain was re-checked, the value of $domain_data
2379 after the final check could be wrong. In particular, if the second check
2380 failed, it could be set empty. This bug probably also applied to
2383 41. The strip_trailing_dot option was not being applied to the address given
2384 with the -f command-line option.
2386 42. The code for reading a message's header from the spool was incrementing
2387 $received_count, but never initializing it. This meant that the value was
2388 incorrect (doubled) while delivering a message in the same process in which
2389 it was received. In the most common configuration of Exim, this never
2390 happens - a fresh exec is done - but it can happen when
2391 deliver_drop_privilege is set.
2393 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2394 it now includes up to 150 characters of the unexpected data in the log
2397 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2398 and building data strings. The size of both of these buffers was 10 000
2399 bytes - far larger than anybody would *ever* want, thought I. Needless to
2400 say, somebody hit the limit. I have increased the maximum line length to
2401 20 000 and the maximum data length of concatenated lines to 100 000. I have
2402 also fixed two bugs, because there was no checking on these buffers. Tsk,
2403 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2404 buffer is too small.
2406 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2407 lsearch lookups. Now it does.
2409 46. When parsing a route_list item in a manualroute router, a fixed-length
2410 buffer was used for the list of hosts. I made this 1024 bytes long,
2411 thinking that nobody would ever have a list of hosts that long. Wrong.
2412 Somebody had a whole pile of complicated expansion conditions, and the
2413 string was silently truncated, leading to an expansion error. It turns out
2414 that it is easier to change to an unlimited length (owing to other changes
2415 that have happened since this code was originally written) than to build
2416 structure for giving a limitation error. The length of the item that
2417 expands into the list of hosts is now unlimited.
2419 47. The lsearch lookup could not handle data where the length of text line was
2420 more than 4095 characters. Such lines were truncated, leading to shortened
2421 data being returned. It should now handle lines of any length.
2423 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2424 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2427 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2428 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2429 no longer happens in comment lines. A list of replacements is now placed
2430 at the head of all of the source files, except those whose only change is
2431 to replace PERL_COMMAND in the very first #! line.
2433 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2434 messages on the queue, with a bottom-up merge sort, using code contributed
2435 by Michael Haardt. This should make operations like -bp somewhat faster on
2436 large queues. It won't affect queue runners, except when queue_run_in_order
2439 51. Installed eximstats 1.31 in the distribution.
2441 52. Added support for SRV lookups to the dnslookup router.
2443 53. If an ACL referred to $message_body or $message_body_end, the value was not
2444 reset for any messages that followed in the same SMTP session.
2446 54. The store-handling optimization for building very long strings was not
2447 differentiating between the different store pools. I don't think this
2448 actually made any difference in practice, but I've tidied it.
2450 55. While running the routers to verify a sender address, $sender_address
2451 was still set to the sender address. This is wrong, because when routing to
2452 send a bounce to the sender, it would be empty. Therefore, I have changed
2453 it so that, while verifying a sender address, $sender_address is set to <>.
2454 (There is no change to what happens when verifying a recipient address.)
2456 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2457 target A or AAAA records (if not already returned) without resetting the
2458 qualify_single or search_parents options of the DNS resolver. These are
2459 inappropriate in this case because the targets of MX and SRV records must
2460 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2461 target that, when qualified, matched something in the local domain. These
2462 two options are now turned off when doing these lookups.
2464 57. It seems that at least some releases of Reiserfs (which does not have the
2465 concept of a fixed number of inodes) returns zero and not -1 for the
2466 number of available inodes. This interacted badly with check_spool_inodes,
2467 which assumed that -1 was the "no such thing" setting. What I have done is
2468 to check that the total number of inodes is greater than zero before doing
2469 the test of how many are available.
2471 58. When a "warn" ACL statement has a log_message modifier, the message is
2472 remembered, and not repeated. This is to avoid a lot of repetition when a
2473 message has many recipients that cause the same warning to be written.
2474 Howewer, Exim was preserving the list of already written lines for an
2475 entire SMTP session, which doesn't seem right. The memory is now reset if a
2476 new message is started.
2478 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2479 debugging output unless log_rewrite was also set.
2481 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2482 of (int)(handle) when we know that handle contains (void *)(-1).
2484 61. The Exim daemon panic-logs an error return when it closes the incoming
2485 connection. However "connection reset by peer" seems to be common, and
2486 isn't really an error worthy of noting specially, so that particular error
2489 62. When Exim is trying to find all the local interfaces, it used to panic and
2490 die if the ioctl to get the interface flags failed. However, it seems that
2491 on at least one OS (Solaris 9) it is possible to have an interface that is
2492 included in the list of interfaces, but for which you get a failure error
2493 for this call. This happens when the interface is not "plumbed" into a
2494 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2495 failure of the "get flags" call assumes that the interface is down.
2497 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2498 makes life easier for people who are doing arithmetic on fields extracted
2499 from dates, where you often get leading zeros that should not be
2500 interpreted as octal.
2502 64. Added qualify_domain to the redirect router, to override the global
2505 65. If a pathologically long header line contained very many addresses (the
2506 report of this problem mentioned 10 000) and each of them was rewritten,
2507 Exim could use up a very large amount of memory. (It kept on making new
2508 copies of the header line as it rewrote, and never released the old ones.)
2509 At the expense of a bit more processing, the header rewriting function has
2510 been changed so that it no longer eats memory in this way.
2512 66. The generation of the Received: header has been moved from the time that a
2513 message starts to be received, to the time that it finishes. The timestamp
2514 in the Received: header should now be very close to that of the <= log
2515 line. There are two side-effects of this change:
2517 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2518 logged header lines no longer include the local Received: line, because
2519 it has not yet been created. The same applies to a copy of the message
2520 that is returned to a non-SMTP sender when a message is rejected.
2522 (b) When a filter file is tested using -bf, no additional Received: header
2523 is added to the test message. After some thought, I decided that this
2526 This change does not affect the value of $received_for. It is still set
2527 after address rewriting, but before local_scan() is called.
2529 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2531 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2532 gave an unhelpful panic error message, and a defer error. I have managed to
2533 change this behaviour so that it now rejects any supplied certificate,
2534 which seems right, as the list of acceptable certificates is empty.
2536 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2537 gave an unhelpful defer error. I have not managed to make this reject any
2538 supplied certificates, but the error message it gives is "no certificate
2539 supplied", which is not helpful.
2541 70. exigrep's output now also includes lines that are not associated with any
2542 message, but which match the given pattern. Implemented by a patch from
2543 Martin Sluka, which also tidied up the Perl a bit.
2545 71. Recipient callout verification, like sender verification, was using <> in
2546 the MAIL FROM command. This isn't really the right thing, since the actual
2547 sender may affect whether the remote host accepts the recipient or not. I
2548 have changed it to use the actual sender in the callout; this means that
2549 the cache record is now keyed on a recipient/sender pair, not just the
2550 recipient address. There doesn't seem to be a real danger of callout loops,
2551 since a callout by the remote host to check the sender would use <>.
2552 [SEE ABOVE: changed after hitting problems.]
2554 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2555 temporary errors. However, in the case of such a code being given after
2556 the end of a data transmission (i.e. after ".") Exim was failing to write
2557 a retry record for the message. (Yes, there was some broken host that was
2558 actually sending 8xx at this point.)
2560 73. An unknown lookup type in a host list could cause Exim to panic-die when
2561 the list was checked. (An example that provoked this was putting <; in the
2562 middle of a list instead of at the start.) If this happened during a DATA
2563 ACL check, a -D file could be left lying around. This kind of configuration
2564 error no longer causes Exim to die; instead it causes a defer errror. The
2565 incident is still logged to the main and panic logs.
2567 74. Buglet left over from Exim 3 conversion. The message "too many messages
2568 in one connection" was written to the rejectlog but not the mainlog, except
2569 when address rewriting (yes!) was being logged.
2571 75. Added write_rejectlog option.
2573 76. When a system filter was run not as root (that is, when system_filter_user
2574 was set), the values of the $n variables were not being returned to the
2575 main process; thus, they were not subsequently available in the $sn
2578 77. Added +return_path_on_delivery log selector.
2580 78. A connection timeout was being treated differently from recipients deferred
2581 when testing hosts_max_try with a message that was older than the host's
2582 retry timeout. (The host should not be counted, thus allowing all hosts to
2583 be tried at least once before bouncing.) This may have been the cause of an
2584 occasionally reported bug whereby a message would remain on the queue
2585 longer than the retry timeout, but would be bounced if a delivery was
2586 forced. I say "may" because I never totally pinned down the problem;
2587 setting up timeout/retry tests is difficult. See also the next item.
2589 79. The ultimate address timeout was not being applied to errors that involved
2590 a combination of host plus message (for example, a timeout on a MAIL
2591 command). When an address resolved to a number of possible hosts, and they
2592 were not all tried for each delivery (e.g. because of hosts_max_try), a
2593 message could remain on the queue longer than the retry timeout.
2595 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2598 81. Fixed an obscure SMTP outgoing bug which required at least the following
2599 conditions: (a) there was another message waiting for the same server;
2600 (b) the server returned 5xx to all RCPT commands in the first message so
2601 that the message was not completed; (c) the server dropped the connection
2602 or gave a negative response to the RSET that Exim sends to abort the
2603 transaction. The observed case was a dropped connection after DATA that had
2604 been sent in pipelining mode. That is, the server had advertised PIPELINING
2605 but was not implementing it correctly. The effect of the bug was incorrect
2606 behaviour, such as trying another host, and this could lead to a crash.
2612 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2613 and daemon.c were passed as pointers to ints; they should have been
2614 pointers to socklen_t variables (which are typically unsigned ints).
2616 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2619 3. Fixed a really odd bug that affected only the testing scheme; patching a
2620 certain fixed string in the binary changed the value of another string that
2621 happened to be identical to the end of the original first string.
2623 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2624 name", it returns that address as the IP address. On some operating
2625 systems (e.g. Solaris), it also passes back the IP address string as the
2626 "host name". However, on others (e.g. Linux), it passes back an empty
2627 string. Exim wasn't checking for this, and was changing the host name to an
2628 empty string, assuming it had been canonicized.
2630 5. Although rare, it is permitted to have more than one PTR record for a given
2631 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2632 all the names associated with an address, because they do in Solaris.
2633 However, it seems that they do not in Linux for data that comes from the
2634 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2635 I found this out when I moved to a new Linux workstation and tried to run
2636 the Exim test suite.
2638 To get round this problem I have changed the code so that it now does its
2639 own call to the DNS to look up PTR records when searching for a host name.
2640 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2641 addresses that are only in /etc/hosts are still found.
2643 This behaviour is, however, controlled by an option called host_lookup_
2644 order, which defaults to "bydns:byaddr". If people want to use the other
2645 order, or indeed, just use one or the other means of lookup, they can
2646 specify it in this variable.
2648 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2649 some operating systems, this comes back from gethostbyaddr() as an empty
2650 string, and this is what Exim used to test for. However, it seems that in
2651 other systems, "." is yielded. Exim now tests for this case too.
2653 7. The values of check_spool_space and check_log_space are now held internally
2654 as a number of kilobytes instead of an absolute number of bytes. If a
2655 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2656 kilobyte. This means that much larger values can be stored.
2658 8. Exim monitor: an attempt to get the action menu when not actually pointing
2659 at a message produces an empty menu entitled "No message selected". This
2660 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2661 no entries in it ("Shell widget menu has zero width and/or height"). So I
2662 have added a single, blank menu entry in this case.
2664 9. Added ${quote_local_part.
2666 10. MIME decoding is now applied to the contents of Subject: header lines when
2669 11. Now that a reference to $sender_host_address automatically causes a reverse
2670 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2671 host lookup before query-style lookups in lists that might use this
2672 variable. This has therefore been abolished, and the "net-" prefix is no
2673 longer necessary for query-style lookups.
2675 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2676 have been a typo for LFLAGS, so it has been changed.
2678 13. The install script calls Exim with "-C /dev/null" in order to find the
2679 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2680 to be output. Howeve, since Exim outputs its version number before the
2681 error, it didn't break the script. It just looked ugly. I fixed this by
2682 always allowing "-C /dev/null" if the caller is root.
2684 14. Ignore overlarge ACL variable number when reading spool file - insurance
2685 against a later release with more variables having written the file.
2687 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2688 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2689 this, as well as the form without the "IPv6" on the front.
2691 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2692 OS/Makefile-Darwin file.
2694 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2695 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2697 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2698 variable. However, some OS use ints for inodes. I've added cast to long int
2699 to get rid of the compiler warning.
2701 19. I had forgotten to lock out "/../" in configuration file names when
2702 ALT_CONFIG_PREFIX was set.
2704 20. Routers used for verification do not need to specify transports. However,
2705 if such a router generated a host list, and callout was configured, Exim
2706 crashed, because it could not find a port number from the (non-existent)
2707 transport. It now assumes port 25 in this circumstance.
2709 21. Added the -t option to exigrep.
2711 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2712 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2713 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2714 them is set, LOOKUP_LSEARCH is forced.
2716 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2718 24. Added sender and host information to the "rejected by local_scan()" log
2719 line; previously there was no indication of these.
2721 25. Added .include_if_exists.
2723 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2724 Linux. It turns out that not all file systems support this. Apparently some
2725 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2726 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2727 not supported on the file descriptor, is now ignored when Exim is trying to
2728 sync a directory. This applies only to Linux.
2730 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2732 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2733 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2736 29. Added support for maildirsize files from supplied patch (modified a bit).
2738 30. The use of :fail: followed by an empty string could lead Exim to respond to
2739 sender verification failures with (e.g.):
2741 550 Verification failed for <xxx>
2742 550 Sender verify failed
2744 where the first response line was missing the '-' that indicates it is not
2745 the final line of the response.
2747 31. The loop for finding the name of the user that called Exim had a hardwired
2748 limit of 10; it now uses the value of finduser_retries, which is used for
2749 all other user lookups.
2751 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2754 33. Exim was neglecting to zero errno before one call of strtol() when
2755 expanding a string and expecting an integer value. On some systems this
2756 resulted in spurious "integer overflow" errors. Also, it was casting the
2757 result into an int without checking.
2759 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2760 did not work. The code looks as if it has *never* worked, though it appears
2761 to have been documented since at least releast 1.62. I have made it work.
2763 35. The "timeout_DNS" error in retry rules, also documented since at least
2764 1.62, also never worked. As it isn't clear exactly what this means, and
2765 clearly it isn't a major issue, I have abolished the feature by treating it
2766 as "timeout", and writing a warning to the main and panic logs.
2768 36. The display of retry rules for -brt wasn't always showing the error code
2771 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2772 timeout_connect_A, timeout_connect_MX.
2774 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2775 to the empty sender.
2777 39. The daemon was not analysing the content of -oX till after it had closed
2778 stderr and disconnected from the controlling terminal. This meant that any
2779 syntax errors were only noted on the panic log, and the return code from
2780 the command was 0. By re-arranging the code a little, I've made the
2781 decoding happen first, so such errors now appear on stderr, and the return
2782 code is 1. However, the actual setting up of the sockets still happens in
2783 the disconnected process, so errors there are still only recorded on the
2786 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2787 connections (as happens on some IP stacks) was logged at start up time as
2788 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2789 it from "IPv6 and IPv4", which means that two separate sockets are being
2792 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2793 says whether AF_INET or AF_INET6 was passed as an argument.
2795 42. Exiwhat output was messed up when time zones were included in log
2798 43. Exiwhat now gives more information about the daemon's listening ports,
2799 and whether -tls-on-connect was used.
2801 44. The "port" option of the smtp transport is now expanded.
2803 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2804 silently ignored. Now an error message is written to the main and panic
2807 46. There's a new ACL modifier called "logwrite" which writes to a log file
2808 as soon as it is encountered.
2810 47. Added $local_user_uid and $local_user_gid at routing time.
2812 48. Exim crashed when trying to verify a sender address that was being
2815 49. Exim was recognizing only a space character after ".include". It now also
2816 recognizes a tab character.
2818 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2819 extracting the relevant information from the specification. The man page no
2820 longer contains scrambled data for the -d option, and I've added a section
2821 at the front about calling Exim under different names.
2823 51. Added "extra_headers" argument to the "mail" command in filter files.
2825 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2828 53. Installed eximstats 1.29.
2830 54. Added transport_filter_timeout as a generic transport option.
2832 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2833 Cc: header lines. This was required by RFC 822, but it not required by RFC
2836 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2837 incoming messages that did not have them. Now it does so only if the
2838 message originates locally, that is, if there is no associated remote host
2839 address. When Resent- header lines are present, this applies to the Resent-
2840 lines rather than the non-Resent- lines.
2842 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2843 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2845 58. Messages for configuration errors now include the name of the main
2846 configuration file - useful now that there may be more than one file in a
2847 list (.included file names were always shown).
2849 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2850 for those rare installations that do not start the daemon as root or run it
2851 setuid root. I've cut out the call to initgroups() if the daemon is not
2854 60. The Exim user and group can now be bound into the binary as text strings
2855 that are looked up at the start of Exim's processing.
2857 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2859 62. Added $mailstore_basename variable.
2861 63. Installed patch to sieve.c from Michael Haardt.
2863 64. When Exim failed to open the panic log after failing to open the main log,
2864 the original message it was trying to log was written to stderr and debug
2865 output, but if they were not available (the usual case in production), it
2866 was lost. Now it is written to syslog before the two lines that record the
2867 failures to open the logs.
2869 65. Users' Exim filters run in subprocesses under the user's uid. It is
2870 possible for a "deliver" command or an alias in a "personal" command to
2871 provoke an address rewrite. If logging of address rewriting is configured,
2872 this fails because the process is not running as root or exim. There may be
2873 a better way of dealing with this, but for the moment (because 4.30 needs
2874 to be released), I have disabled address rewrite logging when running a
2875 filter in a non-root, non-exim process.
2881 1. The buildconfig auxiliary program wasn't quoting the value set for
2882 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2883 not defined. This bug was masked in 4.22 by the effect that was fixed in
2886 2. Some messages that were rejected after a message id was allocated were
2887 shown as "incomplete" by exigrep. It no longer does this for messages that
2888 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2890 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2891 have allow_domain_literals set, the ID did not get logged in the <= line.
2892 Domain literals are now always recognized in Message-ID: header lines.
2894 4. The first argument for a ${extract expansion item is the key name or field
2895 number. Leading and trailing spaces in this item were not being ignored,
2896 causing some misleading effects.
2898 5. When deliver_drop_privilege was set, single queue runner processes started
2899 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2900 same command in the process it spins off) were not dropping privilege.
2902 6. When the daemon running as "exim" started a queue runner, it always
2903 re-executed Exim in the spun-off process. This is a waste of effort when
2904 deliver_drop_privilege is set. The new process now just calls the
2905 queue-runner function directly.
2911 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2914 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2915 ignored. Though the use of -oP was forcing the writing of a pid file, it
2916 was always written to the default place.
2918 3. If the message "no IP address found for host xxxx" is generated during
2919 incoming verification, it is now followed by identification of the incoming
2920 connection (so you can more easily find what provoked it).
2922 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2924 5. Added some features to "harden" Exim a bit more against certain attacks:
2926 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2927 be put in Local/Makefile. This is like the never_users runtime option,
2928 but it cannot be overridden. The default setting is "root".
2930 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2931 prefix string with which any file named in a -C command line option
2934 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2935 is retained for -C and -D only if the caller of Exim is root. Without
2936 it, the exim user may also use -C and -D and retain privilege.
2938 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2939 command line option is disabled.
2941 6. Macro names set by the -D option must start with an upper case letter, just
2942 like macro names defined in the configuration file.
2944 7. Added "dereference=" facility to LDAP.
2946 8. Two instances of the typo "uknown" in the source files are fixed.
2948 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2949 the Configure-Makefile script screwed up while processing it.
2951 10. Incorporated PCRE 4.4.
2953 11. The SMTP synchronization check was not operating right at the start of an
2954 SMTP session. For example, it could not catch a HELO sent before the client
2955 waited for the greeting. There is now a check for outstanding input at the
2956 point when the greeting is written. Because of the duplex, asynchronous
2957 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2958 way, but not yet received, when the check is performed.
2960 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2961 on TCP/IP sockets, because this apparently causes some broken clients to
2964 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2965 unchanged) from the Cygwin maintainer.
2967 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2968 was supported instead of testing for mbx. Effectively a typo.
2970 15. The spa authenticator server code was not checking that the input it
2971 received was valid base64.
2973 16. The debug output line for the "set" modifier in ACLs was not showing the
2974 name of the variable that was being set.
2976 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2978 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2979 lookup on its own. Something else was needed to trigger the lookup. For
2980 example, a match in host_lookup or the need for a host name in a host list.
2981 Now, if $sender_host_name is referenced and the host name has not yet been
2982 looked up, a lookup is performed. If the lookup fails, the variable remains
2983 empty, and $host_lookup_failed is set to "1".
2985 19. Added "eqi" as a case-independent comparison operator.
2987 20. The saslauthd authentication condition could segfault if neither service
2988 nor realm was specified.
2990 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2991 error message that was logged was misleading, and incoming SMTP
2992 connections were dropped. The message is now more accurate, and temporary
2993 errors are given to SMTP connections.
2995 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2996 (except RSET) with a 421 error, until QUIT is received. However, it was
2997 failing to send a response to QUIT.
2999 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3000 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3001 ACL is now run after the helo_try_verify_hosts code.
3003 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3004 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3005 case-independent, so other case variants are also recognized). Apparently
3006 some systems use these upper case variants.
3008 25. If more than two messages were waiting for the same host, and a transport
3009 filter was specified for the transport, Exim sent two messages over the
3010 same TCP/IP connection, and then failed with "socket operation on non-
3011 socket" when it tried to send the third.
3013 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3015 27. The extern definition of crypt16() in expand.c was not being excluded when
3016 the OS had its own crypt16() function.
3018 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3019 as a preferred synonym for return_size_limit, both as an option and as an
3022 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3024 30. Changed the default configuration ACL to relax the local part checking rule
3025 for addresses that are not in any local domains. For these addresses,
3026 slashes and pipe symbols are allowed within local parts, but the sequence
3027 /../ is explicitly forbidden.
3029 31. SPA server authentication was not clearing the challenge buffer before
3032 32. log_message in a "warn" ACL statement was writing to the reject log as
3033 well as to the main log, which contradicts the documentation and doesn't
3034 seem right (because no rejection is happening). So I have stopped it.
3036 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3037 However, I am unable to do any testing of this.
3039 34. Fixed an infelicity in the appendfile transport. When checking directories
3040 for a mailbox, to see if any needed to be created, it was accidentally
3041 using path names with one or more superfluous leading slashes; tracing
3042 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3044 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3045 contents are added to the log line that is written for every discarded
3046 recipient. (Previously a log_message setting was ignored.)
3048 36. The ${quote: operator now quotes the string if it is empty.
3050 37. The install script runs exim in order to find its version number. If for
3051 some reason other than non-existence or emptiness, which it checks, it
3052 could not run './exim', it was installing it with an empty version number,
3053 i.e. as "exim-". This error state is now caught, and the installation is
3056 38. An argument was missing from the function that creates an error message
3057 when Exim fails to connect to the socket for saslauthd authentication.
3058 This could cause Exim to crash, or give a corrupted message.
3060 39. Added isip, isip4, and isip6 to ${if conditions.
3062 40. The ACL variables $acl_xx are now saved with the message, and can be
3063 accessed later in routers, transports, and filters.
3065 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3066 strings in the file are not string-expanded.
3068 42. If a MAIL command specified a SIZE value that was too large to fit into an
3069 int variable, the check against message_size_limit failed. Such values are
3070 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3071 day this will have to be increased, but I don't think I want to be around
3072 when emails are that large.
3079 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3080 iconv() is not standard in FreeBSD.
3082 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3083 IPv6 enabled. The observed symptom was a segmentation fault on return from
3084 the function os_common_find_running_interfaces() in src/os.c.
3086 3. In the check_special_case() function in daemon.c I had used "errno" as an
3087 argument name, which causes warnings on some systems. This was basically a
3088 typo, since it was named "eno" in the comments!
3090 4. The code that waits for the clock to tick (at a resolution of some fraction
3091 of a second) so as to ensure message-id uniqueness was always waiting for
3092 at least one whole tick, when it could have waited for less. [This is
3093 almost certainly not relevant at current processor speeds, where it is
3094 unlikely to ever wait at all. But we try to future-proof.]
3096 5. The function that sleeps for a time interval that includes fractions of a
3097 second contained a race. It did not block SIGALRM between setting the
3098 timer, and suspending (a couple of lines later). If the interval was short
3099 and the sigsuspend() was delayed until after it had expired, the suspension
3100 never ended. On busy systems this could lead to processes getting stuck for
3103 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3104 process, before it forks any delivery processes. The open lookup caching
3105 mechanism meant that the open file or database connection was passed into
3106 the delivery process. The problem was that delivery processes always tidy
3107 up cached lookup data. This could cause a problem for the next delivery
3108 process started by the queue runner, because the external queue runner
3109 process does not know about the closure. So the next delivery process
3110 still has data in the lookup cache. In the case of a file lookup, there was
3111 no problem because closing a file descriptor in a subprocess doesn't affect
3112 the parent. However, if the lookup was caching a connection to a database,
3113 the connection was closed, and the second delivery process was likely to
3114 see errors such as "PGSQL: query failed: server closed the connection
3115 unexpectedly". The problem has been fixed by closing all cached lookups
3116 in a queue runner before running a delivery process.
3118 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3119 seem to have the "const" qualifier which it has on other OS. I've
3122 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3123 *of the same type* (client or server) with the same public name that an
3124 error should be diagnosed.
3126 9. When Exim looked up a host name for an IP address, but failed to find the
3127 original IP address when looking up the host name (a safety check), it
3128 output the message "<ip address> does not match any IP for NULL", which was
3129 confusing, to say the least. The bug was that the host name should have
3130 appeared instead of "NULL".
3132 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3133 user that is built into the binary, and the -C or -D options is used, root
3134 privilege is dropped before the configuration file is read. In addition,
3135 logging is switched to stderr instead of the normal log files. If the
3136 configuration then re-defines the Exim user, the unprivileged environment
3137 is probably not what is expected, so Exim logs a panic warning message (but
3140 However, if deliver_drop_privilege is set, the unprivileged state may well
3141 be exactly what is intended, so the warning has been cut out in that case,
3142 and Exim is allowed to try to write to its normal log files.
3148 1. smtp_return_error_details was not giving details for temporary sender
3149 or receiver verification errors.
3151 2. Diagnose a configuration error if two authenticators have the same public
3154 3. Exim used not to create the message log file for a message until the first
3155 delivery attempt. This could be confusing when incoming messages were held
3156 for policy or load reasons. The message log file is now created at the time
3157 the message is received, and an initial "Received" line is written to it.
3159 4. The automatically generated man page for command line options had a minor
3160 bug that caused no ill effects; however, a more serious problem was that
3161 the procedure for building the man page automatically didn't always
3162 operate. Consequently, release 4.20 contains an out-of-date version. This
3163 shouldn't happen again.
3165 5. When building Exim with embedded Perl support, the script that builds the
3166 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3167 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3169 6. The freeze_tell option was not being used for messages that were frozen on
3170 arrival, either by an ACL or by local_scan().
3172 7. Added the smtp_incomplete_transaction log selector.
3174 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3175 was accepting AUTH without a new EHLO.
3177 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3178 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3181 10. Logging of TCP/IP connections (when configured) now happens in the main
3182 daemon process instead of the child process, so that the TCP/IP connection
3183 count is more accurate (but it can never be perfect).
3185 11. The use of "drop" in a nested ACL was not being handled correctly in the
3186 outer ACL. Now, if condition failure induced by the nested "drop" causes
3187 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3188 or "require"), the connection is dropped.
3190 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3191 that yield "discard" can now be used with an "accept" or a "discard" verb,
3192 but an error is generated for any others (because I can't see a useful way
3193 to define what should happen).
3195 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3196 are now processed in the same way as lines in the Exim configuration file.
3197 In particular, continuation lines are supported.
3199 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3201 15. Added -ti meaning -t -i.
3203 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3204 domains, and warn by logging if others are found.
3206 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3207 returned from GIFCONF when Exim is trying to find the list of interfaces on
3208 a host. The code in os.c has been modified to copy each ifreq to an aligned
3209 structure in all cases.
3211 Also, in some cases, the returned ifreq's were being copied to a 'struct
3212 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3213 means the last couple of bytes of an IPv6 address could be chopped if the
3214 ifreq contained only a normal sockaddr (14 bytes storage).
3216 18. Named domain lists were not supported in the hosts_treat_as_local option.
3217 An entry such as +xxxx was not recognized, and was treated as a literal
3220 19. Ensure that header lines added by a DATA ACL are included in the reject log
3221 if the ACL subsequently rejects the message.
3223 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3224 MD5 (which is deprecated).
3226 21. When testing a filter file using -bf, Exim was writing a message when it
3227 took the sender from a "From " line in the message, but it was not doing so
3228 when it took $return_path from a Return-Path: header line. It now does.
3230 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3231 with a valid header line field name (a series of printing characters
3232 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3234 23. Changed "disc" in the source to "disk" to conform to the documentation and
3235 the book and for uniformity.
3237 24. Ignore Sendmail's -Ooption=value command line item.
3239 25. When execve() failed while trying to run a command in a pipe transport,
3240 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3241 could be confused with a return value of 69 from the command itself. This
3242 has been changed to 127, the value the shell returns if it is asked to run
3243 a non-existent command. The wording for the related log line suggests a
3244 non-existent command as the problem.
3246 26. If received_header_text expands to an empty string, do not add a Received:
3247 header line to the message. (Well, it adds a token one on the spool, but
3248 marks it "old" so that it doesn't get used or transmitted.)
3250 27. Installed eximstats 1.28 (addition of -nt option).
3252 28. There was no check for failure on the call to getsockname() in the daemon
3253 code. This can fail if there is a shortage of resources on the system, with
3254 ENOMEM, for example. A temporary error is now given on failure.
3256 29. Contrary to the C standard, it seems that in some environments, the
3257 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3258 program. Exim now does this explicitly; it affects the formatting of
3259 timestamps using strftime().
3261 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3262 complaints. I've now initialized all the variables, to avoid this.
3264 32. Header lines added by a system filter were not being "seen" during
3265 transport-time rewrites.
3267 33. The info_callback() function passed to OpenSSL is set up with type void
3268 (*)(SSL *, int, int), as described somewhere. However, when calling the
3269 function (actually a macro) that sets it up, the type void(*)() is
3270 expected. I've put in a cast to prevent warnings from picky compilers.
3272 34. If a DNS black list lookup found a CNAME record, but there were no A
3273 records associated with the domain it pointed at, Exim crashed.
3275 35. If a DNS black list lookup returned more than one A record, Exim ignored
3276 all but the first. It now scans all returned addresses if a particular IP
3277 value is being sought. In this situation, the contents of the
3278 $dnslist_value variable are a list of all the addresses, separated by a
3281 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3282 to accept a host name and all its aliases if the forward lookup for any of
3283 them yielded the IP address of the incoming connection. Now it accepts only
3284 those names whose forward lookup yields the correct IP address. Any other
3285 names are discarded. This closes a loophole whereby a rogue DNS
3286 administrator could create reverse DNS records to break through a
3287 wildcarded host restriction in an ACL.
3289 37. If a user filter or a system filter that ran in a subprocess used any of
3290 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3291 the wrong values were passed to the pipe command ($thisaddress had the
3292 value of $0, $0 had the value of $1, etc). This bug was introduced by
3293 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3295 38. Improved the line breaking for long SMTP error messages from ACLs.
3296 Previously, if there was no break point between 40 and 75 characters, Exim
3297 left the rest of the message alone. Two changes have been made: (a) I've
3298 reduced the minimum length to 35 characters; (b) if it can't find a break
3299 point between 35 and 75 characters, it looks ahead and uses the first one
3300 that it finds. This may give the occasional overlong line, but at least the
3301 remaining text gets split now.
3303 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3304 file descriptors might be low, and that setting 1000 would always raise it.
3305 It turns out that in some environments, the limit is already over 1000 and
3306 that lowering it causes trouble. So now Exim takes care not to decrease it.
3308 40. When delivering a message, the value of $return_path is set to $sender_
3309 address at the start of routing (routers may change the value). By an
3310 oversight, this default was not being set up when an address was tested by
3311 -bt or -bv, which affected the outcome if any router or filter referred to
3314 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3315 log only when "message" or "log_message" are set. However, if one of the
3316 conditions was an address verification, or a call to a nested ACL, the
3317 messages generated by the underlying test were being passed through. This
3318 no longer happens. The underlying message is available in $acl_verify_
3319 message for both "message" and "log_message" expansions, so it can be
3320 passed through if needed.
3322 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3323 new expansion $bh_ to give the encoded byte string without charset
3324 translation. Translation happens only if iconv() is available; HAVE_ICONV
3325 indicates this at build time. HEADERS_CHARSET gives the charset to
3326 translate to; headers_charset can change it in the configuration, and
3327 "headers charset" can change it in an individual filter file.
3329 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3330 that creates RFC 2047 encoded "words" labels them as that charset instead
3331 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3332 expansion operator; (ii) when Exim creates a From: line for a local
3333 message; (iii) when a header line is rewritten to include a "phrase" part.
3335 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3336 buggy, causing it to skip the first lines of messages whose message ID
3337 ended in 'D'. This would not have bitten before Exim release 4.14, because
3338 message IDs were unlikely to end in 'D' before then. The effect was to have
3339 incorrect size information for certain domains.
3341 45. #include "config.h" was missing at the start of the crypt16.c module. This
3342 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3345 46. If there was a timeout during a "random" callout check, Exim treated it as
3346 a failure of the random address, and carried on sending RSET and the real
3347 address. If the delay was just some slowness somewhere, the response to the
3348 original RCPT would be taken as a response to RSET and so on, causing
3349 mayhem of various kinds.
3351 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3352 when I implemented it. It didn't allow for the fact that some option values
3353 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3354 the right test for positive values.
3356 48. Domain names in DNS records are case-independent. Exim always looks them up
3357 in lower case. Some resolvers return domain names in exactly the case they
3358 appear in the zone file, that is, they may contain uppercase letters. Not
3359 all resolvers do this - some return always lower case. Exim was treating a
3360 change of case by a resolver as a change of domain, similar to a widening
3361 of a domain abbreviation. This triggered its re-routing code and so it was
3362 trying to route what was effectively the same domain again. This normally
3363 caused routing to fail (because the router wouldn't handle the domain
3364 twice). Now Exim checks for this case specially, and just changes the
3365 casing of the domain that it ultimately uses when it transmits the message
3368 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3371 50. If a filter generated a file delivery with a non-absolute name (possible if
3372 no home directory exists for the router), the forbid_file option was not
3375 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3376 the existing equality matching.
3378 52. Exim was using ints instead of ino_t variables in some places where it was
3379 dealing with inode numbers.
3381 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3382 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3383 called TMPDIR, and if it finds it is different, it changes its value.
3385 54. The smtp_printf() function is now made available to local_scan() so
3386 additional output lines can be written before returning. There is also an
3387 smtp_fflush() function to enable the detection of a dropped connection.
3388 The variables smtp_input and smtp_batched_input are exported to
3391 55. Changed the default runtime configuration: the message "Unknown user"
3392 has been removed from the ACL, and instead placed on the localuser router,
3393 using the cannot_route_message feature. This means that any verification
3394 failures that generate their own messages won't get overridden. Similarly,
3395 the "Unrouteable address" message that was in the ACL for unverifiable
3396 relay addresses has also been removed.
3398 56. Added hosts_avoid_esmtp to the smtp transport.
3400 57. The exicyclog script was not checking for the esoteric option
3401 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3402 will work only if exicyclog is run under the appropriate euid.
3404 58. Following a discussion on the list, the rules by which Exim recognises line
3405 endings on incoming messages have been changed. The -dropcr and drop_cr
3406 options are now no-ops, retained only for backwards compatibility. The
3407 following line terminators are recognized: LF CRLF CR. However, special
3408 processing applies to CR:
3410 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3411 nor a local message in the state where . is a terminator.
3413 (ii) If a bare CR is encountered in a header line, an extra space is added
3414 after the line terminator so as not to end the header. The reasoning
3415 behind this is that bare CRs in header lines are most likely either
3416 to be mistakes, or people trying to play silly games.
3418 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3419 was being incorrectly given as 18 bytes larger than it should have been.
3420 This is a VOB (very old bug).
3422 60. This may never have affected anything current, but just in case it has:
3423 When the local host is found other than at the start of a list of hosts,
3424 the local host, those with the same MX, and any that follow, are discarded.
3425 When the list in question was part of a longer list of hosts, the following
3426 hosts (not currently being processed) were also being discarded. This no
3427 longer happens. I'm not sure if this situation could ever has previously
3430 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3433 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3434 header. This is something that is recommended in a new Internet Draft, and
3435 is something that is documented as being done by Sendmail. There are two
3436 possible values. For messages generated by the autoreply transport, Exim
3439 Auto-Submitted: auto-replied
3441 whereas for all other generated messages (e.g. bounces) it adds
3443 Auto-Submitted: auto-generated
3445 63. The "personal" condition in filters now includes a test for the
3446 Auto-Submitted: header. If it contains the string "auto-" the message it
3447 not considered personal.
3449 64. Added rcpt_include_affixes as a generic transport option.
3451 65. Added queue_only_override (default true).
3453 66. Added the syslog_duplication option.
3455 67. If what should have been the first header line of a message consisted of
3456 a space followed by a colon, Exim was mis-interpreting it as a header line.
3457 It isn't of course - it is syntactically invalid and should therefore be
3458 treated as the start of the message body. The misbehaviour could have
3459 caused a number of strange effects, including loss of data in subsequent
3460 header lines, and spool format errors.
3462 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3463 client host had authenticated. This control can now be exercised by an ACL
3464 for more flexibility.
3466 69. By default, callouts do not happen when testing with -bh. There is now a
3467 variant, -bhc, which does actually run the callout code, including
3468 consulting and updating the callout cache.
3470 70. Added support for saslauthd authentication, courtesy of Alexander
3473 71. If statvfs() failed on the spool or log directories while checking their
3474 size for availability, Exim confusingly gave the error "space shortage".
3475 Furthermore, in debugging mode it crashed with a floating point exception.
3476 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3477 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3478 serious problem, Exim now writes to the main and panic logs when this
3479 happens, with details of the failure. It then refuses to accept the
3480 incoming message, giving the message "spool directory problem" or "log
3481 directory problem" with a 421 code for SMTP messages.
3483 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3484 0, 1, and 2 exist, because some OS complain for execs without them (see
3485 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3486 descriptors. However, the code omitted to check that the open succeeded,
3487 causing mysterious errors if for some reason the permissions on /dev/null
3488 got screwed. Now Exim writes a message to the main and panic logs, and
3489 bombs out if it can't open /dev/null.
3491 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3492 interact so that it is all more flexible. It is supposed to remain
3493 backwards compatible. Also added extra_local_interfaces.
3495 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3496 to bomb out with an assertion failure - to the client this appears as a
3497 connection drop. This problem occurs in the part of the code that was taken
3498 from the Samba project. Fortunately, the assertion is in a very simple
3499 function, so I have fixed this by reproducing the function inline in the
3500 one place where it is called, and arranging for authentication to fail
3501 instead of killing the process with assert().
3503 75. The SPA client code was not working when the server requested OEM rather
3504 than Unicode encoding.
3506 76. Added code to make require_files with a specific uid setting more usable in
3507 the case where statting the file as root fails - usually a non-root-mounted
3508 NFS file system. When this happens and the failure is EACCES, Exim now
3509 forks a subprocess and does the per-uid checking as the relevant uid.
3511 77. Added process_log_path.
3513 78. If log_file_path was not explicitly set, a setting of check_log_space or
3514 check_log_inodes was ignored.
3516 79. If a space check for the spool or log partitions fails, the incident is now
3517 logged. Of course, in the latter case the data may get lost...
3519 80. Added the %p formatting code to string_format() so that it can be used to
3520 print addresses in debug_print(). Adjusted all the address printing in the
3521 debugging in store.c to use %p rather than %d.
3523 81. There was a concern that a line of code in smtp_in.c could overflow a
3524 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3525 initially expressed, the concern was not well-founded, because trailing
3526 spaces are removed early. However, if the trailing spaces were followed by
3527 a NULL, they did not get removed, so the overflow was possible. Two fixes
3530 (a) I re-wrote the offending code in a cleaner fashion.
3531 (b) If an incoming SMTP command contains a NULL character, it is rejected
3534 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3535 runs initgroups(), so that if the Exim user is in any additional groups,
3536 they will be used during message reception.
3542 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob