1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.261 2005/11/15 09:44:33 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 Two changes to the default runtime configuration:
11 (1) Move the checks for relay_from_hosts and authenticated clients from
12 after to before the (commented out) DNS black list checks.
14 (2) Add control=submission to the relay_from_hosts and authenticated
15 clients checks, on the grounds that messages accepted by these
16 statements are most likely to be submissions.
18 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
20 (1) Generate an error if the third argument for the ${prvs expansion is
23 (2) Treat a missing third argument of ${prvscheck as if it were an empty
26 (3) Reset the variables that are obtained from the first argument of
27 ${prvscheck and used in the second argument before leaving the code,
28 because their memory is reclaimed, so using them afterwards may do
31 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
32 one (it's much easier than Tom thought :-).
34 (5) Because of (4), we can now allow for the use of $prvscheck_result
35 inside the third argument.
37 PH/03 For some reason, the default setting of PATH when running a command from
38 a pipe transport was just "/usr/bin". I have changed it to
41 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
42 anything to be listed in the output from -bV.
44 PH/05 When a filter generated an autoreply, the entire To: header line was
45 quoted in the delivery log line, like this:
47 => >A.N.Other <ano@some.domain> <original@ddress> ...
49 This has been changed so that it extracts the operative address. There
50 may be more than one such address. If so, they are comma-separated, like
53 => >ano@some.domain,ona@other.domain <original@ddress> ...
55 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
56 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
57 not being looked up in the rDNS to get a host name, Exim was showing the
58 IP address twice in Received: lines, even though the IP addresses were
59 identical. For example:
61 Received: from [1.2.3.4] (helo=[1.2.3.4])
63 However, if the real host name was known, it was omitting the HELO data
64 if it matched the actual IP address. This has been tidied up so that it
65 doesn't show the same IP address twice.
67 PH/07 When both +timestamp and +memory debugging was on, the value given by
68 $tod_xxx expansions could be wrong, because the tod_stamp() function was
69 called by the debug printing, thereby overwriting the timestamp buffer.
70 Debugging no longer uses the tod_stamp() function when +timestamp is set.
72 PH/08 When the original message was included in an autoreply transport, it
73 always said "this is a copy of the message, including all the headers",
74 even if body_only or headers_only was set. It now gives an appropriate
77 PH/09 Applied a patch from the Sieve maintainer which:
80 o adds the (disabled) notify extension core
81 o adds some debug output for the result of if/elsif tests
82 o points to the current vacation draft in the documentation
83 and documents the missing references header update
87 o fixes a bug in processing the envelope test (when testing
88 multiple envelope elements, the last element determinted the
91 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
92 Electronic Mail") by including:
94 Auto-submitted: auto-generated
96 in the messages that it generates (bounce messages and others, such as
97 warnings). In the case of bounce messages for non-SMTP mesages, there was
98 also a typo: it was using "Auto_submitted" (underscore instead of
99 hyphen). Since every message generated by Exim is necessarily in response
100 to another message, thes have all been changed to:
102 Auto-Submitted: auto-replied
104 in accordance with these statements in the RFC:
106 The auto-replied keyword:
108 - SHOULD be used on messages sent in direct response to another
109 message by an automatic process,
111 - MUST NOT be used on manually-generated messages,
113 - MAY be used on Delivery Status Notifications (DSNs) and Message
114 Disposition Notifications (MDNs),
116 - MUST NOT be used on messages generated by automatic or periodic
117 processes, except for messages which are automatic responses to
120 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
121 to the default Received: header definition.
123 PH/12 Added log selector acl_warn_skipped (default on).
125 PH/13 After a successful wildlsearch lookup, discard the values of numeric
126 variables because (a) they are in the wrong storage pool and (b) even if
127 they were copied, it wouldn't work properly because of the caching.
133 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
134 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
137 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
138 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
140 PH/03 Typo: missing ".o" in src/pcre/Makefile.
142 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
143 header line, restrict the check to what is listed in RFCs 2369 and 2929.
144 Also, for "Auto-Submitted", treat anything other than "no" as
145 non-personal, in accordance with RFC 3834. (Previously it treated
146 anything starting "auto-" as non-personal.)
148 TF/01 The control=submission/name=... option had a problem with syntax
149 errors if the name included a slash character. The /name= option
150 now slurps the rest of the string, so it can include any characters
151 but it must come last in the list of options (after /sender_retain
154 PH/05 Some modifications to the interface to the fake nameserver for the new
157 PH/06 Add ${if def:sender_address {(envelope-from <$sender_address>)\n\t}}
158 into the default Received: header string.
165 TK/01 Added the "success_on_redirect" address verification option. See
166 NewStuff for rationale and an example.
168 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
170 PH/02 Patch to exigrep to allow it to work on syslog lines.
172 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
173 fread() to skip over the body file's header line, because in Cygwin the
174 header line is locked and is inaccessible.
176 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
177 co-exist for some time) to make it clear that it is the Exim ID that is
178 referenced, not the Message-ID: header line.
180 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
181 string_format() function, because snprintf() does not exist on all
184 PH/06 The use of forbid_filter_existstest now also locks out the use of the
185 ${stat: expansion item.
187 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
188 protocol synchronization error", to keep the pedants happy.
190 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
191 well as for IRIX systems, when gcc is being used. See the host.c source
194 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
196 PH/10 Named domain lists were not working if used in a queue_smtp_domains
199 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
200 transport and to the smtp transport in LMTP mode.
202 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
204 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
205 run a filter in a subprocess. This could lead to confusion in subsequent
206 lookups in the parent process. There should also be a search_tidyup() at
207 the end of the subprocess.
209 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
210 only if the host matched helo_try_verify_hosts, which caused the
211 verification to occur when the EHLO/HELO command was issued. The ACL just
212 tested the remembered result. Now, if a previous verification attempt has
213 not happened, "verify = helo" does it there and then.
215 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
217 TK/03 Fix log output including CR from clamd.
219 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
220 exist provoked a memory error which could cause a segfault.
222 PH/15 Installed PCRE 6.2
224 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
226 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
227 of the problem. Specifically, suggested +O2 rather than +O1 for the
230 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
232 PH/20 If a delivery was routed to a non-standard port by means of an SRV
233 record, the port was not correctly logged when the outgoing_port log
234 selector was set (it logged the transort's default port).
236 PH/21 Added support for host-specific ports to manualroute, queryprogram,
237 fallback_hosts, and "hosts" in the smtp transport.
239 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
240 host errors such as "Connection refused".
242 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
243 authentication with radiusclient 0.4.9:
245 - Error returned from rc_read_config was caught wrongly
246 - Username/password not passed on to radius server due to wrong length.
248 The presumption is that some radiusclient API changes for 4.51/PH/17
249 were not taken care of correctly. The code is still untested by me (my
250 Linux distribution still has 0.3.2 of radiusclient), but it was
251 contributed by a Radius user.
253 PH/24 When doing a callout, the value of $domain wasn't set correctly when
254 expanding the "port" option of the smtp transport.
256 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
257 while reading a MIME header. Thanks to Tom Hughes for a patch.
259 PH/24 Include config.h inside local_scan.h so that configuration settings are
262 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
263 that in an ACL for RCPT (for example), you can examine exactly what was
266 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
267 commands, but it was not correctly comparing the address with the actual
268 client host address. Thus, it would show the EHLO address in Received:
269 header lines when this was not necessary.
271 PH/27 Added the % operator to ${eval:}.
273 PH/28 Exim tries to create and chdir to its spool directory when it starts;
274 it should be ignoring failures (because with -C, for example, it has lost
275 privilege). It wasn't ignoring creation failures other than "already
278 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
281 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
282 because an input line was too long, either on its own, or by virtue of
283 too many continuations, the temporary file was not being removed, and the
284 return code was incorrect.
286 PH/31 Missing "BOOL" in function definition in filtertest.c.
288 PH/32 Applied Sieve patches from the maintainer.
290 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
292 PH/33 Added "verify = not_blind".
294 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
295 Local/Makefile (with some defaults set). These are used in built scripts
296 such as exicyclog, but they have never been used in the exim_install
297 script (though there are many overriding facilities there). I have
298 arranged that the exim_install script now takes note of these two
301 PH/35 Installed configuration files for Dragonfly.
303 PH/36 When a locally submitted message by a trusted user did not contain a
304 From: header, and the sender address was obtained from -f or from an SMTP
305 MAIL command, and the trusted user did not use -F to supply a sender
306 name, $originator_name was incorrectly used when constructing a From:
307 header. Furthermore, $originator_name was used for submission mode
308 messages from external hosts without From: headers in a similar way,
309 which is clearly wrong.
311 PH/37 Added control=suppress_local_fixups.
313 PH/38 When log_selector = +received_sender was set, and the addition of the
314 sender made the log line's construction buffer exactly full, or one byte
315 less than full, an overflow happened when the terminating "\n" was
318 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
319 when the result of a list match is failure because a DNS lookup failed.
321 PH/40 RM_COMMAND is now used in the building process.
323 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
324 the "build-* directories that it finds.
326 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
327 address in a domain literal was a prefix of an interface address.
329 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
330 when verifying a sender address, unless rewrite_headers is false.
332 PH/44 Wrote a long comment about why errors_to addresses are verified as
333 recipients, not senders.
335 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
336 the ratelimit ACL was added.
338 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
340 PH/46 Added extra information about PostgreSQL errors to the error string.
342 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
343 avoiding the need to install special zones in a real server. This is
344 backwards compatible; if it can't find the fake resolver, it drops back.
345 Thus, both old and new test suites can be run.
347 TF/02 Added util/ratelimit.pl
349 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
350 clock is set back in time.
352 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
353 Candler <B.Candler@pobox.com>.
355 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
356 OK for addresses that are the result of redirections.
358 PH/48 A number of further additions for the benefit of the new test suite,
359 including a fake gethostbyname() that interfaces to the fake DNS resolver
362 TF/06 The fix for widen_domains has also been applied to qualify_single and
363 search_parents which are the other dnslookup options that can cause
366 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
369 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
371 TF/07 Exim produced the error message "an SRV record indicated no SMTP
372 service" if it encountered an MX record with an empty target hostname.
373 The message is now "an MX or SRV record indicated no SMTP service".
375 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
376 if the DNS of the sending site is misconfigured. This is quite a
377 common situation. This change restores the behaviour of treating a
378 helo verification defer as a failure.
380 PH/51 If self=fail was set on a router, the bounce message did not include the
381 actual error message.
387 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
389 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
390 command itself ended in error, the underlying message about the transport
391 filter timeout was being overwritten with the pipe command error. Now the
392 underlying error message should be appended to the second error message.
394 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
396 PH/02 Reduce the amount of output that "make" produces by default. Full output
397 can still be requested.
399 PH/03 The warning log line about a condition test deferring for a "warn" verb
400 was being output only once per connection, rather than after each
401 occurrence (because it was using the same function as for successful
402 "warn" verbs). This seems wrong, so I have changed it.
404 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
405 it should not have, which might have caused a crash in the right
406 circumstances, but probably never did.
408 PH/04 Installed a modified version of Tony Finch's patch to make submission
409 mode fix the return path as well as the Sender: header line, and to
410 add a /name= option so that you can make the user's friendly name appear
413 TF/03 Added the control = fakedefer ACL modifier.
415 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
416 Mark Lowes for thorough testing.
418 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
420 TK/03 Merged latest SRS patch from Miles Wilton.
422 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
423 with the definition in sysexits.h (which is #included earlier).
424 Fortunately, Exim does not actually use EX_OK. The code used to try to
425 preserve the sysexits.h value, by assumimg that macro definitions were
426 scanned for macro replacements. I have been disabused of this notion,
427 so now the code just undefines EX_OK before #including unistd.h.
429 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
430 in the smtp transport. When a block could not be written in a single
431 write() function, the timeout was being re-applied to each part-write.
432 This seems wrong - if the receiver was accepting one byte at a time it
433 would take for ever. The timeout is now adjusted when this happens. It
434 doesn't have to be particularly precise.
436 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
437 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
439 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
440 without a domain if the check to <postmaster@domain> fails.
442 SC/01 Eximstats: added -xls and the ability to specify output files
443 (patch written by Frank Heydlauf).
445 SC/02 Eximstats: use FileHandles for outputing results.
447 SC/03 Eximstats: allow any combination of xls, txt, and html output.
449 SC/04 Eximstats: fixed display of large numbers with -nvr option
451 SC/05 Eximstats: fixed merging of reports with empty tables.
453 SC/06 Eximstats: added the -include_original_destination flag
455 SC/07 Eximstats: removed tabs and trailing whitespace.
457 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
459 TK/06 MBOX spool code: Add real "From " MBOX separator line
460 so the .eml file is really in mbox format (even though
461 most programs do not really care). Patch from Alex Miller.
463 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
464 The latter is generated from $received_to and is only set if the
465 message has one envelope recipient. SA can use these headers,
466 obviously out-of-the-box. Patch from Alex Miller.
468 PH/08 The ${def test on a variable was returning false if the variable's
469 value was "0", contrary to what the specification has always said!
470 The result should be true unless the variable is empty.
472 PH/09 The syntax error of a character other than { following "${if
473 def:variable_name" (after optional whitespace) was not being diagnosed.
474 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
475 accidental colon was present, for example, could give incorrect results.
477 PH/10 Tidied the code in a number of places where the st_size field of a stat()
478 result is used (not including appendfile, where other changes are about
481 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
482 This involved changing a lot of size variables from int to off_t. It
483 should work with maildirs and everything.
485 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
486 spamd dying while we are connected to it.
488 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
491 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
492 function for BDB 4.3.
494 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
496 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
497 the sources for PCRE have been re-arranged and more files are now
500 PH/15 The code I had for printing potentially long long variables in PH/11
501 above was not the best (it lost precision). The length of off_t variables
502 is now inspected at build time, and an appropriate printing format (%ld
503 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
504 to be "long long int" or "long int". This is needed for the internal
505 formatting function string_vformat().
507 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
508 the configuration file to be ":syslog", then the script "guesses" where
509 the logs files are, rather than using the compiled in default. In our
510 case the guess is not the same as the compiled default, so the script
511 suddenly stopped working when I started to use syslog. The patch checks
512 to see if log_file_path is "". If so, it attempts to read it from exim
513 with no configuration file to get the compiled in version, before it
514 falls back to the previous guessing code."
516 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
517 implementing BATV in an Exim configuration. See NewStuff for the gory
520 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
521 Makefile that are specific to HP-UX.
523 PH/18 If the "use_postmaster" option was set for a recipient callout together
524 with the "random" option, the postmaster address was used as the MAIL
525 FROM address for the random test, but not for the subsequent recipient
526 test. It is now used for both.
528 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
529 patch removes a few documentation additions to RFC 3028, because the
530 latest draft now contains them. It adds the new en;ascii-case comparator
531 and a new error check for 8bit text in MIME parts. Comparator and
532 require names are now matched exactly. I enabled the subaddress
533 extension, but it is not well tested yet (read: it works for me)."
535 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
536 rework some of the code of TK/09 above to avoid the hardwired use of
537 "%lld" and "long long". Replaced the call to snprintf() with a call to
540 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
541 records point to non-existent hosts", "retry timeout exceeded", and
542 "retry time not reached for any host after a long failure period".
544 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
545 experimental DomainKeys support:
547 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
548 (2) On an error such as an illegally used "control", the wrong name for
549 the control was given.
551 These problems did NOT occur unless DomainKeys support was compiled.
553 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
555 PH/24 Added ${if match_ip condition.
557 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
558 databases so that it will be absolutely obvious if a crash occurs in the
559 DB library. This is a regular occurrence (often caused by mis-matched
562 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
563 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
564 picked up on a user's system that detects such things. There doesn't seem
565 to be a gcc warning option for this - only an attribute that has to be
566 put on the function's prototype. It seems that in Fedora Core 4 they have
567 set this on a number of new functions. No doubt there will be more in due
570 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
571 specify a transport. However, if an address that was verified by such a
572 router was the subject of a callout, Exim crashed because it tried to
573 read the rcpt_include_affixes from the non-existent transport. Now it
574 just assumes that the setting of that option is false. This bug was
575 introduced by 4.51/PH/31.
577 PH/28 Changed -d+all to exclude +memory, because that information is very
578 rarely of interest, but it makes the output a lot bigger. People tend to
579 do -d+all out of habit.
581 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
582 code in os-type was giving problems when libc.so lives in lib64, like on
585 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
586 aren't the modern standard, and it seems that some systems' include files
587 don't always have them. Exim was already checking for some of the newer
588 ones like T_AAAA, and defining it itself. I've added checks for all the
589 record types that Exim uses.
591 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
592 not automatically generating a new one, as it is supposed to. This
593 prevented TLS from working. If the file did exist, but contained invalid
594 data, a new version was generated, as expected. It was only the case of a
595 non-existent file that was broken.
597 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
598 with a change in libdomainkeys > 0.64.
600 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
601 from DNS. If the selector record carries the flag, it now has
602 precedence over the domain-wide flag.
604 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
606 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
607 the use of a port name that isn't defined in /etc/services) occurred, the
608 message was deferred as in a normal delivery, and thus remained on the
609 spool, instead of being failed because of the mua_wrapper setting. This
610 is now fixed, and I tidied up some of the mua_wrapper messages at the
613 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
614 the messages in a hash of arrays rather than using individual hashes.
615 This is a bit cleaner and results in dramatic memory savings, albeit
616 at a slight CPU cost.
618 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
619 as requested by Marc Sherman.
621 SC/10 Eximstats: added histograms for user specified patterns as requested
624 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
626 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
627 fopen() in the content-scanning modules that did not already have it.
633 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
634 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
636 TK/02 Fix ACL "control" statement not being available in MIME ACL.
638 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
640 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
641 to test Sieve filters that use "vacation".
643 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
644 that changes the way the GnuTLS parameters are stored in the cache file.
645 The new format can be generated externally. For backward compatibility,
646 if the data in the cache doesn't make sense, Exim assumes it has read an
647 old-format file, and it generates new data and writes a new file. This
648 means that you can't go back to an older release without removing the
651 PH/03 A redirect router that has both "unseen" and "one_time" set does not
652 work if there are any delivery delays because "one_time" forces the
653 parent to be marked "delivered", so its unseen clone is never tried
654 again. For this reason, Exim now forbids the simultaneous setting of
657 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
658 redirected to themselves ("homonym" addresses). Read the long ChangeLog
659 entry if you want to know the details. The fix, however, neglected to
660 consider the case when local delivery batching is involved. The test for
661 "previously delivered" was not happening when checking to see if an
662 address could be batched with a previous (undelivered) one; under
663 certain circumstances this could lead to multiple deliveries to the same
666 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
667 in its include files, and this causes problems building Exim.
669 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
670 header_syntax) but Exim was just ignoring anything given after a slash.
671 In particular, this caused confusion with an attempt to use "verify =
672 reverse_host_lookup/defer_ok". An error is now given when options are
673 supplied for verify items that do not have them. (Maybe reverse_host_
674 lookup should have a defer_ok option, but that's a different point.)
676 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
677 defined by RFC 821) to 2048, because there were problems with some AUTH
678 commands, and RFC 1869 says the size should be increased for extended
679 SMTP commands that take arguments.
681 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
684 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
685 "unknown" error; now it says that the functionality isn't in the binary.
687 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
688 an address' error message when a string expansion fails (syntax or
689 whatever). Otherwise the password may appear in the log. Following change
690 PH/42 below, there is no longer a chance of it appearing in a bounce
693 PH/11 Installed exipick version 20050225.0 from John Jetmore.
695 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
696 of its addresses was ever tried. (Bugzilla bug #2.)
698 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
699 the result incorrectly in the debug output. (It correctly added a newline
700 to what was transported.)
702 TF/01 Added $received_time.
704 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
705 commented out examples of how to interface to a virus scanner and to
706 SpamAssassin. Also added commented examples of av_scanner and
707 spamd_address settings.
709 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
710 and controls are allowed in which ACLs. There were a couple of minor
711 errors. Some of the entries in the conditions table (which is a table of
712 where they are NOT allowed) were getting very unwieldy; rewrote them as a
713 negation of where the condition IS allowed.
715 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
717 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
718 header file does not have a version number, so I've had to invent a new
719 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
720 API. The code is untested by me (my Linux distribution still has 0.3.2 of
721 radiusclient), but it was contributed by a Radius user.
723 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
724 files or directories, for OpenSSL.
726 PH/19 When an Exim process that is running as root has to create an Exim log
727 file, it does so in a subprocess that runs as exim:exim so as to get the
728 ownership right at creation (otherwise, other Exim processes might see
729 the file with the wrong ownership). There was no test for failure of this
730 fork() call, which would lead to the process getting stuck as it waited
731 for a non-existent subprocess. Forks do occasionally fail when resources
732 run out. I reviewed all the other calls to fork(); they all seem to check
735 PH/20 When checking for unexpected SMTP input at connect time (before writing
736 the banner), Exim was not dealing correctly with a non-positive return
737 from the read() function. If the client had disconnected by this time,
738 the result was a log entry for a synchronization error with an empty
739 string after "input=" when read() returned zero. If read() returned -1
740 (an event I could not check), uninitialized data bytes were printed.
741 There were reports of junk text (parts of files, etc) appearing after
744 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
746 PH/22 Added support for macro redefinition, and (re)definition in between
747 driver and ACL definitions.
749 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
750 forgetting to use the resulting value; it was using the unexpanded value.
752 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
753 hadn't been configured. The fix is from Juergen Kreileder, who
754 understands it better than I do:
756 "Here's what I see happening with three configured cyrus_sasl
757 authenticators configured (plain, login, cram-md5):
759 On startup auth_cyrus_sasl_init() gets called for each of these.
760 This means three calls to sasl_listmech() without a specified mech_list.
761 => SASL tests which mechs of all available mechs actually work
762 => three warnings about OTP not working
763 => the returned list contains: plain, login, cram-md5, digest-md5, ...
765 With the patch, sasl_listmech() also gets called three times. But now
766 SASL's mech_list option is set to the server_mech specified in the the
767 authenticator. Or in other words, the answer from sasl_listmech()
768 gets limited to just the mech you're testing for (which is different
770 => the return list contains just 'plain' or 'login', 'cram-md5' or
771 nothing depending on the value of ob->server_mech.
773 I've just tested the patch: Authentication still works fine,
774 unavailable mechs specified in the exim configuration are still
775 caught, and the auth.log warnings about OTP are gone."
777 PH/25 When debugging is enabled, the contents of the command line are added
778 to the debugging output, even when log_selector=+arguments is not
781 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
782 answer is "GNU", and only if the return is "GNU/something" is the answer
785 PH/27 $acl_verify_message is now set immediately after the failure of a
786 verification in an ACL, and so is available in subsequent modifiers. In
787 particular, the message can be preserved by coding like this:
789 warn !verify = sender
790 set acl_m0 = $acl_verify_message
792 Previously, $acl_verify_message was set only while expanding "message"
793 and "log_message" when a very denied access.
795 PH/28 Modified OS/os.c-Linux with
797 -#ifndef OS_LOAD_AVERAGE
798 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
800 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
801 nomenclature these days.)
803 PH/29 Installed patch from the Sieve maintainer that adds the options
804 sieve_useraddress and sieve_subaddress to the redirect router.
806 PH/30 In these circumstances:
807 . Two addresses routed to the same list of hosts;
808 . First host does not offer TLS;
809 . First host accepts first address;
810 . First host gives temporary error to second address;
811 . Second host offers TLS and a TLS session is established;
812 . Second host accepts second address.
813 Exim incorrectly logged both deliveries with the TLS parameters (cipher
814 and peerdn, if requested) that were in fact used only for the second
817 PH/31 When doing a callout as part of verifying an address, Exim was not paying
818 attention to any local part prefix or suffix that was matched by the
819 router that accepted the address. It now behaves in the same way as it
820 does for delivery: the affixes are removed from the local part unless
821 rcpt_include_affixes is set on the transport.
823 PH/32 Add the sender address, as F=<...>, to the log line when logging a
824 timeout during the DATA phase of an incoming message.
826 PH/33 Sieve envelope tests were broken for match types other than :is. I have
827 applied a patch sanctioned by the Sieve maintainer.
829 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
830 the uid or gid is negative. A case of a negative gid caused this to be
831 noticed. The fix allows for either to be negative.
833 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
834 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
837 PH/36 The change PH/12 above was broken. Fixed it.
839 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
840 the grounds that routing the same address twice would always produce the
841 same answer. This might have been true once, but it is certainly no
842 longer true now. Routing a child address may depend on the previous
843 routing that produced that child. Some complicated redirection strategies
844 went wrong when messages had multiple recipients, and made Exim's
845 behaviour dependent on the order in which the addresses were given.
847 I have moved the duplicate checking until after the routing is complete.
848 Exim scans the addresses that are assigned to local and remote
849 transports, and removes any duplicates. This means that more work will be
850 done, as duplicates will always all be routed, but duplicates are
851 presumably rare, so I don't expect this is of any significance.
853 For deliveries to pipes, files, and autoreplies, the duplicate checking
854 still happens during the routing process, since they are not going to be
857 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
858 It corrects a timeout issue with spamd. This is Ian's comment: "The
859 background is that sometimes spamd either never reads data from a
860 connection it has accepted, or it never writes response data. The exiscan
861 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
862 blindly assumes that writes won't block so it may never time out."
864 PH/39 Allow G after quota size as well as K and M.
866 PH/40 The value set for $authenticated_id in an authenticator may not contain
867 binary zeroes or newlines because the value is written to log lines and
868 to spool files. There was no check on this. Now the value is run through
869 the string_printing() function so that such characters are converted to
870 printable escape sequences.
872 PH/41 $message_linecount is a new variable that contains the total number of
873 lines in the message. Compare $body_linecount, which is the count for the
876 PH/42 Exim no longer gives details of delivery errors for specific addresses in
877 bounce and delay warning messages, except in certain special cases, which
880 (a) An SMTP error message from a remote host;
881 (b) A message specified in a :fail: redirection;
882 (c) A message specified in a "fail" command in a system filter;
883 (d) A message specified in a FAIL return from the queryprogram router;
884 (e) A message specified by the cannot_route_message router option.
886 In these cases only, Exim does include the error details in bounce and
887 warning messages. There are also a few cases where bland messages such
888 as "unrouteable address" or "local delivery error" are given.
890 PH/43 $value is now also set for the "else" part of a ${run expansion.
892 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
893 being worked on, but at least Exim now implements the latest version to
896 PH/45 In a pipe transport, although a timeout while waiting for the pipe
897 process to complete was treated as a delivery failure, a timeout while
898 writing the message to the pipe was logged, but erroneously treated as a
899 successful delivery. Such timeouts include transport filter timeouts. For
900 consistency with the overall process timeout, these timeouts are now
901 treated as errors, giving rise to delivery failures by default. However,
902 there is now a new Boolean option for the pipe transport called
903 timeout_defer, which, if set TRUE, converts the failures into defers for
904 both kinds of timeout. A transport filter timeout is now identified in
907 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
908 systems where "make" and "gmake" are different, calling "gmake" at top
909 level broke things. I've arranged for the value of $(MAKE) to be passed
910 from the Makefile to this script so that it can call the same version of
914 A note about Exim versions 4.44 and 4.50
915 ----------------------------------------
917 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
918 changes of various kinds. As a consequence, a big documentation update was
919 needed. This delayed the release for rather longer than seemed good, especially
920 in the light of a couple of (minor) security issues. Therefore, the changes
921 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
922 release. So 4.44 and 4.50 are in effect two different branches that both start
925 I have left the 4.50 change log unchanged; it contains all the changes since
926 4.43. The change log for 4.44 is below; many of its items are identical to
927 those for 4.50. This seems to be the most sensible way to preserve the
928 historical information.
934 1. Minor wording change to the doc/README.SIEVE file.
936 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
937 computation of the current number of files was incorrect.
939 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
940 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
943 4. Give more explanation in the error message when the command for a transport
944 filter fails to execute.
946 5. There are several places where Exim runs a non-Exim command in a
947 subprocess. The SIGUSR1 signal should be disabled for these processes. This
948 was being done only for the command run by the queryprogram router. It is
949 now done for all such subprocesses. The other cases are: ${run, transport
950 filters, and the commands run by the lmtp and pipe transports.
952 6. Added CONFIGURE_GROUP build-time option.
954 7. Some older OS have a limit of 256 on the maximum number of file
955 descriptors. Exim was using setrlimit() to set 1000 as a large value
956 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
957 systems. I've change it so that if it can't get 1000, it tries for 256.
959 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
960 was an oversight, and furthermore, ever since the addition of extra
961 controls (e.g. 4.43/32), the checks on when to allow different forms of
962 "control" were broken. There should now be diagnostics for all cases when a
963 control that does not make sense is encountered.
965 9. Added the /retain_sender option to "control=submission".
967 10. $recipients is now available in the predata ACL (oversight).
969 11. Tidy the search cache before the fork to do a delivery from a message
970 received from the command line. Otherwise the child will trigger a lookup
971 failure and thereby defer the delivery if it tries to use (for example) a
972 cached ldap connection that the parent has called unbind on.
974 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
975 of $address_data from the recipient verification was clobbered by the
978 13. The value of address_data from a sender verification is now available in
979 $sender_address_data in subsequent conditions in the ACL statement.
981 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
983 15. Added a new option "connect=<time>" to callout options, to set a different
986 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
987 was its contents. (It was OK if the option was not defined at all.)
989 17. A "Completed" log line is now written for messages that are removed from
990 the spool by the -Mrm option.
992 18. New variables $sender_verify_failure and $recipient_verify_failure contain
993 information about exactly what failed.
995 19. Added -dd to debug only the daemon process.
997 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
998 handles timeouts, both on the server side and network timeouts. Renamed the
999 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1001 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1002 It is now set to "smtps".
1004 22. $host_address is now set to the target address during the checking of
1005 ignore_target_hosts.
1007 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1008 being passed; this would have caused $sender_host_name to have been used if
1009 matching the list had actually called for a host name (not very likely,
1010 since this list is usually IP addresses). A host name is now passed as
1013 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1014 code that specifies a non-restarting handler (typically sigaction() in
1015 modern systems) in an attempt to fix a rare and obscure crash bug.
1017 25. Narrowed the window for a race in the daemon that could cause it to ignore
1018 SIGCHLD signals. This is not a major problem, because they are used only to
1019 wake it up if nothing else does.
1021 26. A malformed maildirsize file could cause Exim to calculate negative values
1022 for the mailbox size or file count. Odd effects could occur as a result.
1023 The maildirsize information is now recalculated if the size or filecount
1026 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1027 support for a long time. Removed HAVE_SYS_VFS_H.
1029 28. Installed the latest version of exipick from John Jetmore.
1031 29. In an address list, if the pattern was not a regular expression, an empty
1032 subject address (from a bounce message) matched only if the pattern was an
1033 empty string. Non-empty patterns were not even tested. This was the wrong
1034 because it is perfectly reasonable to use an empty address as part of a
1035 database query. An empty address is now tested by patterns that are
1036 lookups. However, all the other forms of pattern expect the subject to
1037 contain a local part and a domain, and therefore, for them, an empty
1038 address still always fails if the pattern is not itself empty.
1040 30. Exim went into a mad DNS loop when attempting to do a callout where the
1041 host was specified on an smtp transport, and looking it up yielded more
1042 than one IP address.
1044 31. Re-factored the code for checking spool and log partition space into a
1045 function that finds that data and another that does the check. The former
1046 is then used to implement four new variables: $spool_space, $log_space,
1047 $spool_inodes, and $log_inodes.
1049 32. The RFC2047 encoding function was originally intended for short strings
1050 such as real names; it was not keeping to the 75-character limit for
1051 encoded words that the RFC imposes. It now respects the limit, and
1052 generates multiple encoded words if necessary. To be on the safe side, I
1053 have increased the buffer size for the ${rfc2047: expansion operator from
1056 33. It is now permitted to omit both strings after an "if" condition; if the
1057 condition is true, the result is "true". As before, when the second string
1058 is omitted, a false condition yields an empty string. This makes it less
1059 cumbersome to write custom ACL and router conditions.
1061 34. Failure to deliver a bounce message always caused it to be frozen, even if
1062 there was an errors_to setting on the router. The errors_to setting is now
1065 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1066 canonical form (fully expanded) before being placed in
1067 $sender_host_address.
1069 36. The table in the code that translates DNS record types into text (T_A to
1070 "A" for instance) was missing entries for NS and CNAME. It is just possible
1071 that this could have caused confusion if both these types were looked up
1072 for the same domain, because the text type is used as part of Exim's
1073 per-process caching. But the chance of anyone hitting this buglet seems
1076 37. The dnsdb lookup has been extended in a number of ways.
1078 (1) There is a new type, "zns", which walks up the domain tree until it
1079 finds some nameserver records. It should be used with care.
1081 (2) There is a new type, "mxh", which is like "mx" except that it returns
1082 just the host names, not the priorities.
1084 (3) It is now possible to give a list of domains (or IP addresses) to be
1085 looked up. The behaviour when one of the lookups defers can be
1086 controlled by a keyword.
1088 (4) It is now possible to specify the separator character for use when
1089 multiple records are returned.
1091 38. The dnslists ACL condition has been extended: it is now possible to supply
1092 a list of IP addresses and/or domains to be looked up in a particular DNS
1095 39. Added log_selector=+queue_time_overall.
1097 40. When running the queue in the test harness, wait just a tad after forking a
1098 delivery process, to get repeatability of debugging output.
1100 41. Include certificate and key file names in error message when GnuTLS fails
1101 to set them up, because the GnuTLS error message doesn't include the name
1102 of the failing file when there is a problem reading it.
1104 42. Allow both -bf and -bF in the same test run.
1106 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1108 44. The "Exiscan patch" is now merged into the mainline Exim source.
1110 45. Sometimes the final signoff response after QUIT could fail to get
1111 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1112 < 0 before doing a fflush(). This bug looks as though it goes back to the
1113 introduction of TLS in release 3.20, but "sometimes" must have been rare
1114 because the tests only now provoked it.
1116 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1117 (this can affect the format of dates).
1119 47. exim_tidydb, when checking for the continued existence of a message for
1120 which it has found a message-specific retry record, was not finding
1121 messages that were in split spool directories. Consequently, it was
1122 deleting retry records that should have stayed in existence.
1124 48. Steve fixed some bugs in eximstats.
1126 49. The SPA authentication driver was not abandoning authentication and moving
1127 on to the next authenticator when an expansion was forced to fail,
1128 contradicting the general specification for all authenticators. Instead it
1129 was generating a temporary error. It now behaves as specified.
1131 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1132 (the order specifies the preference for clients). The order is now AES256,
1133 AES128, 3DES, ARCFOUR128.
1135 51. Small patch to Sieve code - explicitly set From: when generating an
1138 52. Exim crashed if a remote delivery caused a very long error message to be
1139 recorded - for instance if somebody sent an entire SpamAssassin report back
1140 as a large number of 550 error lines. This bug was coincidentally fixed by
1141 increasing the size of one of Exim's internal buffers (big_buffer) that
1142 happened as part of the Exiscan merge. However, to be on the safe side, I
1143 have made the code more robust (and fixed the comments that describe what
1146 53. Now that there can be additional text after "Completed" in log lines (if
1147 the queue_time_overall log selector is set), a one-byte patch to exigrep
1148 was needed to allow it to recognize "Completed" as not the last thing in
1151 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1152 patch that reportedly fixes this has been added. I am not expert enough to
1153 create a test for it. This is what the patch creator wrote:
1155 "I found a little strange behaviour of ldap code when working with
1156 Windows 2003 AD Domain, where users was placed in more than one
1157 Organization Units. When I tried to give exim partial DN, the exit code
1158 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1159 But simultaneously result of request was absolutely normal ldap result,
1160 so I produce this patch..."
1162 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1163 REFERENCE, so I have modified the code to exclude the patch when that macro
1166 55. Some experimental protocols are using DNS PTR records for new purposes. The
1167 keys for these records are domain names, not reversed IP addresses. The
1168 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1169 leaves it alone. Component reversal etc. now happens only for IP addresses.
1172 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1174 57. Double the size of the debug message buffer (to 2048) so that more of very
1175 long debug lines gets shown.
1177 58. The exicyclog utility now does better if the number of log files to keep
1178 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1180 59. Two changes related to the smtp_active_hostname option:
1182 (1) $smtp_active_hostname is now available as a variable.
1183 (2) The default for smtp_banner uses $smtp_active_hostname instead
1184 of $primary_hostname.
1186 60. The host_aton() function is supposed to be passed a string that is known
1187 to be a valid IP address. However, in the case of IPv6 addresses, it was
1188 not checking this. This is a hostage to fortune. Exim now panics and dies
1189 if the condition is not met. A case was found where this could be provoked
1190 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1191 components; fortuitously, this particular loophole had already been fixed
1192 by change 4.50/55 above.
1194 If there are any other similar loopholes, the new check in host_aton()
1195 itself should stop them being exploited. The report I received stated that
1196 data on the command line could provoke the exploit when Exim was running as
1197 exim, but did not say which command line option was involved. All I could
1198 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1199 running as the user.
1202 61. There was a buffer overflow vulnerability in the SPA authentication code
1203 (which came originally from the Samba project). I have added a test to the
1204 spa_base64_to_bits() function which I hope fixes it.
1207 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1208 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1210 63. The daemon start-up calls getloadavg() while still root for those OS that
1211 need the first call to be done as root, but it missed one case: when
1212 deliver_queue_load_max is set with deliver_drop_privilege. This is
1213 necessary for the benefit of the queue runner, because there is no re-exec
1214 when deliver_drop_privilege is set.
1216 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1217 This has been fixed.
1219 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1220 was in use, was not putting the data itself into the right store pool;
1221 consequently, it could be overwritten for a subsequent message in the same
1222 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1225 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1227 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1228 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1229 the function were treating the return as a boolean value, which happened to
1230 work because 0=false and not-0=true, but is not correct code.
1232 68. The host_aton() function was not handling scoped IPv6 addresses (those
1233 with, for example, "%eth0" on the end) correctly.
1235 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1236 negated items (that is, ~something) in unsigned ints. Some compilers
1237 apparently mutter when there is no cast.
1239 70. If an address verification called from an ACL failed, and did not produce a
1240 user-specific message (i.e. there was only a "system" message), nothing was
1241 put in $acl_verify_message. In this situation, it now puts the system
1244 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1245 session; change 4.31/43 added the unwanted input to the log line - except
1246 that it did not do this in the start of session case. It now does.
1248 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1249 This is wrong and can cause the other end to generate a synchronization
1250 error if it is another Exim or anything else that does the synchronization
1251 check. A QUIT command is no longer sent after a timeout.
1253 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1254 during host lookups.
1256 74. The defer_ok option of callout verification was not working if it was used
1257 when verifying addresses in header lines, that is, for this case:
1259 verify = header_sender/callout=defer_ok
1261 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1262 those file descriptors could be used for SMTP connections. If anything
1263 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1264 could be sent to the SMTP client, causing chaos. The daemon now opens
1265 stdin, stdout, and stderr to /dev/null when it puts itself into the
1268 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1269 log by default. The user can override this with suitable Perl magic.
1271 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1272 the log message when discard triggers, was not working for the DATA ACL or
1273 for the non-SMTP ACL.
1275 78. Error message wording change in sieve.c.
1277 79. If smtp_accept_max_per_host was set, the number of connections could be
1278 restricted to fewer than expected, because the daemon was trying to set up
1279 a new connection before checking whether the processes handling previous
1280 connections had finished. The check for completed processes is now done
1281 earlier. On busy systems, this bug wouldn't be noticed because something
1282 else would have woken the daemon, and it would have reaped the completed
1285 80. If a message was submitted locally by a user whose login name contained one
1286 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1287 It caused a spool format error. I have fixed the spool reading code. A
1288 related problem was that the "from" clause in the Received: line became
1289 illegal because of the space(s). It is now covered by ${quote_local_part.
1291 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1294 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1296 83. Patch from Sieve maintainer to fix unterminated string problem in
1297 "vacation" handling.
1299 84. Some minor changes to the Linux configuration files to help with other
1300 OS variants using glibc.
1302 85. One more patch for Sieve to update vacation handling to latest spec.
1305 ----------------------------------------------------
1306 See the note above about the 4.44 and 4.50 releases.
1307 ----------------------------------------------------
1313 1. Change 4.43/35 introduced a bug that caused file counts to be
1314 incorrectly computed when quota_filecount was set in an appendfile
1317 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1318 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1321 3. Give more explanation in the error message when the command for a transport
1322 filter fails to execute.
1324 4. There are several places where Exim runs a non-Exim command in a
1325 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1326 was being done only for the command run by the queryprogram router. It is
1327 now done for all such subprocesses. The other cases are: ${run, transport
1328 filters, and the commands run by the lmtp and pipe transports.
1330 5. Some older OS have a limit of 256 on the maximum number of file
1331 descriptors. Exim was using setrlimit() to set 1000 as a large value
1332 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1333 systems. I've change it so that if it can't get 1000, it tries for 256.
1335 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1336 was an oversight, and furthermore, ever since the addition of extra
1337 controls (e.g. 4.43/32), the checks on when to allow different forms of
1338 "control" were broken. There should now be diagnostics for all cases when a
1339 control that does not make sense is encountered.
1341 7. $recipients is now available in the predata ACL (oversight).
1343 8. Tidy the search cache before the fork to do a delivery from a message
1344 received from the command line. Otherwise the child will trigger a lookup
1345 failure and thereby defer the delivery if it tries to use (for example) a
1346 cached ldap connection that the parent has called unbind on.
1348 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1349 of $address_data from the recipient verification was clobbered by the
1350 sender verification.
1352 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1353 was its contents. (It was OK if the option was not defined at all.)
1355 11. A "Completed" log line is now written for messages that are removed from
1356 the spool by the -Mrm option.
1358 12. $host_address is now set to the target address during the checking of
1359 ignore_target_hosts.
1361 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1362 being passed; this would have caused $sender_host_name to have been used if
1363 matching the list had actually called for a host name (not very likely,
1364 since this list is usually IP addresses). A host name is now passed as
1367 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1368 code that specifies a non-restarting handler (typically sigaction() in
1369 modern systems) in an attempt to fix a rare and obscure crash bug.
1371 15. Narrowed the window for a race in the daemon that could cause it to ignore
1372 SIGCHLD signals. This is not a major problem, because they are used only to
1373 wake it up if nothing else does.
1375 16. A malformed maildirsize file could cause Exim to calculate negative values
1376 for the mailbox size or file count. Odd effects could occur as a result.
1377 The maildirsize information is now recalculated if the size or filecount
1380 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1381 support for a long time. Removed HAVE_SYS_VFS_H.
1383 18. Updated exipick to current release from John Jetmore.
1385 19. Allow an empty sender to be matched against a lookup in an address list.
1386 Previously the only cases considered were a regular expression, or an
1389 20. Exim went into a mad DNS lookup loop when doing a callout where the
1390 host was specified on the transport, if the DNS lookup yielded more than
1393 21. The RFC2047 encoding function was originally intended for short strings
1394 such as real names; it was not keeping to the 75-character limit for
1395 encoded words that the RFC imposes. It now respects the limit, and
1396 generates multiple encoded words if necessary. To be on the safe side, I
1397 have increased the buffer size for the ${rfc2047: expansion operator from
1400 22. Failure to deliver a bounce message always caused it to be frozen, even if
1401 there was an errors_to setting on the router. The errors_to setting is now
1404 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1405 canonical form (fully expanded) before being placed in
1406 $sender_host_address.
1408 24. Updated eximstats to version 1.33
1410 25. Include certificate and key file names in error message when GnuTLS fails
1411 to set them up, because the GnuTLS error message doesn't include the name
1412 of the failing file when there is a problem reading it.
1414 26. Expand error message when OpenSSL has problems setting up cert/key files.
1417 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1418 (this can affect the format of dates).
1420 28. exim_tidydb, when checking for the continued existence of a message for
1421 which it has found a message-specific retry record, was not finding
1422 messages that were in split spool directories. Consequently, it was
1423 deleting retry records that should have stayed in existence.
1425 29. eximstats updated to version 1.35
1426 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1427 1.35 - bugfix such that pie charts by volume are generated correctly
1429 30. The SPA authentication driver was not abandoning authentication and moving
1430 on to the next authenticator when an expansion was forced to fail,
1431 contradicting the general specification for all authenticators. Instead it
1432 was generating a temporary error. It now behaves as specified.
1434 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1435 (the order specifies the preference for clients). The order is now AES256,
1436 AES128, 3DES, ARCFOUR128.
1438 31. Small patch to Sieve code - explicitly set From: when generating an
1441 32. Exim crashed if a remote delivery caused a very long error message to be
1442 recorded - for instance if somebody sent an entire SpamAssassin report back
1443 as a large number of 550 error lines. This bug was coincidentally fixed by
1444 increasing the size of one of Exim's internal buffers (big_buffer) that
1445 happened as part of the Exiscan merge. However, to be on the safe side, I
1446 have made the code more robust (and fixed the comments that describe what
1449 33. Some experimental protocols are using DNS PTR records for new purposes. The
1450 keys for these records are domain names, not reversed IP addresses. The
1451 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1452 leaves it alone. Component reversal etc. now happens only for IP addresses.
1455 34. The host_aton() function is supposed to be passed a string that is known
1456 to be a valid IP address. However, in the case of IPv6 addresses, it was
1457 not checking this. This is a hostage to fortune. Exim now panics and dies
1458 if the condition is not met. A case was found where this could be provoked
1459 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1460 components; fortuitously, this particular loophole had already been fixed
1461 by change 4.50/55 or 4.44/33 above.
1463 If there are any other similar loopholes, the new check in host_aton()
1464 itself should stop them being exploited. The report I received stated that
1465 data on the command line could provoke the exploit when Exim was running as
1466 exim, but did not say which command line option was involved. All I could
1467 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1468 running as the user.
1471 35. There was a buffer overflow vulnerability in the SPA authentication code
1472 (which came originally from the Samba project). I have added a test to the
1473 spa_base64_to_bits() function which I hope fixes it.
1476 36. The daemon start-up calls getloadavg() while still root for those OS that
1477 need the first call to be done as root, but it missed one case: when
1478 deliver_queue_load_max is set with deliver_drop_privilege. This is
1479 necessary for the benefit of the queue runner, because there is no re-exec
1480 when deliver_drop_privilege is set.
1482 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1483 was in use, was not putting the data itself into the right store pool;
1484 consequently, it could be overwritten for a subsequent message in the same
1485 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1488 38. Sometimes the final signoff response after QUIT could fail to get
1489 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1490 < 0 before doing a fflush(). This bug looks as though it goes back to the
1491 introduction of TLS in release 3.20, but "sometimes" must have been rare
1492 because the tests only now provoked it.
1498 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1499 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1500 Now it returns an integer. A number of calls were still expecting a T/F
1501 return. Fortuitously, in all cases, the tests worked in OK situations,
1502 which is the norm. However, things would have gone wrong on any write
1503 failures on the smtp file descriptor. This function is used when sending
1504 messages over SMTP and also when doing verify callouts.
1506 2. When Exim is called to do synchronous delivery of a locally submitted
1507 message (the -odf or -odi options), it no longer closes stderr before doing
1510 3. Implemented the mua_wrapper option.
1512 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1514 5. Implemented the functions header_remove(), header_testname(),
1515 header_add_at_position(), and receive_remove_recipient(), and exported them
1518 6. If an ACL "warn" statement specified the addition of headers, Exim already
1519 inserted X-ACL-Warn: at the start if there was no header name. However, it
1520 was not making this test for the second and subsequent header lines if
1521 there were newlines in the string. This meant that an invalid header could
1522 be inserted if Exim was badly configured.
1524 7. Allow an ACL "warn" statement to add header lines at the start or after all
1525 the Received: headers, as well as at the end.
1527 8. Added the rcpt_4xx retry error code.
1529 9. Added postmaster_mailfrom=xxx to callout verification option.
1531 10. Added mailfrom=xxxx to the callout verification option, for verify=
1534 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1535 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1537 12. Inserted some casts to stop certain compilers warning when using pointer
1538 differences as field lengths or precisions in printf-type calls (mostly
1539 affecting debugging statements).
1541 13. Added optional readline() support for -be (dynamically loaded).
1543 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1544 same clock tick as a message's arrival, so that its received time was the
1545 same as the "first fail" time on the retry record, and that message
1546 remained on the queue past the ultimate address timeout, every queue runner
1547 would try a delivery (because it was past the ultimate address timeout) but
1548 after another failure, the ultimate address timeout, which should have then
1549 bounced the address, did not kick in. This was a "< instead of <=" error;
1550 in most cases the first failure would have been in the next clock tick
1551 after the received time, and all would be well.
1553 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1554 being recognized when the domain list was tested by the match_domain
1555 condition in an expansion string.
1557 16. Added the ${str2b64: operator.
1559 17. Exim was always calling setrlimit() to set a large limit for the number of
1560 processes, without checking whether the existing limit was already
1561 adequate. (It did check for the limit on file descriptors.) Furthermore,
1562 errors from getrlimit() and setrlimit() were being ignored. Now they are
1563 logged to the main and panic logs, but Exim does carry on, to try to do its
1564 job under whatever limits there are.
1566 18. Imported PCRE 5.0.
1568 19. Trivial typo in log message " temporarily refused connection" (the leading
1571 20. If the log selector return_path_on_delivery was set and an address was
1572 redirected to /dev/null, the delivery process crashed because it assumed
1573 that a return path would always be set for a "successful" delivery. In this
1574 case, the whole delivery is bypassed as an optimization, and therefore no
1577 21. Internal re-arrangement: the function for sending a