1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.258 2005/11/14 13:56:49 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
126 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
127 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
130 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
131 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
133 PH/03 Typo: missing ".o" in src/pcre/Makefile.
135 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
136 header line, restrict the check to what is listed in RFCs 2369 and 2929.
137 Also, for "Auto-Submitted", treat anything other than "no" as
138 non-personal, in accordance with RFC 3834. (Previously it treated
139 anything starting "auto-" as non-personal.)
141 TF/01 The control=submission/name=... option had a problem with syntax
142 errors if the name included a slash character. The /name= option
143 now slurps the rest of the string, so it can include any characters
144 but it must come last in the list of options (after /sender_retain
147 PH/05 Some modifications to the interface to the fake nameserver for the new
154 TK/01 Added the "success_on_redirect" address verification option. See
155 NewStuff for rationale and an example.
157 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
159 PH/02 Patch to exigrep to allow it to work on syslog lines.
161 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
162 fread() to skip over the body file's header line, because in Cygwin the
163 header line is locked and is inaccessible.
165 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
166 co-exist for some time) to make it clear that it is the Exim ID that is
167 referenced, not the Message-ID: header line.
169 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
170 string_format() function, because snprintf() does not exist on all
173 PH/06 The use of forbid_filter_existstest now also locks out the use of the
174 ${stat: expansion item.
176 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
177 protocol synchronization error", to keep the pedants happy.
179 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
180 well as for IRIX systems, when gcc is being used. See the host.c source
183 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
185 PH/10 Named domain lists were not working if used in a queue_smtp_domains
188 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
189 transport and to the smtp transport in LMTP mode.
191 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
193 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
194 run a filter in a subprocess. This could lead to confusion in subsequent
195 lookups in the parent process. There should also be a search_tidyup() at
196 the end of the subprocess.
198 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
199 only if the host matched helo_try_verify_hosts, which caused the
200 verification to occur when the EHLO/HELO command was issued. The ACL just
201 tested the remembered result. Now, if a previous verification attempt has
202 not happened, "verify = helo" does it there and then.
204 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
206 TK/03 Fix log output including CR from clamd.
208 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
209 exist provoked a memory error which could cause a segfault.
211 PH/15 Installed PCRE 6.2
213 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
215 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
216 of the problem. Specifically, suggested +O2 rather than +O1 for the
219 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
221 PH/20 If a delivery was routed to a non-standard port by means of an SRV
222 record, the port was not correctly logged when the outgoing_port log
223 selector was set (it logged the transort's default port).
225 PH/21 Added support for host-specific ports to manualroute, queryprogram,
226 fallback_hosts, and "hosts" in the smtp transport.
228 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
229 host errors such as "Connection refused".
231 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
232 authentication with radiusclient 0.4.9:
234 - Error returned from rc_read_config was caught wrongly
235 - Username/password not passed on to radius server due to wrong length.
237 The presumption is that some radiusclient API changes for 4.51/PH/17
238 were not taken care of correctly. The code is still untested by me (my
239 Linux distribution still has 0.3.2 of radiusclient), but it was
240 contributed by a Radius user.
242 PH/24 When doing a callout, the value of $domain wasn't set correctly when
243 expanding the "port" option of the smtp transport.
245 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
246 while reading a MIME header. Thanks to Tom Hughes for a patch.
248 PH/24 Include config.h inside local_scan.h so that configuration settings are
251 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
252 that in an ACL for RCPT (for example), you can examine exactly what was
255 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
256 commands, but it was not correctly comparing the address with the actual
257 client host address. Thus, it would show the EHLO address in Received:
258 header lines when this was not necessary.
260 PH/27 Added the % operator to ${eval:}.
262 PH/28 Exim tries to create and chdir to its spool directory when it starts;
263 it should be ignoring failures (because with -C, for example, it has lost
264 privilege). It wasn't ignoring creation failures other than "already
267 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
270 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
271 because an input line was too long, either on its own, or by virtue of
272 too many continuations, the temporary file was not being removed, and the
273 return code was incorrect.
275 PH/31 Missing "BOOL" in function definition in filtertest.c.
277 PH/32 Applied Sieve patches from the maintainer.
279 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
281 PH/33 Added "verify = not_blind".
283 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
284 Local/Makefile (with some defaults set). These are used in built scripts
285 such as exicyclog, but they have never been used in the exim_install
286 script (though there are many overriding facilities there). I have
287 arranged that the exim_install script now takes note of these two
290 PH/35 Installed configuration files for Dragonfly.
292 PH/36 When a locally submitted message by a trusted user did not contain a
293 From: header, and the sender address was obtained from -f or from an SMTP
294 MAIL command, and the trusted user did not use -F to supply a sender
295 name, $originator_name was incorrectly used when constructing a From:
296 header. Furthermore, $originator_name was used for submission mode
297 messages from external hosts without From: headers in a similar way,
298 which is clearly wrong.
300 PH/37 Added control=suppress_local_fixups.
302 PH/38 When log_selector = +received_sender was set, and the addition of the
303 sender made the log line's construction buffer exactly full, or one byte
304 less than full, an overflow happened when the terminating "\n" was
307 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
308 when the result of a list match is failure because a DNS lookup failed.
310 PH/40 RM_COMMAND is now used in the building process.
312 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
313 the "build-* directories that it finds.
315 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
316 address in a domain literal was a prefix of an interface address.
318 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
319 when verifying a sender address, unless rewrite_headers is false.
321 PH/44 Wrote a long comment about why errors_to addresses are verified as
322 recipients, not senders.
324 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
325 the ratelimit ACL was added.
327 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
329 PH/46 Added extra information about PostgreSQL errors to the error string.
331 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
332 avoiding the need to install special zones in a real server. This is
333 backwards compatible; if it can't find the fake resolver, it drops back.
334 Thus, both old and new test suites can be run.
336 TF/02 Added util/ratelimit.pl
338 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
339 clock is set back in time.
341 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
342 Candler <B.Candler@pobox.com>.
344 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
345 OK for addresses that are the result of redirections.
347 PH/48 A number of further additions for the benefit of the new test suite,
348 including a fake gethostbyname() that interfaces to the fake DNS resolver
351 TF/06 The fix for widen_domains has also been applied to qualify_single and
352 search_parents which are the other dnslookup options that can cause
355 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
358 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
360 TF/07 Exim produced the error message "an SRV record indicated no SMTP
361 service" if it encountered an MX record with an empty target hostname.
362 The message is now "an MX or SRV record indicated no SMTP service".
364 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
365 if the DNS of the sending site is misconfigured. This is quite a
366 common situation. This change restores the behaviour of treating a
367 helo verification defer as a failure.
369 PH/51 If self=fail was set on a router, the bounce message did not include the
370 actual error message.
376 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
378 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
379 command itself ended in error, the underlying message about the transport
380 filter timeout was being overwritten with the pipe command error. Now the
381 underlying error message should be appended to the second error message.
383 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
385 PH/02 Reduce the amount of output that "make" produces by default. Full output
386 can still be requested.
388 PH/03 The warning log line about a condition test deferring for a "warn" verb
389 was being output only once per connection, rather than after each
390 occurrence (because it was using the same function as for successful
391 "warn" verbs). This seems wrong, so I have changed it.
393 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
394 it should not have, which might have caused a crash in the right
395 circumstances, but probably never did.
397 PH/04 Installed a modified version of Tony Finch's patch to make submission
398 mode fix the return path as well as the Sender: header line, and to
399 add a /name= option so that you can make the user's friendly name appear
402 TF/03 Added the control = fakedefer ACL modifier.
404 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
405 Mark Lowes for thorough testing.
407 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
409 TK/03 Merged latest SRS patch from Miles Wilton.
411 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
412 with the definition in sysexits.h (which is #included earlier).
413 Fortunately, Exim does not actually use EX_OK. The code used to try to
414 preserve the sysexits.h value, by assumimg that macro definitions were
415 scanned for macro replacements. I have been disabused of this notion,
416 so now the code just undefines EX_OK before #including unistd.h.
418 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
419 in the smtp transport. When a block could not be written in a single
420 write() function, the timeout was being re-applied to each part-write.
421 This seems wrong - if the receiver was accepting one byte at a time it
422 would take for ever. The timeout is now adjusted when this happens. It
423 doesn't have to be particularly precise.
425 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
426 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
428 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
429 without a domain if the check to <postmaster@domain> fails.
431 SC/01 Eximstats: added -xls and the ability to specify output files
432 (patch written by Frank Heydlauf).
434 SC/02 Eximstats: use FileHandles for outputing results.
436 SC/03 Eximstats: allow any combination of xls, txt, and html output.
438 SC/04 Eximstats: fixed display of large numbers with -nvr option
440 SC/05 Eximstats: fixed merging of reports with empty tables.
442 SC/06 Eximstats: added the -include_original_destination flag
444 SC/07 Eximstats: removed tabs and trailing whitespace.
446 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
448 TK/06 MBOX spool code: Add real "From " MBOX separator line
449 so the .eml file is really in mbox format (even though
450 most programs do not really care). Patch from Alex Miller.
452 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
453 The latter is generated from $received_to and is only set if the
454 message has one envelope recipient. SA can use these headers,
455 obviously out-of-the-box. Patch from Alex Miller.
457 PH/08 The ${def test on a variable was returning false if the variable's
458 value was "0", contrary to what the specification has always said!
459 The result should be true unless the variable is empty.
461 PH/09 The syntax error of a character other than { following "${if
462 def:variable_name" (after optional whitespace) was not being diagnosed.
463 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
464 accidental colon was present, for example, could give incorrect results.
466 PH/10 Tidied the code in a number of places where the st_size field of a stat()
467 result is used (not including appendfile, where other changes are about
470 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
471 This involved changing a lot of size variables from int to off_t. It
472 should work with maildirs and everything.
474 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
475 spamd dying while we are connected to it.
477 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
480 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
481 function for BDB 4.3.
483 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
485 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
486 the sources for PCRE have been re-arranged and more files are now
489 PH/15 The code I had for printing potentially long long variables in PH/11
490 above was not the best (it lost precision). The length of off_t variables
491 is now inspected at build time, and an appropriate printing format (%ld
492 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
493 to be "long long int" or "long int". This is needed for the internal
494 formatting function string_vformat().
496 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
497 the configuration file to be ":syslog", then the script "guesses" where
498 the logs files are, rather than using the compiled in default. In our
499 case the guess is not the same as the compiled default, so the script
500 suddenly stopped working when I started to use syslog. The patch checks
501 to see if log_file_path is "". If so, it attempts to read it from exim
502 with no configuration file to get the compiled in version, before it
503 falls back to the previous guessing code."
505 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
506 implementing BATV in an Exim configuration. See NewStuff for the gory
509 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
510 Makefile that are specific to HP-UX.
512 PH/18 If the "use_postmaster" option was set for a recipient callout together
513 with the "random" option, the postmaster address was used as the MAIL
514 FROM address for the random test, but not for the subsequent recipient
515 test. It is now used for both.
517 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
518 patch removes a few documentation additions to RFC 3028, because the
519 latest draft now contains them. It adds the new en;ascii-case comparator
520 and a new error check for 8bit text in MIME parts. Comparator and
521 require names are now matched exactly. I enabled the subaddress
522 extension, but it is not well tested yet (read: it works for me)."
524 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
525 rework some of the code of TK/09 above to avoid the hardwired use of
526 "%lld" and "long long". Replaced the call to snprintf() with a call to
529 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
530 records point to non-existent hosts", "retry timeout exceeded", and
531 "retry time not reached for any host after a long failure period".
533 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
534 experimental DomainKeys support:
536 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
537 (2) On an error such as an illegally used "control", the wrong name for
538 the control was given.
540 These problems did NOT occur unless DomainKeys support was compiled.
542 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
544 PH/24 Added ${if match_ip condition.
546 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
547 databases so that it will be absolutely obvious if a crash occurs in the
548 DB library. This is a regular occurrence (often caused by mis-matched
551 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
552 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
553 picked up on a user's system that detects such things. There doesn't seem
554 to be a gcc warning option for this - only an attribute that has to be
555 put on the function's prototype. It seems that in Fedora Core 4 they have
556 set this on a number of new functions. No doubt there will be more in due
559 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
560 specify a transport. However, if an address that was verified by such a
561 router was the subject of a callout, Exim crashed because it tried to
562 read the rcpt_include_affixes from the non-existent transport. Now it
563 just assumes that the setting of that option is false. This bug was
564 introduced by 4.51/PH/31.
566 PH/28 Changed -d+all to exclude +memory, because that information is very
567 rarely of interest, but it makes the output a lot bigger. People tend to
568 do -d+all out of habit.
570 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
571 code in os-type was giving problems when libc.so lives in lib64, like on
574 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
575 aren't the modern standard, and it seems that some systems' include files
576 don't always have them. Exim was already checking for some of the newer
577 ones like T_AAAA, and defining it itself. I've added checks for all the
578 record types that Exim uses.
580 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
581 not automatically generating a new one, as it is supposed to. This
582 prevented TLS from working. If the file did exist, but contained invalid
583 data, a new version was generated, as expected. It was only the case of a
584 non-existent file that was broken.
586 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
587 with a change in libdomainkeys > 0.64.
589 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
590 from DNS. If the selector record carries the flag, it now has
591 precedence over the domain-wide flag.
593 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
595 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
596 the use of a port name that isn't defined in /etc/services) occurred, the
597 message was deferred as in a normal delivery, and thus remained on the
598 spool, instead of being failed because of the mua_wrapper setting. This
599 is now fixed, and I tidied up some of the mua_wrapper messages at the
602 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
603 the messages in a hash of arrays rather than using individual hashes.
604 This is a bit cleaner and results in dramatic memory savings, albeit
605 at a slight CPU cost.
607 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
608 as requested by Marc Sherman.
610 SC/10 Eximstats: added histograms for user specified patterns as requested
613 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
615 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
616 fopen() in the content-scanning modules that did not already have it.
622 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
623 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
625 TK/02 Fix ACL "control" statement not being available in MIME ACL.
627 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
629 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
630 to test Sieve filters that use "vacation".
632 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
633 that changes the way the GnuTLS parameters are stored in the cache file.
634 The new format can be generated externally. For backward compatibility,
635 if the data in the cache doesn't make sense, Exim assumes it has read an
636 old-format file, and it generates new data and writes a new file. This
637 means that you can't go back to an older release without removing the
640 PH/03 A redirect router that has both "unseen" and "one_time" set does not
641 work if there are any delivery delays because "one_time" forces the
642 parent to be marked "delivered", so its unseen clone is never tried
643 again. For this reason, Exim now forbids the simultaneous setting of
646 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
647 redirected to themselves ("homonym" addresses). Read the long ChangeLog
648 entry if you want to know the details. The fix, however, neglected to
649 consider the case when local delivery batching is involved. The test for
650 "previously delivered" was not happening when checking to see if an
651 address could be batched with a previous (undelivered) one; under
652 certain circumstances this could lead to multiple deliveries to the same
655 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
656 in its include files, and this causes problems building Exim.
658 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
659 header_syntax) but Exim was just ignoring anything given after a slash.
660 In particular, this caused confusion with an attempt to use "verify =
661 reverse_host_lookup/defer_ok". An error is now given when options are
662 supplied for verify items that do not have them. (Maybe reverse_host_
663 lookup should have a defer_ok option, but that's a different point.)
665 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
666 defined by RFC 821) to 2048, because there were problems with some AUTH
667 commands, and RFC 1869 says the size should be increased for extended
668 SMTP commands that take arguments.
670 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
673 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
674 "unknown" error; now it says that the functionality isn't in the binary.
676 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
677 an address' error message when a string expansion fails (syntax or
678 whatever). Otherwise the password may appear in the log. Following change
679 PH/42 below, there is no longer a chance of it appearing in a bounce
682 PH/11 Installed exipick version 20050225.0 from John Jetmore.
684 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
685 of its addresses was ever tried. (Bugzilla bug #2.)
687 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
688 the result incorrectly in the debug output. (It correctly added a newline
689 to what was transported.)
691 TF/01 Added $received_time.
693 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
694 commented out examples of how to interface to a virus scanner and to
695 SpamAssassin. Also added commented examples of av_scanner and
696 spamd_address settings.
698 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
699 and controls are allowed in which ACLs. There were a couple of minor
700 errors. Some of the entries in the conditions table (which is a table of
701 where they are NOT allowed) were getting very unwieldy; rewrote them as a
702 negation of where the condition IS allowed.
704 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
706 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
707 header file does not have a version number, so I've had to invent a new
708 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
709 API. The code is untested by me (my Linux distribution still has 0.3.2 of
710 radiusclient), but it was contributed by a Radius user.
712 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
713 files or directories, for OpenSSL.
715 PH/19 When an Exim process that is running as root has to create an Exim log
716 file, it does so in a subprocess that runs as exim:exim so as to get the
717 ownership right at creation (otherwise, other Exim processes might see
718 the file with the wrong ownership). There was no test for failure of this
719 fork() call, which would lead to the process getting stuck as it waited
720 for a non-existent subprocess. Forks do occasionally fail when resources
721 run out. I reviewed all the other calls to fork(); they all seem to check
724 PH/20 When checking for unexpected SMTP input at connect time (before writing
725 the banner), Exim was not dealing correctly with a non-positive return
726 from the read() function. If the client had disconnected by this time,
727 the result was a log entry for a synchronization error with an empty
728 string after "input=" when read() returned zero. If read() returned -1
729 (an event I could not check), uninitialized data bytes were printed.
730 There were reports of junk text (parts of files, etc) appearing after
733 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
735 PH/22 Added support for macro redefinition, and (re)definition in between
736 driver and ACL definitions.
738 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
739 forgetting to use the resulting value; it was using the unexpanded value.
741 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
742 hadn't been configured. The fix is from Juergen Kreileder, who
743 understands it better than I do:
745 "Here's what I see happening with three configured cyrus_sasl
746 authenticators configured (plain, login, cram-md5):
748 On startup auth_cyrus_sasl_init() gets called for each of these.
749 This means three calls to sasl_listmech() without a specified mech_list.
750 => SASL tests which mechs of all available mechs actually work
751 => three warnings about OTP not working
752 => the returned list contains: plain, login, cram-md5, digest-md5, ...
754 With the patch, sasl_listmech() also gets called three times. But now
755 SASL's mech_list option is set to the server_mech specified in the the
756 authenticator. Or in other words, the answer from sasl_listmech()
757 gets limited to just the mech you're testing for (which is different
759 => the return list contains just 'plain' or 'login', 'cram-md5' or
760 nothing depending on the value of ob->server_mech.
762 I've just tested the patch: Authentication still works fine,
763 unavailable mechs specified in the exim configuration are still
764 caught, and the auth.log warnings about OTP are gone."
766 PH/25 When debugging is enabled, the contents of the command line are added
767 to the debugging output, even when log_selector=+arguments is not
770 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
771 answer is "GNU", and only if the return is "GNU/something" is the answer
774 PH/27 $acl_verify_message is now set immediately after the failure of a
775 verification in an ACL, and so is available in subsequent modifiers. In
776 particular, the message can be preserved by coding like this:
778 warn !verify = sender
779 set acl_m0 = $acl_verify_message
781 Previously, $acl_verify_message was set only while expanding "message"
782 and "log_message" when a very denied access.
784 PH/28 Modified OS/os.c-Linux with
786 -#ifndef OS_LOAD_AVERAGE
787 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
789 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
790 nomenclature these days.)
792 PH/29 Installed patch from the Sieve maintainer that adds the options
793 sieve_useraddress and sieve_subaddress to the redirect router.
795 PH/30 In these circumstances:
796 . Two addresses routed to the same list of hosts;
797 . First host does not offer TLS;
798 . First host accepts first address;
799 . First host gives temporary error to second address;
800 . Second host offers TLS and a TLS session is established;
801 . Second host accepts second address.
802 Exim incorrectly logged both deliveries with the TLS parameters (cipher
803 and peerdn, if requested) that were in fact used only for the second
806 PH/31 When doing a callout as part of verifying an address, Exim was not paying
807 attention to any local part prefix or suffix that was matched by the
808 router that accepted the address. It now behaves in the same way as it
809 does for delivery: the affixes are removed from the local part unless
810 rcpt_include_affixes is set on the transport.
812 PH/32 Add the sender address, as F=<...>, to the log line when logging a
813 timeout during the DATA phase of an incoming message.
815 PH/33 Sieve envelope tests were broken for match types other than :is. I have
816 applied a patch sanctioned by the Sieve maintainer.
818 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
819 the uid or gid is negative. A case of a negative gid caused this to be
820 noticed. The fix allows for either to be negative.
822 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
823 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
826 PH/36 The change PH/12 above was broken. Fixed it.
828 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
829 the grounds that routing the same address twice would always produce the
830 same answer. This might have been true once, but it is certainly no
831 longer true now. Routing a child address may depend on the previous
832 routing that produced that child. Some complicated redirection strategies
833 went wrong when messages had multiple recipients, and made Exim's
834 behaviour dependent on the order in which the addresses were given.
836 I have moved the duplicate checking until after the routing is complete.
837 Exim scans the addresses that are assigned to local and remote
838 transports, and removes any duplicates. This means that more work will be
839 done, as duplicates will always all be routed, but duplicates are
840 presumably rare, so I don't expect this is of any significance.
842 For deliveries to pipes, files, and autoreplies, the duplicate checking
843 still happens during the routing process, since they are not going to be
846 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
847 It corrects a timeout issue with spamd. This is Ian's comment: "The
848 background is that sometimes spamd either never reads data from a
849 connection it has accepted, or it never writes response data. The exiscan
850 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
851 blindly assumes that writes won't block so it may never time out."
853 PH/39 Allow G after quota size as well as K and M.
855 PH/40 The value set for $authenticated_id in an authenticator may not contain
856 binary zeroes or newlines because the value is written to log lines and
857 to spool files. There was no check on this. Now the value is run through
858 the string_printing() function so that such characters are converted to
859 printable escape sequences.
861 PH/41 $message_linecount is a new variable that contains the total number of
862 lines in the message. Compare $body_linecount, which is the count for the
865 PH/42 Exim no longer gives details of delivery errors for specific addresses in
866 bounce and delay warning messages, except in certain special cases, which
869 (a) An SMTP error message from a remote host;
870 (b) A message specified in a :fail: redirection;
871 (c) A message specified in a "fail" command in a system filter;
872 (d) A message specified in a FAIL return from the queryprogram router;
873 (e) A message specified by the cannot_route_message router option.
875 In these cases only, Exim does include the error details in bounce and
876 warning messages. There are also a few cases where bland messages such
877 as "unrouteable address" or "local delivery error" are given.
879 PH/43 $value is now also set for the "else" part of a ${run expansion.
881 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
882 being worked on, but at least Exim now implements the latest version to
885 PH/45 In a pipe transport, although a timeout while waiting for the pipe
886 process to complete was treated as a delivery failure, a timeout while
887 writing the message to the pipe was logged, but erroneously treated as a
888 successful delivery. Such timeouts include transport filter timeouts. For
889 consistency with the overall process timeout, these timeouts are now
890 treated as errors, giving rise to delivery failures by default. However,
891 there is now a new Boolean option for the pipe transport called
892 timeout_defer, which, if set TRUE, converts the failures into defers for
893 both kinds of timeout. A transport filter timeout is now identified in
896 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
897 systems where "make" and "gmake" are different, calling "gmake" at top
898 level broke things. I've arranged for the value of $(MAKE) to be passed
899 from the Makefile to this script so that it can call the same version of
903 A note about Exim versions 4.44 and 4.50
904 ----------------------------------------
906 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
907 changes of various kinds. As a consequence, a big documentation update was
908 needed. This delayed the release for rather longer than seemed good, especially
909 in the light of a couple of (minor) security issues. Therefore, the changes
910 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
911 release. So 4.44 and 4.50 are in effect two different branches that both start
914 I have left the 4.50 change log unchanged; it contains all the changes since
915 4.43. The change log for 4.44 is below; many of its items are identical to
916 those for 4.50. This seems to be the most sensible way to preserve the
917 historical information.
923 1. Minor wording change to the doc/README.SIEVE file.
925 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
926 computation of the current number of files was incorrect.
928 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
929 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
932 4. Give more explanation in the error message when the command for a transport
933 filter fails to execute.
935 5. There are several places where Exim runs a non-Exim command in a
936 subprocess. The SIGUSR1 signal should be disabled for these processes. This
937 was being done only for the command run by the queryprogram router. It is
938 now done for all such subprocesses. The other cases are: ${run, transport
939 filters, and the commands run by the lmtp and pipe transports.
941 6. Added CONFIGURE_GROUP build-time option.
943 7. Some older OS have a limit of 256 on the maximum number of file
944 descriptors. Exim was using setrlimit() to set 1000 as a large value
945 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
946 systems. I've change it so that if it can't get 1000, it tries for 256.
948 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
949 was an oversight, and furthermore, ever since the addition of extra
950 controls (e.g. 4.43/32), the checks on when to allow different forms of
951 "control" were broken. There should now be diagnostics for all cases when a
952 control that does not make sense is encountered.
954 9. Added the /retain_sender option to "control=submission".
956 10. $recipients is now available in the predata ACL (oversight).
958 11. Tidy the search cache before the fork to do a delivery from a message
959 received from the command line. Otherwise the child will trigger a lookup
960 failure and thereby defer the delivery if it tries to use (for example) a
961 cached ldap connection that the parent has called unbind on.
963 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
964 of $address_data from the recipient verification was clobbered by the
967 13. The value of address_data from a sender verification is now available in
968 $sender_address_data in subsequent conditions in the ACL statement.
970 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
972 15. Added a new option "connect=<time>" to callout options, to set a different
975 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
976 was its contents. (It was OK if the option was not defined at all.)
978 17. A "Completed" log line is now written for messages that are removed from
979 the spool by the -Mrm option.
981 18. New variables $sender_verify_failure and $recipient_verify_failure contain
982 information about exactly what failed.
984 19. Added -dd to debug only the daemon process.
986 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
987 handles timeouts, both on the server side and network timeouts. Renamed the
988 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
990 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
991 It is now set to "smtps".
993 22. $host_address is now set to the target address during the checking of
996 23. When checking ignore_target_hosts for an ipliteral router, no host name was
997 being passed; this would have caused $sender_host_name to have been used if
998 matching the list had actually called for a host name (not very likely,
999 since this list is usually IP addresses). A host name is now passed as
1002 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1003 code that specifies a non-restarting handler (typically sigaction() in
1004 modern systems) in an attempt to fix a rare and obscure crash bug.
1006 25. Narrowed the window for a race in the daemon that could cause it to ignore
1007 SIGCHLD signals. This is not a major problem, because they are used only to
1008 wake it up if nothing else does.
1010 26. A malformed maildirsize file could cause Exim to calculate negative values
1011 for the mailbox size or file count. Odd effects could occur as a result.
1012 The maildirsize information is now recalculated if the size or filecount
1015 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1016 support for a long time. Removed HAVE_SYS_VFS_H.
1018 28. Installed the latest version of exipick from John Jetmore.
1020 29. In an address list, if the pattern was not a regular expression, an empty
1021 subject address (from a bounce message) matched only if the pattern was an
1022 empty string. Non-empty patterns were not even tested. This was the wrong
1023 because it is perfectly reasonable to use an empty address as part of a
1024 database query. An empty address is now tested by patterns that are
1025 lookups. However, all the other forms of pattern expect the subject to
1026 contain a local part and a domain, and therefore, for them, an empty
1027 address still always fails if the pattern is not itself empty.
1029 30. Exim went into a mad DNS loop when attempting to do a callout where the
1030 host was specified on an smtp transport, and looking it up yielded more
1031 than one IP address.
1033 31. Re-factored the code for checking spool and log partition space into a
1034 function that finds that data and another that does the check. The former
1035 is then used to implement four new variables: $spool_space, $log_space,
1036 $spool_inodes, and $log_inodes.
1038 32. The RFC2047 encoding function was originally intended for short strings
1039 such as real names; it was not keeping to the 75-character limit for
1040 encoded words that the RFC imposes. It now respects the limit, and
1041 generates multiple encoded words if necessary. To be on the safe side, I
1042 have increased the buffer size for the ${rfc2047: expansion operator from
1045 33. It is now permitted to omit both strings after an "if" condition; if the
1046 condition is true, the result is "true". As before, when the second string
1047 is omitted, a false condition yields an empty string. This makes it less
1048 cumbersome to write custom ACL and router conditions.
1050 34. Failure to deliver a bounce message always caused it to be frozen, even if
1051 there was an errors_to setting on the router. The errors_to setting is now
1054 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1055 canonical form (fully expanded) before being placed in
1056 $sender_host_address.
1058 36. The table in the code that translates DNS record types into text (T_A to
1059 "A" for instance) was missing entries for NS and CNAME. It is just possible
1060 that this could have caused confusion if both these types were looked up
1061 for the same domain, because the text type is used as part of Exim's
1062 per-process caching. But the chance of anyone hitting this buglet seems
1065 37. The dnsdb lookup has been extended in a number of ways.
1067 (1) There is a new type, "zns", which walks up the domain tree until it
1068 finds some nameserver records. It should be used with care.
1070 (2) There is a new type, "mxh", which is like "mx" except that it returns
1071 just the host names, not the priorities.
1073 (3) It is now possible to give a list of domains (or IP addresses) to be
1074 looked up. The behaviour when one of the lookups defers can be
1075 controlled by a keyword.
1077 (4) It is now possible to specify the separator character for use when
1078 multiple records are returned.
1080 38. The dnslists ACL condition has been extended: it is now possible to supply
1081 a list of IP addresses and/or domains to be looked up in a particular DNS
1084 39. Added log_selector=+queue_time_overall.
1086 40. When running the queue in the test harness, wait just a tad after forking a
1087 delivery process, to get repeatability of debugging output.
1089 41. Include certificate and key file names in error message when GnuTLS fails
1090 to set them up, because the GnuTLS error message doesn't include the name
1091 of the failing file when there is a problem reading it.
1093 42. Allow both -bf and -bF in the same test run.
1095 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1097 44. The "Exiscan patch" is now merged into the mainline Exim source.
1099 45. Sometimes the final signoff response after QUIT could fail to get
1100 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1101 < 0 before doing a fflush(). This bug looks as though it goes back to the
1102 introduction of TLS in release 3.20, but "sometimes" must have been rare
1103 because the tests only now provoked it.
1105 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1106 (this can affect the format of dates).
1108 47. exim_tidydb, when checking for the continued existence of a message for
1109 which it has found a message-specific retry record, was not finding
1110 messages that were in split spool directories. Consequently, it was
1111 deleting retry records that should have stayed in existence.
1113 48. Steve fixed some bugs in eximstats.
1115 49. The SPA authentication driver was not abandoning authentication and moving
1116 on to the next authenticator when an expansion was forced to fail,
1117 contradicting the general specification for all authenticators. Instead it
1118 was generating a temporary error. It now behaves as specified.
1120 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1121 (the order specifies the preference for clients). The order is now AES256,
1122 AES128, 3DES, ARCFOUR128.
1124 51. Small patch to Sieve code - explicitly set From: when generating an
1127 52. Exim crashed if a remote delivery caused a very long error message to be
1128 recorded - for instance if somebody sent an entire SpamAssassin report back
1129 as a large number of 550 error lines. This bug was coincidentally fixed by
1130 increasing the size of one of Exim's internal buffers (big_buffer) that
1131 happened as part of the Exiscan merge. However, to be on the safe side, I
1132 have made the code more robust (and fixed the comments that describe what
1135 53. Now that there can be additional text after "Completed" in log lines (if
1136 the queue_time_overall log selector is set), a one-byte patch to exigrep
1137 was needed to allow it to recognize "Completed" as not the last thing in
1140 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1141 patch that reportedly fixes this has been added. I am not expert enough to
1142 create a test for it. This is what the patch creator wrote:
1144 "I found a little strange behaviour of ldap code when working with
1145 Windows 2003 AD Domain, where users was placed in more than one
1146 Organization Units. When I tried to give exim partial DN, the exit code
1147 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1148 But simultaneously result of request was absolutely normal ldap result,
1149 so I produce this patch..."
1151 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1152 REFERENCE, so I have modified the code to exclude the patch when that macro
1155 55. Some experimental protocols are using DNS PTR records for new purposes. The
1156 keys for these records are domain names, not reversed IP addresses. The
1157 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1158 leaves it alone. Component reversal etc. now happens only for IP addresses.
1161 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1163 57. Double the size of the debug message buffer (to 2048) so that more of very
1164 long debug lines gets shown.
1166 58. The exicyclog utility now does better if the number of log files to keep
1167 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1169 59. Two changes related to the smtp_active_hostname option:
1171 (1) $smtp_active_hostname is now available as a variable.
1172 (2) The default for smtp_banner uses $smtp_active_hostname instead
1173 of $primary_hostname.
1175 60. The host_aton() function is supposed to be passed a string that is known
1176 to be a valid IP address. However, in the case of IPv6 addresses, it was
1177 not checking this. This is a hostage to fortune. Exim now panics and dies
1178 if the condition is not met. A case was found where this could be provoked
1179 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1180 components; fortuitously, this particular loophole had already been fixed
1181 by change 4.50/55 above.
1183 If there are any other similar loopholes, the new check in host_aton()
1184 itself should stop them being exploited. The report I received stated that
1185 data on the command line could provoke the exploit when Exim was running as
1186 exim, but did not say which command line option was involved. All I could
1187 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1188 running as the user.
1191 61. There was a buffer overflow vulnerability in the SPA authentication code
1192 (which came originally from the Samba project). I have added a test to the
1193 spa_base64_to_bits() function which I hope fixes it.
1196 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1197 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1199 63. The daemon start-up calls getloadavg() while still root for those OS that
1200 need the first call to be done as root, but it missed one case: when
1201 deliver_queue_load_max is set with deliver_drop_privilege. This is
1202 necessary for the benefit of the queue runner, because there is no re-exec
1203 when deliver_drop_privilege is set.
1205 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1206 This has been fixed.
1208 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1209 was in use, was not putting the data itself into the right store pool;
1210 consequently, it could be overwritten for a subsequent message in the same
1211 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1214 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1216 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1217 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1218 the function were treating the return as a boolean value, which happened to
1219 work because 0=false and not-0=true, but is not correct code.
1221 68. The host_aton() function was not handling scoped IPv6 addresses (those
1222 with, for example, "%eth0" on the end) correctly.
1224 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1225 negated items (that is, ~something) in unsigned ints. Some compilers
1226 apparently mutter when there is no cast.
1228 70. If an address verification called from an ACL failed, and did not produce a
1229 user-specific message (i.e. there was only a "system" message), nothing was
1230 put in $acl_verify_message. In this situation, it now puts the system
1233 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1234 session; change 4.31/43 added the unwanted input to the log line - except
1235 that it did not do this in the start of session case. It now does.
1237 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1238 This is wrong and can cause the other end to generate a synchronization
1239 error if it is another Exim or anything else that does the synchronization
1240 check. A QUIT command is no longer sent after a timeout.
1242 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1243 during host lookups.
1245 74. The defer_ok option of callout verification was not working if it was used
1246 when verifying addresses in header lines, that is, for this case:
1248 verify = header_sender/callout=defer_ok
1250 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1251 those file descriptors could be used for SMTP connections. If anything
1252 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1253 could be sent to the SMTP client, causing chaos. The daemon now opens
1254 stdin, stdout, and stderr to /dev/null when it puts itself into the
1257 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1258 log by default. The user can override this with suitable Perl magic.
1260 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1261 the log message when discard triggers, was not working for the DATA ACL or
1262 for the non-SMTP ACL.
1264 78. Error message wording change in sieve.c.
1266 79. If smtp_accept_max_per_host was set, the number of connections could be
1267 restricted to fewer than expected, because the daemon was trying to set up
1268 a new connection before checking whether the processes handling previous
1269 connections had finished. The check for completed processes is now done
1270 earlier. On busy systems, this bug wouldn't be noticed because something
1271 else would have woken the daemon, and it would have reaped the completed
1274 80. If a message was submitted locally by a user whose login name contained one
1275 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1276 It caused a spool format error. I have fixed the spool reading code. A
1277 related problem was that the "from" clause in the Received: line became
1278 illegal because of the space(s). It is now covered by ${quote_local_part.
1280 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1283 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1285 83. Patch from Sieve maintainer to fix unterminated string problem in
1286 "vacation" handling.
1288 84. Some minor changes to the Linux configuration files to help with other
1289 OS variants using glibc.
1291 85. One more patch for Sieve to update vacation handling to latest spec.
1294 ----------------------------------------------------
1295 See the note above about the 4.44 and 4.50 releases.
1296 ----------------------------------------------------
1302 1. Change 4.43/35 introduced a bug that caused file counts to be
1303 incorrectly computed when quota_filecount was set in an appendfile
1306 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1307 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1310 3. Give more explanation in the error message when the command for a transport
1311 filter fails to execute.
1313 4. There are several places where Exim runs a non-Exim command in a
1314 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1315 was being done only for the command run by the queryprogram router. It is
1316 now done for all such subprocesses. The other cases are: ${run, transport
1317 filters, and the commands run by the lmtp and pipe transports.
1319 5. Some older OS have a limit of 256 on the maximum number of file
1320 descriptors. Exim was using setrlimit() to set 1000 as a large value
1321 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1322 systems. I've change it so that if it can't get 1000, it tries for 256.
1324 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1325 was an oversight, and furthermore, ever since the addition of extra
1326 controls (e.g. 4.43/32), the checks on when to allow different forms of
1327 "control" were broken. There should now be diagnostics for all cases when a
1328 control that does not make sense is encountered.
1330 7. $recipients is now available in the predata ACL (oversight).
1332 8. Tidy the search cache before the fork to do a delivery from a message
1333 received from the command line. Otherwise the child will trigger a lookup
1334 failure and thereby defer the delivery if it tries to use (for example) a
1335 cached ldap connection that the parent has called unbind on.
1337 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1338 of $address_data from the recipient verification was clobbered by the
1339 sender verification.
1341 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1342 was its contents. (It was OK if the option was not defined at all.)
1344 11. A "Completed" log line is now written for messages that are removed from
1345 the spool by the -Mrm option.
1347 12. $host_address is now set to the target address during the checking of
1348 ignore_target_hosts.
1350 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1351 being passed; this would have caused $sender_host_name to have been used if
1352 matching the list had actually called for a host name (not very likely,
1353 since this list is usually IP addresses). A host name is now passed as
1356 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1357 code that specifies a non-restarting handler (typically sigaction() in
1358 modern systems) in an attempt to fix a rare and obscure crash bug.
1360 15. Narrowed the window for a race in the daemon that could cause it to ignore
1361 SIGCHLD signals. This is not a major problem, because they are used only to
1362 wake it up if nothing else does.
1364 16. A malformed maildirsize file could cause Exim to calculate negative values
1365 for the mailbox size or file count. Odd effects could occur as a result.
1366 The maildirsize information is now recalculated if the size or filecount
1369 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1370 support for a long time. Removed HAVE_SYS_VFS_H.
1372 18. Updated exipick to current release from John Jetmore.
1374 19. Allow an empty sender to be matched against a lookup in an address list.
1375 Previously the only cases considered were a regular expression, or an
1378 20. Exim went into a mad DNS lookup loop when doing a callout where the
1379 host was specified on the transport, if the DNS lookup yielded more than
1382 21. The RFC2047 encoding function was originally intended for short strings
1383 such as real names; it was not keeping to the 75-character limit for
1384 encoded words that the RFC imposes. It now respects the limit, and
1385 generates multiple encoded words if necessary. To be on the safe side, I
1386 have increased the buffer size for the ${rfc2047: expansion operator from
1389 22. Failure to deliver a bounce message always caused it to be frozen, even if
1390 there was an errors_to setting on the router. The errors_to setting is now
1393 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1394 canonical form (fully expanded) before being placed in
1395 $sender_host_address.
1397 24. Updated eximstats to version 1.33
1399 25. Include certificate and key file names in error message when GnuTLS fails
1400 to set them up, because the GnuTLS error message doesn't include the name
1401 of the failing file when there is a problem reading it.
1403 26. Expand error message when OpenSSL has problems setting up cert/key files.
1406 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1407 (this can affect the format of dates).
1409 28. exim_tidydb, when checking for the continued existence of a message for
1410 which it has found a message-specific retry record, was not finding
1411 messages that were in split spool directories. Consequently, it was
1412 deleting retry records that should have stayed in existence.
1414 29. eximstats updated to version 1.35
1415 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1416 1.35 - bugfix such that pie charts by volume are generated correctly
1418 30. The SPA authentication driver was not abandoning authentication and moving
1419 on to the next authenticator when an expansion was forced to fail,
1420 contradicting the general specification for all authenticators. Instead it
1421 was generating a temporary error. It now behaves as specified.
1423 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1424 (the order specifies the preference for clients). The order is now AES256,
1425 AES128, 3DES, ARCFOUR128.
1427 31. Small patch to Sieve code - explicitly set From: when generating an
1430 32. Exim crashed if a remote delivery caused a very long error message to be
1431 recorded - for instance if somebody sent an entire SpamAssassin report back
1432 as a large number of 550 error lines. This bug was coincidentally fixed by
1433 increasing the size of one of Exim's internal buffers (big_buffer) that
1434 happened as part of the Exiscan merge. However, to be on the safe side, I
1435 have made the code more robust (and fixed the comments that describe what
1438 33. Some experimental protocols are using DNS PTR records for new purposes. The
1439 keys for these records are domain names, not reversed IP addresses. The
1440 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1441 leaves it alone. Component reversal etc. now happens only for IP addresses.
1444 34. The host_aton() function is supposed to be passed a string that is known
1445 to be a valid IP address. However, in the case of IPv6 addresses, it was
1446 not checking this. This is a hostage to fortune. Exim now panics and dies
1447 if the condition is not met. A case was found where this could be provoked
1448 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1449 components; fortuitously, this particular loophole had already been fixed
1450 by change 4.50/55 or 4.44/33 above.
1452 If there are any other similar loopholes, the new check in host_aton()
1453 itself should stop them being exploited. The report I received stated that
1454 data on the command line could provoke the exploit when Exim was running as
1455 exim, but did not say which command line option was involved. All I could
1456 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1457 running as the user.
1460 35. There was a buffer overflow vulnerability in the SPA authentication code
1461 (which came originally from the Samba project). I have added a test to the
1462 spa_base64_to_bits() function which I hope fixes it.
1465 36. The daemon start-up calls getloadavg() while still root for those OS that
1466 need the first call to be done as root, but it missed one case: when
1467 deliver_queue_load_max is set with deliver_drop_privilege. This is
1468 necessary for the benefit of the queue runner, because there is no re-exec
1469 when deliver_drop_privilege is set.
1471 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1472 was in use, was not putting the data itself into the right store pool;
1473 consequently, it could be overwritten for a subsequent message in the same
1474 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1477 38. Sometimes the final signoff response after QUIT could fail to get
1478 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1479 < 0 before doing a fflush(). This bug looks as though it goes back to the
1480 introduction of TLS in release 3.20, but "sometimes" must have been rare
1481 because the tests only now provoked it.
1487 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1488 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1489 Now it returns an integer. A number of calls were still expecting a T/F
1490 return. Fortuitously, in all cases, the tests worked in OK situations,
1491 which is the norm. However, things would have gone wrong on any write
1492 failures on the smtp file descriptor. This function is used when sending
1493 messages over SMTP and also when doing verify callouts.
1495 2. When Exim is called to do synchronous delivery of a locally submitted
1496 message (the -odf or -odi options), it no longer closes stderr before doing
1499 3. Implemented the mua_wrapper option.
1501 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1503 5. Implemented the functions header_remove(), header_testname(),
1504 header_add_at_position(), and receive_remove_recipient(), and exported them
1507 6. If an ACL "warn" statement specified the addition of headers, Exim already
1508 inserted X-ACL-Warn: at the start if there was no header name. However, it
1509 was not making this test for the second and subsequent header lines if
1510 there were newlines in the string. This meant that an invalid header could
1511 be inserted if Exim was badly configured.
1513 7. Allow an ACL "warn" statement to add header lines at the start or after all
1514 the Received: headers, as well as at the end.
1516 8. Added the rcpt_4xx retry error code.
1518 9. Added postmaster_mailfrom=xxx to callout verification option.
1520 10. Added mailfrom=xxxx to the callout verification option, for verify=
1523 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1524 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1526 12. Inserted some casts to stop certain compilers warning when using pointer
1527 differences as field lengths or precisions in printf-type calls (mostly
1528 affecting debugging statements).
1530 13. Added optional readline() support for -be (dynamically loaded).
1532 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1533 same clock tick as a message's arrival, so that its received time was the
1534 same as the "first fail" time on the retry record, and that message
1535 remained on the queue past the ultimate address timeout, every queue runner
1536 would try a delivery (because it was past the ultimate address timeout) but
1537 after another failure, the ultimate address timeout, which should have then
1538 bounced the address, did not kick in. This was a "< instead of <=" error;
1539 in most cases the first failure would have been in the next clock tick
1540 after the received time, and all would be well.
1542 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1543 being recognized when the domain list was tested by the match_domain
1544 condition in an expansion string.
1546 16. Added the ${str2b64: operator.
1548 17. Exim was always calling setrlimit() to set a large limit for the number of
1549 processes, without checking whether the existing limit was already
1550 adequate. (It did check for the limit on file descriptors.) Furthermore,
1551 errors from getrlimit() and setrlimit() were being ignored. Now they are
1552 logged to the main and panic logs, but Exim does carry on, to try to do its
1553 job under whatever limits there are.
1555 18. Imported PCRE 5.0.
1557 19. Trivial typo in log message " temporarily refused connection" (the leading
1560 20. If the log selector return_path_on_delivery was set and an address was
1561 redirected to /dev/null, the delivery process crashed because it assumed
1562 that a return path would always be set for a "successful" delivery. In this
1563 case, the whole delivery is bypassed as an optimization, and therefore no
1566 21. Internal re-arrangement: the function for sending a challenge and reading
1567 a response while authentication was assuming a zero-terminated challenge
1568 string. It's now changed to take a pointer and a length, to allow for
1569 binary data in such strings.
1571 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1573 23. Exim was not respecting finduser_retries when seeking the login of the
1574 uid under which it was called; it was always trying 10 times. (The default
1575 setting of finduser_retries is zero.) Also, it was sleeping after the final
1576 failure, which is pointless.
1578 24. Implemented tls_on_connect_ports.
1580 25. Implemented acl_smtp_predata.
1582 26. If the domain in control=submission is set empty, Exim assumes that the
1583 authenticated id is a complete email address when it generates From: or
1584 Sender: header lines.
1586 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1587 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1588 chown and chgrp in /bin and hostname in /usr/bin.
1590 28. Exim was keeping the "process log" file open after each use, just as it
1591 does for the main log. This opens the possibility of it remaining open for
1592 long periods when the USR1 signal hits a daemon. Occasional processlog
1593 errors were reported, that could have been caused by this. Anyway, it seems
1594 much more sensible not to leave this file open at all, so that is what now
1597 29. The long-running daemon process does not normally write to the log once it
1598 has entered its main loop, and it closes the log before doing so. This is
1599 so that log files can straightforwardly be renamed and moved. However,
1600 there are a couple of unusual error situations where the daemon does write
1601 log entries, and I had neglected to close the log afterwards.
1603 30. The text of an SMTP error response that was received during a remote
1604 delivery was being truncated at 512 bytes. This is too short for some of
1605 the long messages that one sometimes sees. I've increased the limit to
1608 31. It is now possible to make retry rules that apply only when a message has a
1609 specific sender, in particular, an empty sender.
1611 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1612 it possible to be selective about when SMTP synchronization is enforced.
1614 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1616 32. Implemented hosts_connection_nolog.
1618 33. Added an ACL for QUIT.
1620 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1623 35. Added mailbox_size and mailbox_filecount to appendfile.
1625 36. Added control = no_multiline_responses to ACLs.
1627 37. There was a bug in the logic of the code that waits for the clock to tick
1628 in the case where the clock went backwards by a substantial amount such
1629 that the microsecond fraction of "now" was more than the microsecond
1630 fraction of "then" (but the whole seconds number was less).
1632 38. Added support for the libradius Radius client library this is found on
1633 FreeBSD (previously only the radiusclient library was supported).
1639 1. When certain lookups returned multiple values in the form name=value, the
1640 quoting of the values was not always being done properly. Specifically:
1641 (a) If the value started with a double quote, but contained no whitespace,
1643 (b) If the value contained whitespace other than a space character (i.e.
1644 tabs or newlines or carriage returns) it was not quoted.
1645 This fix has been applied to the mysql and pgsql lookups by writing a
1646 separate quoting function and calling it from the lookup code. The fix
1647 should probably also be applied to nisplus, ibase and oracle lookups, but
1648 since I cannot test any of those, I have not disturbed their existing code.
1650 2. A hit in the callout cache for a specific address caused a log line with no
1651 reason for rejecting RCPT. Now it says "Previous (cached) callout
1652 verification failure".
1654 3. There was an off-by-one bug in the queryprogram router. An over-long
1655 return line was truncated at 256 instead of 255 characters, thereby
1656 overflowing its buffer with the terminating zero. As well as fixing this, I
1657 have increased the buffer size to 1024 (and made a note to document this).
1659 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1660 when Exim is waiting for an SMTP response from a remote server, Exim
1661 restarts its select() call on the socket, thereby resetting its timeout.
1662 This is not a problem when such interrupts are rare. Somebody set up a cron
1663 job to run exiwhat every 2 minutes, which is less than the normal select()
1664 timeout (5 or 10 minutes). This meant that the select() timeout never
1665 kicked in because it was always reset. I have fixed this by comparing the
1666 time when an interrupt arrives with the time at the start of the first call
1667 to select(). If more time than the timeout has elapsed, the interrupt is
1668 treated as a timeout.
1670 5. Some internal re-factoring in preparation for the addition of Sieve
1671 extensions (by MH). In particular, the "personal" test is moved to a
1672 separate function, and given an option for scanning Cc: and Bcc: (which is
1673 not set for Exim filters).
1675 6. When Exim created an email address using the login of the caller as the
1676 local part (e.g. when creating a From: or Sender: header line), it was not
1677 quoting the local part when it contained special characters such as @.
1679 7. Installed new OpenBSD configuration files.
1681 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1682 try to make them clearer.
1684 9. Callout options, other than the timeout value, were being ignored when
1685 verifying sender addresses in header lines. For example, when using
1687 verify = header_sender/callout=no_cache
1689 the cache was (incorrectly) being used.
1691 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1692 only the Cygwin environment.
1694 11. When return_path_on_delivery was set as a log selector, if different remote
1695 addresses in the same message used different return paths and parallel
1696 remote delivery occurred, the wrong values would sometimes be logged.
1697 (Whenever a remote delivery process finished, the return path value from
1698 the most recently started remote delivery process was logged.)
1700 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1701 lines when AUTH and/or TLS are in use. This is the "received protocol"
1702 field. Exim used to use "asmtp" for authenticated SMTP, without any
1703 indication (in the protocol name) for TLS use. Now it follows the RFC and
1704 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1705 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1706 names appear in log lines as well as in Received: header lines.
1708 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1709 extensions, and comparison tests, and to fix some bugs.
1711 14. Changes to the "personal" filter test:
1713 (1) The test was buggy in that it was just doing the equivalent of
1714 "contains" tests on header lines. For example, if a user's address was
1715 anne@some.where, the "personal" test would incorrectly be true for
1717 To: susanne@some.where
1719 This test is now done by extracting each address from the header in turn,
1720 and checking the entire address. Other tests that are part of "personal"
1721 are now done using regular expressions (for example, to check local parts
1722 of addresses in From: header lines).
1724 (2) The list of non-personal local parts in From: addresses has been
1725 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1726 taken from the Sieve specification recommendations.
1728 (3) If the message contains any header line starting with "List-" it is
1729 treated as non-personal.
1731 (4) The test for "circular" in the Subject: header line has been removed
1732 because it now seems ill-conceived.
1734 15. Minor typos in src/EDITME comments corrected.
1736 16. Installed latest exipick from John Jetmore.
1738 17. If headers_add on a router specified a text string that was too long for
1739 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1740 of string_sprintf() is now avoided.
1742 18. $message_body_size was not set (it was always zero) when running the DATA
1743 ACL and the local_scan() function.
1745 19. For the "mail" command in an Exim filter, no default was being set for
1746 the once_repeat time, causing a random time value to be used if "once" was
1747 specified. (If the value happened to be <= 0, no repeat happened.) The
1748 default is now 0s, meaning "never repeat". The "vacation" command was OK
1749 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1750 (I found it when inspecting the code).
1752 20. There is now an overall timeout for performing a callout verification. It
1753 defaults to 4 times the callout timeout, which applies to individual SMTP
1754 commands during the callout. The overall timeout applies when there is more
1755 than one host that can be tried. The timeout is checked before trying the
1756 next host. This prevents very long delays if there are a large number of
1757 hosts and all are timing out (e.g. when the network connections are timing
1758 out). The value of the overall timeout can be changed by specifying an
1759 additional sub-option for "callout", called "maxwait". For example:
1761 verify = sender/callout=5s,maxwait=20s
1763 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1764 to the end before writing, but this should make it even safer).
1766 22. Exim was forgetting that it had advertised PIPELINING for the second and
1767 subsequent messages on an SMTP connection. It was also not resetting its
1768 memory on STARTTLS and an internal HELO.
1770 23. When Exim logs an SMTP synchronization error within a session, it now
1771 records whether PIPELINING has been advertised or not.
1773 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1774 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1775 rather than long int.
1777 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1779 26. Added the never_mail option to autoreply.
1785 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1786 crash if the getsockname() call failed; this can happen if a connection is
1787 closed very soon after it is established. The problem was simply in the
1788 order in which certain operations were done, causing Exim to try to write
1789 to the SMTP stream before it had set up the file descriptor. The bug has
1790 been fixed by making things happen in the correct order.
1796 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1797 before the connection was closed, thus losing the rejection response.
1799 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1800 some early Solaris releases, but causes trouble in current releases where
1801 socklen_t is defined.
1803 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1806 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1809 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1810 that releasing the top of it at the end releases what was used for sub-
1811 expansions (unless the block got too big). However, discard this block if
1812 the first thing is a variable or header, so that we can use its block when
1813 it is dynamic (useful for very large $message_headers, for example).
1815 6. Lookups now cache *every* query, not just the most recent. A new, separate
1816 store pool is used for this. It can be recovered when all lookup caches are
1817 flushed. Lookups now release memory at the end of their result strings.
1818 This has involved some general refactoring of the lookup sources.
1820 7. Some code has been added to the store_xxx() functions to reduce the amount
1821 of flapping under certain conditions.
1823 8. log_incoming_interface used to affect only the <= reception log lines. Now
1824 it causes the local interface and port to be added to several more SMTP log
1825 lines, for example "SMTP connection from", and rejection lines.
1827 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1829 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1831 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1832 could be overwritten at the end of the current message (or the start of a
1833 new message if it was set in a HELO ACL). The value is now preserved for
1834 the duration of the SMTP connection.
1836 12. If a transport had a headers_rewrite setting, and a matching header line
1837 contained an unqualified address, that address was qualified, even if it
1838 did not match any rewriting rules. The underlying bug was that the values
1839 of the flags that permit the existence of unqualified sender and recipient
1840 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1841 non-local messages, and by -bnq for local messages) were not being
1842 preserved with the message after it was received.
1844 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1845 "next input=" as part of the text comprising the host identity instead of
1846 the correct text. The code was using the same buffer for two different
1847 strings. However, depending on which order the printing function evaluated
1848 its arguments, the bug did not always show up. Under Linux, for example, my
1849 test suite worked just fine.
1851 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1852 doesn't work with some older versions of Perl. It has been changed to "my",
1853 which in any case is probably the better facility to use.
1855 15. A really picky compiler found some instances of statements for creating
1856 error messages that either had too many or two few arguments for the format
1859 16. The size of the buffer for calls to the DNS resolver has been increased
1860 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1861 for addresses that have a lot of PTR records. This alleviates a problem; it
1862 does not fully solve it.
1864 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1865 buffer now truncates the list and logs the incident, which is the same
1866 action as happens when Exim is looking up a host name and its aliases.
1867 Previously in this situation something unpredictable would happen;
1868 sometimes it was "internal error: store_reset failed".
1870 18. If a server dropped the connection unexpectedly when an Exim client was
1871 using GnuTLS and trying to read a response, the client delivery process
1872 crashed while trying to generate an error log message.
1874 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1875 string, for example:
1877 warn message = H1: something\nH2: something
1879 the text was added as a single header line from Exim's point of view
1880 though it ended up OK in the delivered message. However, searching for the
1881 second and subsequent header lines using $h_h2: did not work. This has been
1882 fixed. Similarly, if a system filter added multiple headers in this way,
1883 the routers could not see them.
1885 20. Expanded the error message when iplsearch is called with an invalid key to
1886 suggest using net-iplsearch in a host list.
1888 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1889 ACLs are no longer actually imposed (and a message to that effect is
1892 22. If a "gecos" field in a passwd entry contained escaped characters, in
1893 particular, if it contained a \" sequence, Exim got it wrong when building
1894 a From: or a Sender: header from that name. A second bug also caused
1895 incorrect handling when an unquoted " was present following a character
1896 that needed quoting.
1898 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1899 was not being matched caselessly.
1901 24. Arranged for all hyphens in the exim.8 source to be escaped with
1904 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1905 properly. Recipient callout cache records were still being keyed to include
1906 the sender, even when use_sender was set false. This led to far more
1907 callouts that were necessary. The sender is no longer included in the key
1908 when use_sender is false.
1910 26. Added "control = submission" modifier to ACLs.
1912 27. Added the ${base62d: operator to decode base 62 numbers.
1914 28. dnsdb lookups can now access SRV records.
1916 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1917 the configuration file.
1919 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1920 (-v) mode. This makes the output for a verbose queue run more intelligible.
1922 31. Added a use_postmaster feature to recipient callouts.
1924 32. Added the $body_zerocount variable, containing the number of binary zero
1925 bytes in the message body.
1927 33. The time of last modification of the "new" subdirectory is now used as the
1928 "mailbox time last read" when there is a quota error for a maildir
1931 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1933 35. Added +ignore_unknown as a special item in host lists.
1935 36. Code for decoding IPv6 addresses in host lists is now included, even if
1936 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1937 address was recognized as an IP address, but was then not correctly decoded
1938 into binary, causing unexpected and incorrect effects when compared with
1945 1. Very minor rewording of debugging text in manualroute to say "list of
1946 hosts" instead of "hostlist".
1948 2. If verify=header_syntax was set, and a header line with an unqualified
1949 address (no domain) and a large number of spaces between the end of the
1950 name and the colon was received, the reception process suffered a buffer
1951 overflow, and (when I tested it) crashed. This was caused by some obsolete
1952 code that should have been removed. The fix is to remove it!
1954 3. When running in the test harness, delay a bit after writing a bounce
1955 message to get a bit more predictability in the log output.
1957 4. Added a call to search_tidyup() just before forking a reception process. In
1958 theory, someone could use a lookup in the expansion of smtp_accept_max_
1959 per_host which, without the tidyup, could leave open a database connection.
1961 5. Added the variables $recipient_data and $sender_data which get set from a
1962 lookup success in an ACL "recipients" or "senders" condition, or a router
1963 "senders" option, similar to $domain_data and $local_part_data.
1965 6. Moved the writing of debug_print from before to after the "senders" test
1968 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1969 problems for message scanning, either using a data ACL, or using
1970 local_scan() because the Received: header was not generated till after they
1971 were called (in order to set the time as the time of reception completion).
1972 I have revised the way this works. The header is now generated after the
1973 body is received, but before the ACL or local_scan() are called. After they
1974 are run, the timestamp in the header is updated.
1980 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1981 before starting a queue runner without re-exec. This happened only when
1982 deliver_drop_privilege was set or when the Exim user was set to root. The
1983 effect of the bug was that timeouts during subsequent deliveries caused
1984 crashes instead of being properly handled. The handler is now left at its
1985 default (and expected) setting.
1987 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1988 message, again when deliver_drop_privilege is set or Exim is run as root.
1989 The bug described in (1) was not present in this case, but the tidying up
1990 of the other signals was missing. I have made the two cases consistent.
1992 3. The ignore_target_hosts setting on a manualroute router was being ignored
1993 for hosts that were looked up using the /MX notation.
1995 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1998 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1999 operated on the sender address. After changing the $sender_address to <>
2000 for the sender address verify, Exim was re-instated it as the original
2001 (before rewriting) address, but remembering that it had rewritten it, so it
2002 wasn't rewriting it again. This bug also had the effect of breaking the
2003 sender address verification caching when the sender address was rewritten.
2005 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2006 This has been changed so that if the ip literal address matches
2007 ignore_target_hosts, the router declines.
2009 7. Added expansion conditions match_domain, match_address, and match_local_
2010 part (NOT match_host).
2012 8. The placeholder for the Received: header didn't have a length field set.
2014 9. Added code to Exim itself and to exim_lock to test for a specific race
2015 condition that could lead to file corruption when using MBX delivery. The
2016 issue is with the lockfile that is created in /tmp. If this file is removed
2017 after a process has opened it but before that process has acquired a lock,
2018 there is the potential for a second process to recreate the file and also
2019 acquire a lock. This could lead to two Exim processes writing to the file
2020 at the same time. The added code performs the same test as UW imapd; it
2021 checks after acquiring the lock that its file descriptor still refers to
2022 the same named file.
2024 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2025 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2028 11. Added the smtp_active_hostname option. If used, this will typically be made
2029 to depend on the incoming interface address. Because $interface_address is
2030 not set up until the daemon has forked a reception process, error responses
2031 that can happen earlier (such as "too many connections") no longer contain
2034 12. If an expansion in a condition on a "warn" statement fails because a lookup
2035 defers, the "warn" statement is abandoned, and the next ACL statement is
2036 processed. Previously this caused the whole ACL to be aborted.
2038 13. Added the iplsearch lookup type.
2040 14. Added ident_timeout as a log selector.
2042 15. Added tls_certificate_verified as a log selector.
2044 16. Added a global option tls_require_ciphers (compare the smtp transport
2045 option of the same name). This controls incoming TLS connections.
2047 17. I finally figured out how to make tls_require_ciphers do a similar thing
2048 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2049 before starting the TLS session.
2051 18. Tabs are now shown as \t in -bP output.
2053 19. If the log selector return_path_on_delivery was set, Exim crashed when
2054 bouncing a message because it had too many Received: header lines.
2056 20. If two routers both had headers_remove settings, and the first one included
2057 a superfluous trailing colon, the final name in the first list and the
2058 first name in the second list were incorrectly joined into one item (with a
2059 colon in the middle).
2065 1. Added -C and -D options to the exinext utility, mainly to make it easier
2066 to include in the automated testing, but these could be helpful when
2067 multiple configurations are in use.
2069 2. The exinext utility was not formatting the output nicely when there was
2070 an alternate port involved in the retry record key, nor when there was a
2071 message id as well (for retries that were specific to a specific message
2072 and a specific host). It was also confused by IPv6 addresses, because of
2073 the additional colons they contain. I have fixed the IPv4 problem, and
2074 patched it up to do a reasonable job for IPv6.
2076 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2077 delivery, the log line now contains "pipelined" if PIPELINING was used.
2079 4. An SMTP transport process used to panic and die if the bind() call to set
2080 an explicit outgoing interface failed. This has been changed; it is now
2081 treated in the same way as a connect() failure.
2083 5. A reference to $sender_host_name in the part of a conditional expansion
2084 that was being skipped was still causing a DNS lookup. This no longer
2087 6. The def: expansion condition was not recognizing references to header lines
2088 that used bh_ and bheader_.
2090 7. Added the _cache feature to named lists.
2092 8. The code for checking quota_filecount in the appendfile transport was
2093 allowing one more file than it should have been.
2095 9. For compatibility with Sendmail, the command line option
2103 and sets the incoming protocol and host name (for trusted callers). The
2104 host name and its colon can be omitted when only the protocol is to be set.
2105 Note the Exim already has two private options, -pd and -ps, that refer to
2106 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2107 "s", but I don't think that's a major issue.
2109 10. A number of refactoring changes to the code, none of which should affect
2112 (a) The number of logging options was getting close to filling up the
2113 32-bit word that was used as a bit map. I have split them into two classes:
2114 those that are passed in the argument to log_write(), and those that are
2115 only ever tested independently outside of that function. These are now in
2116 separate 32-bit words, so there is plenty of room for expansion again.
2117 There is no change in the user interface or the logging behaviour.
2119 (b) When building, for example, log lines, the code previously used a
2120 macro that called string_cat() twice, in order to add two strings. This is
2121 not really sufficiently general. Furthermore, there was one instance where
2122 it was actually wrong because one of the argument was used twice, and in
2123 one call a function was used. (As it happened, calling the function twice
2124 did not affect the overall behaviour.) The macro has been replaced by a
2125 function that can join an arbitrary number of extra strings onto a growing
2128 (c) The code for expansion conditions now uses a table and a binary chop
2129 instead of a serial search (which was left over from when there were very
2130 few conditions). Also, it now recognizes conditions like "pam" even when
2131 the relevant support is not compiled in: a suitably worded error message is
2132 given if an attempt is made to use such a condition.
2134 11. Added ${time_interval:xxxxx}.
2136 12. A bug was causing one of the ddress fields not to be passed back correctly
2137 from remote delivery subprocesses. The field in question was not being
2138 subsequently used, so this caused to problems in practice.
2140 13. Added new log selectors queue_time and deliver_time.
2142 14. Might have fixed a bug in maildirsizefile handling that threw up
2143 "unexpected character" debug warnings, and recalculated the data
2144 unnecessarily. In any case, I expanded the warning message to give more
2147 15. Added the message "Restricted characters in address" to the statements in
2148 the default ACL that block characters like @ and % in local parts.
2150 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2151 Three changes have been made:
2153 (a) There was a serious bug; a negative response to MAIL caused the whole
2154 recipient domain to be cached as invalid, thereby blocking all messages
2155 to all local parts at the same domain, from all senders. This bug has
2156 been fixed. The domain is no longer cached after a negative response to
2157 MAIL if the sender used is not empty.
2159 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2162 (c) A new callout option, "use_sender" has been added for people who want
2163 the modified behaviour.
2169 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2172 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2173 indeed breaks things for older releases.
2175 3. Added additional logging to the case where there is a problem reading data
2176 from a filter that is running in a subprocess using a pipe, in order to
2177 try to track down a specific problem.
2179 4. Testing facility fudge: when running in the test harness and attempting
2180 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2181 getting "No route to host". Convert this to a timeout.
2183 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2186 6. Some OS don't have socklen_t but use size_t instead. This affects the
2187 fifth argument of getsockopt() amongst other things. This is now
2188 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2189 can be set for individual OS. I have set it for SunOS5, OSF1, and
2190 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2191 some earlier ones do not.
2193 7. Change 4.30/15 was not doing the test caselessly.
2195 8. The standard form for an IPv6 address literal was being rejected by address
2196 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2197 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2198 this, as well as the form without the "IPv6" on the front (but only when
2199 address literals are enabled, of course).
2201 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2203 10. Exim crashed if a message with an empty sender address specified by -f
2204 encountered a router with an errors_to setting. This could be provoked only
2205 by a command such as
2209 where an empty string was supplied; "<>" did not hit this bug.
2211 11. Installed PCRE release 4.5.
2213 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2214 remained set. It is now erased.
2216 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2217 times from message ids (which are base 36 rather than the normal 62).
2219 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2220 were being counted as actual protocol errors, and logged if the log
2221 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2222 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2223 and DATA following a set of rejected RCPTs do not count as protocol errors.
2224 In other words, Exim assumes they were pipelined, though this may not
2225 actually be the case. Of course, in all cases the client gets an
2226 appropriate error code.
2228 15. If a lookup fails in an ACL condition, a message about the failure may
2229 be available; it is used if testing the ACL cannot continue, because most
2230 such messages specify what the cause of the deferral is. However, some
2231 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2232 that caused an old message to be retained and used if a later statement
2233 caused a defer, replacing the real cause of the deferral.
2235 16. If an IP address had so many PTR records that the DNS lookup buffer
2236 was not large enough to hold them, Exim could crash while trying to process
2237 the truncated data. It now detects and logs this case.
2239 17. Further to 4.21/58, another change has been made: if (and only if) the
2240 first line of a message (the first header line) ends with CRLF, a bare LF
2241 in a subsequent header line has a space inserted after it, so as not to
2242 terminate the header.
2244 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2245 unnecessarily, used atoi() instead of strtol(), and didn't check the
2246 termination when getting file sizes from file names by regex.
2248 19. Completely re-implemented the support for maildirsize files, in the light
2249 of a number of problems with the previous contributed implementation
2250 (4.30/29). In particular:
2252 . If the quota is zero, the maildirsize file is maintained, but no quota is
2255 . If the maildir directory does not exist, it is created before any attempt
2256 to write a maildirsize file.
2258 . The quota value in the file is just a cache; if the quota is changed in
2259 the transport, the new value overrides.
2261 . A regular expression is available for excluding directories from the
2264 20. The autoreply transport checks the characters in options that define the
2265 message's headers; it allows continued headers, but it was checking with
2266 isspace() after an embedded newline instead of explicitly looking for a
2269 21. If all the "regular" hosts to which an address was routed had passed their
2270 expiry times, and had not reached their retry times, the address was
2271 bounced, even if fallback hosts were defined. Now Exim should go on to try
2274 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2275 equivalent code in the SMTP transport. Some hosts send humungous responses
2276 to HELO/EHLO, more than 1024 it seems.
2278 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2279 gives compiler warnings in some environments. I've now done it "properly",
2282 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2283 (because of problems with the built-in one) was declared to return uschar *
2284 instead of char *, causing compiler failure.
2286 25. Fixed a file descriptor leak when processing alias/forward files.
2288 26. Fixed a minor format string issue in dbfn.c.
2290 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2292 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2293 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2296 29. When a sender address is verified, it is cached, to save repeating the test
2297 when there is more than one recipient in a message. However, when the
2298 verification involves a callout, it is possible for different callout
2299 options to be set for different recipients. It is too complicated to keep
2300 track of this in the cache, so now Exim always runs a verification when a
2301 callout is required, relying on the callout cache for the optimization.
2302 The overhead is duplication of the address routing, but this should not be
2305 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2306 to be verified with callout=postmaster, and the main callout worked but the
2307 postmaster check failed, the verification correctly failed. However, if a
2308 subsequent RCPT command asked for sender verification *without* the
2309 postmaster check, incorrect caching caused this verification also to fail,
2312 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2313 it was not caching the DNS options (qualify_single, search_parents) that
2314 were used when the lookup failed. A subsequent lookup with different
2315 options therefore always gave the same answer, though there were cases
2316 where it should not have. (Example: a "domains = !$mx_any" option on a
2317 dnslookup router: the "domains" option is always processed without any
2318 widening, but the router might have qualify_single set.) Now Exim uses the
2319 cached value only when the same options are set.
2321 32. Added John Jetmore's "exipick" utility to the distribution.
2323 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2324 than a timeout (e.g. a certificate is required, and is not provided), an
2325 Exim server now closes the connection immediately. Previously it waited for
2326 the client to close - but if the client is SSL, it seems that they each
2327 wait for each other, leading to a delay before one of them times out.
2329 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2330 maintained 0.8.x compatibility because I don't think many are using it, and
2331 it is clearly obsolete.
2333 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2336 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2337 client certificate was expired. A simple patch fixes this, though I don't
2338 understand the full logic of why the verify callback is called multiple
2341 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2342 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2343 which causes problems with some clients (such as the Certicom SSL Plus
2344 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2345 disables the coutermeasure allowing Eudora to connect."
2347 38. Exim was not checking that a write() to a log file succeeded. This could
2348 lead to Bad Things if a log got too big, in particular if it hit a file
2349 size limit. Exim now panics and dies if it cannot write to a log file, just
2350 as it does if it cannot open a log file.
2352 39. Modified OS/Makefile-Linux so that it now contains
2354 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2356 The two -D definitions ensure that Exim is compiled with large file
2357 support, which makes it possible to handle log files that are bigger than
2360 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2361 instance) a domain was checked against a named list that involved a lookup,
2362 causing $domain_data to be set, then another domain was checked against the
2363 same list, then the first domain was re-checked, the value of $domain_data
2364 after the final check could be wrong. In particular, if the second check
2365 failed, it could be set empty. This bug probably also applied to
2368 41. The strip_trailing_dot option was not being applied to the address given
2369 with the -f command-line option.
2371 42. The code for reading a message's header from the spool was incrementing
2372 $received_count, but never initializing it. This meant that the value was
2373 incorrect (doubled) while delivering a message in the same process in which
2374 it was received. In the most common configuration of Exim, this never
2375 happens - a fresh exec is done - but it can happen when
2376 deliver_drop_privilege is set.
2378 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2379 it now includes up to 150 characters of the unexpected data in the log
2382 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2383 and building data strings. The size of both of these buffers was 10 000
2384 bytes - far larger than anybody would *ever* want, thought I. Needless to
2385 say, somebody hit the limit. I have increased the maximum line length to
2386 20 000 and the maximum data length of concatenated lines to 100 000. I have
2387 also fixed two bugs, because there was no checking on these buffers. Tsk,
2388 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2389 buffer is too small.
2391 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2392 lsearch lookups. Now it does.
2394 46. When parsing a route_list item in a manualroute router, a fixed-length
2395 buffer was used for the list of hosts. I made this 1024 bytes long,
2396 thinking that nobody would ever have a list of hosts that long. Wrong.
2397 Somebody had a whole pile of complicated expansion conditions, and the
2398 string was silently truncated, leading to an expansion error. It turns out
2399 that it is easier to change to an unlimited length (owing to other changes
2400 that have happened since this code was originally written) than to build
2401 structure for giving a limitation error. The length of the item that
2402 expands into the list of hosts is now unlimited.
2404 47. The lsearch lookup could not handle data where the length of text line was
2405 more than 4095 characters. Such lines were truncated, leading to shortened
2406 data being returned. It should now handle lines of any length.
2408 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2409 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2412 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2413 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2414 no longer happens in comment lines. A list of replacements is now placed
2415 at the head of all of the source files, except those whose only change is
2416 to replace PERL_COMMAND in the very first #! line.
2418 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2419 messages on the queue, with a bottom-up merge sort, using code contributed
2420 by Michael Haardt. This should make operations like -bp somewhat faster on
2421 large queues. It won't affect queue runners, except when queue_run_in_order
2424 51. Installed eximstats 1.31 in the distribution.
2426 52. Added support for SRV lookups to the dnslookup router.
2428 53. If an ACL referred to $message_body or $message_body_end, the value was not
2429 reset for any messages that followed in the same SMTP session.
2431 54. The store-handling optimization for building very long strings was not
2432 differentiating between the different store pools. I don't think this
2433 actually made any difference in practice, but I've tidied it.
2435 55. While running the routers to verify a sender address, $sender_address
2436 was still set to the sender address. This is wrong, because when routing to
2437 send a bounce to the sender, it would be empty. Therefore, I have changed
2438 it so that, while verifying a sender address, $sender_address is set to <>.
2439 (There is no change to what happens when verifying a recipient address.)
2441 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2442 target A or AAAA records (if not already returned) without resetting the
2443 qualify_single or search_parents options of the DNS resolver. These are
2444 inappropriate in this case because the targets of MX and SRV records must
2445 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2446 target that, when qualified, matched something in the local domain. These
2447 two options are now turned off when doing these lookups.
2449 57. It seems that at least some releases of Reiserfs (which does not have the
2450 concept of a fixed number of inodes) returns zero and not -1 for the
2451 number of available inodes. This interacted badly with check_spool_inodes,
2452 which assumed that -1 was the "no such thing" setting. What I have done is
2453 to check that the total number of inodes is greater than zero before doing
2454 the test of how many are available.
2456 58. When a "warn" ACL statement has a log_message modifier, the message is
2457 remembered, and not repeated. This is to avoid a lot of repetition when a
2458 message has many recipients that cause the same warning to be written.
2459 Howewer, Exim was preserving the list of already written lines for an
2460 entire SMTP session, which doesn't seem right. The memory is now reset if a
2461 new message is started.
2463 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2464 debugging output unless log_rewrite was also set.
2466 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2467 of (int)(handle) when we know that handle contains (void *)(-1).
2469 61. The Exim daemon panic-logs an error return when it closes the incoming
2470 connection. However "connection reset by peer" seems to be common, and
2471 isn't really an error worthy of noting specially, so that particular error
2474 62. When Exim is trying to find all the local interfaces, it used to panic and
2475 die if the ioctl to get the interface flags failed. However, it seems that
2476 on at least one OS (Solaris 9) it is possible to have an interface that is
2477 included in the list of interfaces, but for which you get a failure error
2478 for this call. This happens when the interface is not "plumbed" into a
2479 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2480 failure of the "get flags" call assumes that the interface is down.
2482 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2483 makes life easier for people who are doing arithmetic on fields extracted
2484 from dates, where you often get leading zeros that should not be
2485 interpreted as octal.
2487 64. Added qualify_domain to the redirect router, to override the global
2490 65. If a pathologically long header line contained very many addresses (the
2491 report of this problem mentioned 10 000) and each of them was rewritten,
2492 Exim could use up a very large amount of memory. (It kept on making new
2493 copies of the header line as it rewrote, and never released the old ones.)
2494 At the expense of a bit more processing, the header rewriting function has
2495 been changed so that it no longer eats memory in this way.
2497 66. The generation of the Received: header has been moved from the time that a
2498 message starts to be received, to the time that it finishes. The timestamp
2499 in the Received: header should now be very close to that of the <= log
2500 line. There are two side-effects of this change:
2502 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2503 logged header lines no longer include the local Received: line, because
2504 it has not yet been created. The same applies to a copy of the message
2505 that is returned to a non-SMTP sender when a message is rejected.
2507 (b) When a filter file is tested using -bf, no additional Received: header
2508 is added to the test message. After some thought, I decided that this
2511 This change does not affect the value of $received_for. It is still set
2512 after address rewriting, but before local_scan() is called.
2514 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2516 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2517 gave an unhelpful panic error message, and a defer error. I have managed to
2518 change this behaviour so that it now rejects any supplied certificate,
2519 which seems right, as the list of acceptable certificates is empty.
2521 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2522 gave an unhelpful defer error. I have not managed to make this reject any
2523 supplied certificates, but the error message it gives is "no certificate
2524 supplied", which is not helpful.
2526 70. exigrep's output now also includes lines that are not associated with any
2527 message, but which match the given pattern. Implemented by a patch from
2528 Martin Sluka, which also tidied up the Perl a bit.
2530 71. Recipient callout verification, like sender verification, was using <> in
2531 the MAIL FROM command. This isn't really the right thing, since the actual
2532 sender may affect whether the remote host accepts the recipient or not. I
2533 have changed it to use the actual sender in the callout; this means that
2534 the cache record is now keyed on a recipient/sender pair, not just the
2535 recipient address. There doesn't seem to be a real danger of callout loops,
2536 since a callout by the remote host to check the sender would use <>.
2537 [SEE ABOVE: changed after hitting problems.]
2539 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2540 temporary errors. However, in the case of such a code being given after
2541 the end of a data transmission (i.e. after ".") Exim was failing to write
2542 a retry record for the message. (Yes, there was some broken host that was
2543 actually sending 8xx at this point.)
2545 73. An unknown lookup type in a host list could cause Exim to panic-die when
2546 the list was checked. (An example that provoked this was putting <; in the
2547 middle of a list instead of at the start.) If this happened during a DATA
2548 ACL check, a -D file could be left lying around. This kind of configuration
2549 error no longer causes Exim to die; instead it causes a defer errror. The
2550 incident is still logged to the main and panic logs.
2552 74. Buglet left over from Exim 3 conversion. The message "too many messages
2553 in one connection" was written to the rejectlog but not the mainlog, except
2554 when address rewriting (yes!) was being logged.
2556 75. Added write_rejectlog option.
2558 76. When a system filter was run not as root (that is, when system_filter_user
2559 was set), the values of the $n variables were not being returned to the
2560 main process; thus, they were not subsequently available in the $sn
2563 77. Added +return_path_on_delivery log selector.
2565 78. A connection timeout was being treated differently from recipients deferred
2566 when testing hosts_max_try with a message that was older than the host's
2567 retry timeout. (The host should not be counted, thus allowing all hosts to
2568 be tried at least once before bouncing.) This may have been the cause of an
2569 occasionally reported bug whereby a message would remain on the queue
2570 longer than the retry timeout, but would be bounced if a delivery was
2571 forced. I say "may" because I never totally pinned down the problem;
2572 setting up timeout/retry tests is difficult. See also the next item.
2574 79. The ultimate address timeout was not being applied to errors that involved
2575 a combination of host plus message (for example, a timeout on a MAIL
2576 command). When an address resolved to a number of possible hosts, and they
2577 were not all tried for each delivery (e.g. because of hosts_max_try), a
2578 message could remain on the queue longer than the retry timeout.
2580 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2583 81. Fixed an obscure SMTP outgoing bug which required at least the following
2584 conditions: (a) there was another message waiting for the same server;
2585 (b) the server returned 5xx to all RCPT commands in the first message so
2586 that the message was not completed; (c) the server dropped the connection
2587 or gave a negative response to the RSET that Exim sends to abort the
2588 transaction. The observed case was a dropped connection after DATA that had
2589 been sent in pipelining mode. That is, the server had advertised PIPELINING
2590 but was not implementing it correctly. The effect of the bug was incorrect
2591 behaviour, such as trying another host, and this could lead to a crash.
2597 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2598 and daemon.c were passed as pointers to ints; they should have been
2599 pointers to socklen_t variables (which are typically unsigned ints).
2601 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2604 3. Fixed a really odd bug that affected only the testing scheme; patching a
2605 certain fixed string in the binary changed the value of another string that
2606 happened to be identical to the end of the original first string.
2608 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2609 name", it returns that address as the IP address. On some operating
2610 systems (e.g. Solaris), it also passes back the IP address string as the
2611 "host name". However, on others (e.g. Linux), it passes back an empty
2612 string. Exim wasn't checking for this, and was changing the host name to an
2613 empty string, assuming it had been canonicized.
2615 5. Although rare, it is permitted to have more than one PTR record for a given
2616 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2617 all the names associated with an address, because they do in Solaris.
2618 However, it seems that they do not in Linux for data that comes from the
2619 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2620 I found this out when I moved to a new Linux workstation and tried to run
2621 the Exim test suite.
2623 To get round this problem I have changed the code so that it now does its
2624 own call to the DNS to look up PTR records when searching for a host name.
2625 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2626 addresses that are only in /etc/hosts are still found.
2628 This behaviour is, however, controlled by an option called host_lookup_
2629 order, which defaults to "bydns:byaddr". If people want to use the other
2630 order, or indeed, just use one or the other means of lookup, they can
2631 specify it in this variable.
2633 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2634 some operating systems, this comes back from gethostbyaddr() as an empty
2635 string, and this is what Exim used to test for. However, it seems that in
2636 other systems, "." is yielded. Exim now tests for this case too.
2638 7. The values of check_spool_space and check_log_space are now held internally
2639 as a number of kilobytes instead of an absolute number of bytes. If a
2640 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2641 kilobyte. This means that much larger values can be stored.
2643 8. Exim monitor: an attempt to get the action menu when not actually pointing
2644 at a message produces an empty menu entitled "No message selected". This
2645 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2646 no entries in it ("Shell widget menu has zero width and/or height"). So I
2647 have added a single, blank menu entry in this case.
2649 9. Added ${quote_local_part.
2651 10. MIME decoding is now applied to the contents of Subject: header lines when
2654 11. Now that a reference to $sender_host_address automatically causes a reverse
2655 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2656 host lookup before query-style lookups in lists that might use this
2657 variable. This has therefore been abolished, and the "net-" prefix is no
2658 longer necessary for query-style lookups.
2660 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2661 have been a typo for LFLAGS, so it has been changed.
2663 13. The install script calls Exim with "-C /dev/null" in order to find the
2664 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2665 to be output. Howeve, since Exim outputs its version number before the
2666 error, it didn't break the script. It just looked ugly. I fixed this by
2667 always allowing "-C /dev/null" if the caller is root.
2669 14. Ignore overlarge ACL variable number when reading spool file - insurance
2670 against a later release with more variables having written the file.
2672 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2673 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2674 this, as well as the form without the "IPv6" on the front.
2676 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2677 OS/Makefile-Darwin file.
2679 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2680 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2682 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2683 variable. However, some OS use ints for inodes. I've added cast to long int
2684 to get rid of the compiler warning.
2686 19. I had forgotten to lock out "/../" in configuration file names when
2687 ALT_CONFIG_PREFIX was set.
2689 20. Routers used for verification do not need to specify transports. However,
2690 if such a router generated a host list, and callout was configured, Exim
2691 crashed, because it could not find a port number from the (non-existent)
2692 transport. It now assumes port 25 in this circumstance.
2694 21. Added the -t option to exigrep.
2696 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2697 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2698 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2699 them is set, LOOKUP_LSEARCH is forced.
2701 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2703 24. Added sender and host information to the "rejected by local_scan()" log
2704 line; previously there was no indication of these.
2706 25. Added .include_if_exists.
2708 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2709 Linux. It turns out that not all file systems support this. Apparently some
2710 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2711 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2712 not supported on the file descriptor, is now ignored when Exim is trying to
2713 sync a directory. This applies only to Linux.
2715 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2717 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2718 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2721 29. Added support for maildirsize files from supplied patch (modified a bit).
2723 30. The use of :fail: followed by an empty string could lead Exim to respond to
2724 sender verification failures with (e.g.):
2726 550 Verification failed for <xxx>
2727 550 Sender verify failed
2729 where the first response line was missing the '-' that indicates it is not
2730 the final line of the response.
2732 31. The loop for finding the name of the user that called Exim had a hardwired
2733 limit of 10; it now uses the value of finduser_retries, which is used for
2734 all other user lookups.
2736 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2739 33. Exim was neglecting to zero errno before one call of strtol() when
2740 expanding a string and expecting an integer value. On some systems this
2741 resulted in spurious "integer overflow" errors. Also, it was casting the
2742 result into an int without checking.
2744 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2745 did not work. The code looks as if it has *never* worked, though it appears
2746 to have been documented since at least releast 1.62. I have made it work.
2748 35. The "timeout_DNS" error in retry rules, also documented since at least
2749 1.62, also never worked. As it isn't clear exactly what this means, and
2750 clearly it isn't a major issue, I have abolished the feature by treating it
2751 as "timeout", and writing a warning to the main and panic logs.
2753 36. The display of retry rules for -brt wasn't always showing the error code
2756 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2757 timeout_connect_A, timeout_connect_MX.
2759 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2760 to the empty sender.
2762 39. The daemon was not analysing the content of -oX till after it had closed
2763 stderr and disconnected from the controlling terminal. This meant that any
2764 syntax errors were only noted on the panic log, and the return code from
2765 the command was 0. By re-arranging the code a little, I've made the
2766 decoding happen first, so such errors now appear on stderr, and the return
2767 code is 1. However, the actual setting up of the sockets still happens in
2768 the disconnected process, so errors there are still only recorded on the
2771 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2772 connections (as happens on some IP stacks) was logged at start up time as
2773 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2774 it from "IPv6 and IPv4", which means that two separate sockets are being
2777 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2778 says whether AF_INET or AF_INET6 was passed as an argument.
2780 42. Exiwhat output was messed up when time zones were included in log
2783 43. Exiwhat now gives more information about the daemon's listening ports,
2784 and whether -tls-on-connect was used.
2786 44. The "port" option of the smtp transport is now expanded.
2788 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2789 silently ignored. Now an error message is written to the main and panic
2792 46. There's a new ACL modifier called "logwrite" which writes to a log file
2793 as soon as it is encountered.
2795 47. Added $local_user_uid and $local_user_gid at routing time.
2797 48. Exim crashed when trying to verify a sender address that was being
2800 49. Exim was recognizing only a space character after ".include". It now also
2801 recognizes a tab character.
2803 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2804 extracting the relevant information from the specification. The man page no
2805 longer contains scrambled data for the -d option, and I've added a section
2806 at the front about calling Exim under different names.
2808 51. Added "extra_headers" argument to the "mail" command in filter files.
2810 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2813 53. Installed eximstats 1.29.
2815 54. Added transport_filter_timeout as a generic transport option.
2817 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2818 Cc: header lines. This was required by RFC 822, but it not required by RFC
2821 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2822 incoming messages that did not have them. Now it does so only if the
2823 message originates locally, that is, if there is no associated remote host
2824 address. When Resent- header lines are present, this applies to the Resent-
2825 lines rather than the non-Resent- lines.
2827 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2828 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2830 58. Messages for configuration errors now include the name of the main
2831 configuration file - useful now that there may be more than one file in a
2832 list (.included file names were always shown).
2834 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2835 for those rare installations that do not start the daemon as root or run it
2836 setuid root. I've cut out the call to initgroups() if the daemon is not
2839 60. The Exim user and group can now be bound into the binary as text strings
2840 that are looked up at the start of Exim's processing.
2842 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2844 62. Added $mailstore_basename variable.
2846 63. Installed patch to sieve.c from Michael Haardt.
2848 64. When Exim failed to open the panic log after failing to open the main log,
2849 the original message it was trying to log was written to stderr and debug
2850 output, but if they were not available (the usual case in production), it
2851 was lost. Now it is written to syslog before the two lines that record the
2852 failures to open the logs.
2854 65. Users' Exim filters run in subprocesses under the user's uid. It is
2855 possible for a "deliver" command or an alias in a "personal" command to
2856 provoke an address rewrite. If logging of address rewriting is configured,
2857 this fails because the process is not running as root or exim. There may be
2858 a better way of dealing with this, but for the moment (because 4.30 needs
2859 to be released), I have disabled address rewrite logging when running a
2860 filter in a non-root, non-exim process.
2866 1. The buildconfig auxiliary program wasn't quoting the value set for
2867 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2868 not defined. This bug was masked in 4.22 by the effect that was fixed in
2871 2. Some messages that were rejected after a message id was allocated were
2872 shown as "incomplete" by exigrep. It no longer does this for messages that
2873 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2875 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2876 have allow_domain_literals set, the ID did not get logged in the <= line.
2877 Domain literals are now always recognized in Message-ID: header lines.
2879 4. The first argument for a ${extract expansion item is the key name or field
2880 number. Leading and trailing spaces in this item were not being ignored,
2881 causing some misleading effects.
2883 5. When deliver_drop_privilege was set, single queue runner processes started
2884 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2885 same command in the process it spins off) were not dropping privilege.
2887 6. When the daemon running as "exim" started a queue runner, it always
2888 re-executed Exim in the spun-off process. This is a waste of effort when
2889 deliver_drop_privilege is set. The new process now just calls the
2890 queue-runner function directly.
2896 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2899 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2900 ignored. Though the use of -oP was forcing the writing of a pid file, it
2901 was always written to the default place.
2903 3. If the message "no IP address found for host xxxx" is generated during
2904 incoming verification, it is now followed by identification of the incoming
2905 connection (so you can more easily find what provoked it).
2907 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2909 5. Added some features to "harden" Exim a bit more against certain attacks:
2911 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2912 be put in Local/Makefile. This is like the never_users runtime option,
2913 but it cannot be overridden. The default setting is "root".
2915 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2916 prefix string with which any file named in a -C command line option
2919 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2920 is retained for -C and -D only if the caller of Exim is root. Without
2921 it, the exim user may also use -C and -D and retain privilege.
2923 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2924 command line option is disabled.
2926 6. Macro names set by the -D option must start with an upper case letter, just
2927 like macro names defined in the configuration file.
2929 7. Added "dereference=" facility to LDAP.
2931 8. Two instances of the typo "uknown" in the source files are fixed.
2933 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2934 the Configure-Makefile script screwed up while processing it.
2936 10. Incorporated PCRE 4.4.
2938 11. The SMTP synchronization check was not operating right at the start of an
2939 SMTP session. For example, it could not catch a HELO sent before the client
2940 waited for the greeting. There is now a check for outstanding input at the
2941 point when the greeting is written. Because of the duplex, asynchronous
2942 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2943 way, but not yet received, when the check is performed.
2945 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2946 on TCP/IP sockets, because this apparently causes some broken clients to
2949 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2950 unchanged) from the Cygwin maintainer.
2952 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2953 was supported instead of testing for mbx. Effectively a typo.
2955 15. The spa authenticator server code was not checking that the input it
2956 received was valid base64.
2958 16. The debug output line for the "set" modifier in ACLs was not showing the
2959 name of the variable that was being set.
2961 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2963 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2964 lookup on its own. Something else was needed to trigger the lookup. For
2965 example, a match in host_lookup or the need for a host name in a host list.
2966 Now, if $sender_host_name is referenced and the host name has not yet been
2967 looked up, a lookup is performed. If the lookup fails, the variable remains
2968 empty, and $host_lookup_failed is set to "1".
2970 19. Added "eqi" as a case-independent comparison operator.
2972 20. The saslauthd authentication condition could segfault if neither service
2973 nor realm was specified.
2975 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2976 error message that was logged was misleading, and incoming SMTP
2977 connections were dropped. The message is now more accurate, and temporary
2978 errors are given to SMTP connections.
2980 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2981 (except RSET) with a 421 error, until QUIT is received. However, it was
2982 failing to send a response to QUIT.
2984 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2985 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2986 ACL is now run after the helo_try_verify_hosts code.
2988 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2989 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2990 case-independent, so other case variants are also recognized). Apparently
2991 some systems use these upper case variants.
2993 25. If more than two messages were waiting for the same host, and a transport
2994 filter was specified for the transport, Exim sent two messages over the
2995 same TCP/IP connection, and then failed with "socket operation on non-
2996 socket" when it tried to send the third.
2998 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3000 27. The extern definition of crypt16() in expand.c was not being excluded when
3001 the OS had its own crypt16() function.
3003 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3004 as a preferred synonym for return_size_limit, both as an option and as an
3007 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3009 30. Changed the default configuration ACL to relax the local part checking rule
3010 for addresses that are not in any local domains. For these addresses,
3011 slashes and pipe symbols are allowed within local parts, but the sequence
3012 /../ is explicitly forbidden.
3014 31. SPA server authentication was not clearing the challenge buffer before
3017 32. log_message in a "warn" ACL statement was writing to the reject log as
3018 well as to the main log, which contradicts the documentation and doesn't
3019 seem right (because no rejection is happening). So I have stopped it.
3021 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3022 However, I am unable to do any testing of this.
3024 34. Fixed an infelicity in the appendfile transport. When checking directories
3025 for a mailbox, to see if any needed to be created, it was accidentally
3026 using path names with one or more superfluous leading slashes; tracing
3027 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3029 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3030 contents are added to the log line that is written for every discarded
3031 recipient. (Previously a log_message setting was ignored.)
3033 36. The ${quote: operator now quotes the string if it is empty.
3035 37. The install script runs exim in order to find its version number. If for
3036 some reason other than non-existence or emptiness, which it checks, it
3037 could not run './exim', it was installing it with an empty version number,
3038 i.e. as "exim-". This error state is now caught, and the installation is
3041 38. An argument was missing from the function that creates an error message
3042 when Exim fails to connect to the socket for saslauthd authentication.
3043 This could cause Exim to crash, or give a corrupted message.
3045 39. Added isip, isip4, and isip6 to ${if conditions.
3047 40. The ACL variables $acl_xx are now saved with the message, and can be
3048 accessed later in routers, transports, and filters.
3050 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3051 strings in the file are not string-expanded.
3053 42. If a MAIL command specified a SIZE value that was too large to fit into an
3054 int variable, the check against message_size_limit failed. Such values are
3055 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3056 day this will have to be increased, but I don't think I want to be around
3057 when emails are that large.
3064 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3065 iconv() is not standard in FreeBSD.
3067 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3068 IPv6 enabled. The observed symptom was a segmentation fault on return from
3069 the function os_common_find_running_interfaces() in src/os.c.
3071 3. In the check_special_case() function in daemon.c I had used "errno" as an
3072 argument name, which causes warnings on some systems. This was basically a
3073 typo, since it was named "eno" in the comments!
3075 4. The code that waits for the clock to tick (at a resolution of some fraction
3076 of a second) so as to ensure message-id uniqueness was always waiting for
3077 at least one whole tick, when it could have waited for less. [This is
3078 almost certainly not relevant at current processor speeds, where it is
3079 unlikely to ever wait at all. But we try to future-proof.]
3081 5. The function that sleeps for a time interval that includes fractions of a
3082 second contained a race. It did not block SIGALRM between setting the
3083 timer, and suspending (a couple of lines later). If the interval was short
3084 and the sigsuspend() was delayed until after it had expired, the suspension
3085 never ended. On busy systems this could lead to processes getting stuck for
3088 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3089 process, before it forks any delivery processes. The open lookup caching
3090 mechanism meant that the open file or database connection was passed into
3091 the delivery process. The problem was that delivery processes always tidy
3092 up cached lookup data. This could cause a problem for the next delivery
3093 process started by the queue runner, because the external queue runner
3094 process does not know about the closure. So the next delivery process
3095 still has data in the lookup cache. In the case of a file lookup, there was
3096 no problem because closing a file descriptor in a subprocess doesn't affect
3097 the parent. However, if the lookup was caching a connection to a database,
3098 the connection was closed, and the second delivery process was likely to
3099 see errors such as "PGSQL: query failed: server closed the connection
3100 unexpectedly". The problem has been fixed by closing all cached lookups
3101 in a queue runner before running a delivery process.
3103 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3104 seem to have the "const" qualifier which it has on other OS. I've
3107 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3108 *of the same type* (client or server) with the same public name that an
3109 error should be diagnosed.
3111 9. When Exim looked up a host name for an IP address, but failed to find the
3112 original IP address when looking up the host name (a safety check), it
3113 output the message "<ip address> does not match any IP for NULL", which was
3114 confusing, to say the least. The bug was that the host name should have
3115 appeared instead of "NULL".
3117 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3118 user that is built into the binary, and the -C or -D options is used, root
3119 privilege is dropped before the configuration file is read. In addition,
3120 logging is switched to stderr instead of the normal log files. If the
3121 configuration then re-defines the Exim user, the unprivileged environment
3122 is probably not what is expected, so Exim logs a panic warning message (but
3125 However, if deliver_drop_privilege is set, the unprivileged state may well
3126 be exactly what is intended, so the warning has been cut out in that case,
3127 and Exim is allowed to try to write to its normal log files.
3133 1. smtp_return_error_details was not giving details for temporary sender
3134 or receiver verification errors.
3136 2. Diagnose a configuration error if two authenticators have the same public
3139 3. Exim used not to create the message log file for a message until the first
3140 delivery attempt. This could be confusing when incoming messages were held
3141 for policy or load reasons. The message log file is now created at the time
3142 the message is received, and an initial "Received" line is written to it.
3144 4. The automatically generated man page for command line options had a minor
3145 bug that caused no ill effects; however, a more serious problem was that
3146 the procedure for building the man page automatically didn't always
3147 operate. Consequently, release 4.20 contains an out-of-date version. This
3148 shouldn't happen again.
3150 5. When building Exim with embedded Perl support, the script that builds the
3151 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3152 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3154 6. The freeze_tell option was not being used for messages that were frozen on
3155 arrival, either by an ACL or by local_scan().
3157 7. Added the smtp_incomplete_transaction log selector.
3159 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3160 was accepting AUTH without a new EHLO.
3162 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3163 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3166 10. Logging of TCP/IP connections (when configured) now happens in the main
3167 daemon process instead of the child process, so that the TCP/IP connection
3168 count is more accurate (but it can never be perfect).
3170 11. The use of "drop" in a nested ACL was not being handled correctly in the
3171 outer ACL. Now, if condition failure induced by the nested "drop" causes
3172 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3173 or "require"), the connection is dropped.
3175 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3176 that yield "discard" can now be used with an "accept" or a "discard" verb,
3177 but an error is generated for any others (because I can't see a useful way
3178 to define what should happen).
3180 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3181 are now processed in the same way as lines in the Exim configuration file.
3182 In particular, continuation lines are supported.
3184 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3186 15. Added -ti meaning -t -i.
3188 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3189 domains, and warn by logging if others are found.
3191 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3192 returned from GIFCONF when Exim is trying to find the list of interfaces on
3193 a host. The code in os.c has been modified to copy each ifreq to an aligned
3194 structure in all cases.
3196 Also, in some cases, the returned ifreq's were being copied to a 'struct
3197 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3198 means the last couple of bytes of an IPv6 address could be chopped if the
3199 ifreq contained only a normal sockaddr (14 bytes storage).
3201 18. Named domain lists were not supported in the hosts_treat_as_local option.
3202 An entry such as +xxxx was not recognized, and was treated as a literal
3205 19. Ensure that header lines added by a DATA ACL are included in the reject log
3206 if the ACL subsequently rejects the message.
3208 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3209 MD5 (which is deprecated).
3211 21. When testing a filter file using -bf, Exim was writing a message when it
3212 took the sender from a "From " line in the message, but it was not doing so
3213 when it took $return_path from a Return-Path: header line. It now does.
3215 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3216 with a valid header line field name (a series of printing characters
3217 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3219 23. Changed "disc" in the source to "disk" to conform to the documentation and
3220 the book and for uniformity.
3222 24. Ignore Sendmail's -Ooption=value command line item.
3224 25. When execve() failed while trying to run a command in a pipe transport,
3225 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3226 could be confused with a return value of 69 from the command itself. This
3227 has been changed to 127, the value the shell returns if it is asked to run
3228 a non-existent command. The wording for the related log line suggests a
3229 non-existent command as the problem.
3231 26. If received_header_text expands to an empty string, do not add a Received:
3232 header line to the message. (Well, it adds a token one on the spool, but
3233 marks it "old" so that it doesn't get used or transmitted.)
3235 27. Installed eximstats 1.28 (addition of -nt option).
3237 28. There was no check for failure on the call to getsockname() in the daemon
3238 code. This can fail if there is a shortage of resources on the system, with
3239 ENOMEM, for example. A temporary error is now given on failure.
3241 29. Contrary to the C standard, it seems that in some environments, the
3242 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3243 program. Exim now does this explicitly; it affects the formatting of
3244 timestamps using strftime().
3246 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3247 complaints. I've now initialized all the variables, to avoid this.
3249 32. Header lines added by a system filter were not being "seen" during
3250 transport-time rewrites.
3252 33. The info_callback() function passed to OpenSSL is set up with type void
3253 (*)(SSL *, int, int), as described somewhere. However, when calling the
3254 function (actually a macro) that sets it up, the type void(*)() is
3255 expected. I've put in a cast to prevent warnings from picky compilers.
3257 34. If a DNS black list lookup found a CNAME record, but there were no A
3258 records associated with the domain it pointed at, Exim crashed.
3260 35. If a DNS black list lookup returned more than one A record, Exim ignored
3261 all but the first. It now scans all returned addresses if a particular IP
3262 value is being sought. In this situation, the contents of the
3263 $dnslist_value variable are a list of all the addresses, separated by a
3266 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3267 to accept a host name and all its aliases if the forward lookup for any of
3268 them yielded the IP address of the incoming connection. Now it accepts only
3269 those names whose forward lookup yields the correct IP address. Any other
3270 names are discarded. This closes a loophole whereby a rogue DNS
3271 administrator could create reverse DNS records to break through a
3272 wildcarded host restriction in an ACL.
3274 37. If a user filter or a system filter that ran in a subprocess used any of
3275 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3276 the wrong values were passed to the pipe command ($thisaddress had the
3277 value of $0, $0 had the value of $1, etc). This bug was introduced by
3278 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3280 38. Improved the line breaking for long SMTP error messages from ACLs.
3281 Previously, if there was no break point between 40 and 75 characters, Exim
3282 left the rest of the message alone. Two changes have been made: (a) I've
3283 reduced the minimum length to 35 characters; (b) if it can't find a break
3284 point between 35 and 75 characters, it looks ahead and uses the first one
3285 that it finds. This may give the occasional overlong line, but at least the
3286 remaining text gets split now.
3288 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3289 file descriptors might be low, and that setting 1000 would always raise it.
3290 It turns out that in some environments, the limit is already over 1000 and
3291 that lowering it causes trouble. So now Exim takes care not to decrease it.
3293 40. When delivering a message, the value of $return_path is set to $sender_
3294 address at the start of routing (routers may change the value). By an
3295 oversight, this default was not being set up when an address was tested by
3296 -bt or -bv, which affected the outcome if any router or filter referred to
3299 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3300 log only when "message" or "log_message" are set. However, if one of the
3301 conditions was an address verification, or a call to a nested ACL, the
3302 messages generated by the underlying test were being passed through. This
3303 no longer happens. The underlying message is available in $acl_verify_
3304 message for both "message" and "log_message" expansions, so it can be
3305 passed through if needed.
3307 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3308 new expansion $bh_ to give the encoded byte string without charset
3309 translation. Translation happens only if iconv() is available; HAVE_ICONV
3310 indicates this at build time. HEADERS_CHARSET gives the charset to
3311 translate to; headers_charset can change it in the configuration, and
3312 "headers charset" can change it in an individual filter file.
3314 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3315 that creates RFC 2047 encoded "words" labels them as that charset instead
3316 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3317 expansion operator; (ii) when Exim creates a From: line for a local
3318 message; (iii) when a header line is rewritten to include a "phrase" part.
3320 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3321 buggy, causing it to skip the first lines of messages whose message ID
3322 ended in 'D'. This would not have bitten before Exim release 4.14, because
3323 message IDs were unlikely to end in 'D' before then. The effect was to have
3324 incorrect size information for certain domains.
3326 45. #include "config.h" was missing at the start of the crypt16.c module. This
3327 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3330 46. If there was a timeout during a "random" callout check, Exim treated it as
3331 a failure of the random address, and carried on sending RSET and the real
3332 address. If the delay was just some slowness somewhere, the response to the
3333 original RCPT would be taken as a response to RSET and so on, causing
3334 mayhem of various kinds.
3336 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3337 when I implemented it. It didn't allow for the fact that some option values
3338 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3339 the right test for positive values.
3341 48. Domain names in DNS records are case-independent. Exim always looks them up
3342 in lower case. Some resolvers return domain names in exactly the case they
3343 appear in the zone file, that is, they may contain uppercase letters. Not
3344 all resolvers do this - some return always lower case. Exim was treating a
3345 change of case by a resolver as a change of domain, similar to a widening
3346 of a domain abbreviation. This triggered its re-routing code and so it was
3347 trying to route what was effectively the same domain again. This normally
3348 caused routing to fail (because the router wouldn't handle the domain
3349 twice). Now Exim checks for this case specially, and just changes the
3350 casing of the domain that it ultimately uses when it transmits the message
3353 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3356 50. If a filter generated a file delivery with a non-absolute name (possible if
3357 no home directory exists for the router), the forbid_file option was not
3360 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3361 the existing equality matching.
3363 52. Exim was using ints instead of ino_t variables in some places where it was
3364 dealing with inode numbers.
3366 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3367 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3368 called TMPDIR, and if it finds it is different, it changes its value.
3370 54. The smtp_printf() function is now made available to local_scan() so
3371 additional output lines can be written before returning. There is also an
3372 smtp_fflush() function to enable the detection of a dropped connection.
3373 The variables smtp_input and smtp_batched_input are exported to
3376 55. Changed the default runtime configuration: the message "Unknown user"
3377 has been removed from the ACL, and instead placed on the localuser router,
3378 using the cannot_route_message feature. This means that any verification
3379 failures that generate their own messages won't get overridden. Similarly,
3380 the "Unrouteable address" message that was in the ACL for unverifiable
3381 relay addresses has also been removed.
3383 56. Added hosts_avoid_esmtp to the smtp transport.
3385 57. The exicyclog script was not checking for the esoteric option
3386 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3387 will work only if exicyclog is run under the appropriate euid.
3389 58. Following a discussion on the list, the rules by which Exim recognises line
3390 endings on incoming messages have been changed. The -dropcr and drop_cr
3391 options are now no-ops, retained only for backwards compatibility. The
3392 following line terminators are recognized: LF CRLF CR. However, special
3393 processing applies to CR:
3395 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3396 nor a local message in the state where . is a terminator.
3398 (ii) If a bare CR is encountered in a header line, an extra space is added
3399 after the line terminator so as not to end the header. The reasoning
3400 behind this is that bare CRs in header lines are most likely either
3401 to be mistakes, or people trying to play silly games.
3403 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3404 was being incorrectly given as 18 bytes larger than it should have been.
3405 This is a VOB (very old bug).
3407 60. This may never have affected anything current, but just in case it has:
3408 When the local host is found other than at the start of a list of hosts,
3409 the local host, those with the same MX, and any that follow, are discarded.
3410 When the list in question was part of a longer list of hosts, the following
3411 hosts (not currently being processed) were also being discarded. This no
3412 longer happens. I'm not sure if this situation could ever has previously
3415 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3418 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3419 header. This is something that is recommended in a new Internet Draft, and
3420 is something that is documented as being done by Sendmail. There are two
3421 possible values. For messages generated by the autoreply transport, Exim
3424 Auto-Submitted: auto-replied
3426 whereas for all other generated messages (e.g. bounces) it adds
3428 Auto-Submitted: auto-generated
3430 63. The "personal" condition in filters now includes a test for the
3431 Auto-Submitted: header. If it contains the string "auto-" the message it
3432 not considered personal.
3434 64. Added rcpt_include_affixes as a generic transport option.
3436 65. Added queue_only_override (default true).
3438 66. Added the syslog_duplication option.
3440 67. If what should have been the first header line of a message consisted of
3441 a space followed by a colon, Exim was mis-interpreting it as a header line.
3442 It isn't of course - it is syntactically invalid and should therefore be
3443 treated as the start of the message body. The misbehaviour could have
3444 caused a number of strange effects, including loss of data in subsequent
3445 header lines, and spool format errors.
3447 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3448 client host had authenticated. This control can now be exercised by an ACL
3449 for more flexibility.
3451 69. By default, callouts do not happen when testing with -bh. There is now a
3452 variant, -bhc, which does actually run the callout code, including
3453 consulting and updating the callout cache.
3455 70. Added support for saslauthd authentication, courtesy of Alexander
3458 71. If statvfs() failed on the spool or log directories while checking their
3459 size for availability, Exim confusingly gave the error "space shortage".
3460 Furthermore, in debugging mode it crashed with a floating point exception.
3461 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3462 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3463 serious problem, Exim now writes to the main and panic logs when this
3464 happens, with details of the failure. It then refuses to accept the
3465 incoming message, giving the message "spool directory problem" or "log
3466 directory problem" with a 421 code for SMTP messages.
3468 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3469 0, 1, and 2 exist, because some OS complain for execs without them (see
3470 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3471 descriptors. However, the code omitted to check that the open succeeded,
3472 causing mysterious errors if for some reason the permissions on /dev/null
3473 got screwed. Now Exim writes a message to the main and panic logs, and
3474 bombs out if it can't open /dev/null.
3476 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3477 interact so that it is all more flexible. It is supposed to remain
3478 backwards compatible. Also added extra_local_interfaces.
3480 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3481 to bomb out with an assertion failure - to the client this appears as a
3482 connection drop. This problem occurs in the part of the code that was taken
3483 from the Samba project. Fortunately, the assertion is in a very simple
3484 function, so I have fixed this by reproducing the function inline in the
3485 one place where it is called, and arranging for authentication to fail
3486 instead of killing the process with assert().
3488 75. The SPA client code was not working when the server requested OEM rather
3489 than Unicode encoding.
3491 76. Added code to make require_files with a specific uid setting more usable in
3492 the case where statting the file as root fails - usually a non-root-mounted
3493 NFS file system. When this happens and the failure is EACCES, Exim now
3494 forks a subprocess and does the per-uid checking as the relevant uid.
3496 77. Added process_log_path.
3498 78. If log_file_path was not explicitly set, a setting of check_log_space or
3499 check_log_inodes was ignored.
3501 79. If a space check for the spool or log partitions fails, the incident is now
3502 logged. Of course, in the latter case the data may get lost...
3504 80. Added the %p formatting code to string_format() so that it can be used to
3505 print addresses in debug_print(). Adjusted all the address printing in the
3506 debugging in store.c to use %p rather than %d.
3508 81. There was a concern that a line of code in smtp_in.c could overflow a
3509 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3510 initially expressed, the concern was not well-founded, because trailing
3511 spaces are removed early. However, if the trailing spaces were followed by
3512 a NULL, they did not get removed, so the overflow was possible. Two fixes
3515 (a) I re-wrote the offending code in a cleaner fashion.
3516 (b) If an incoming SMTP command contains a NULL character, it is rejected
3519 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3520 runs initgroups(), so that if the Exim user is in any additional groups,
3521 they will be used during message reception.
3527 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob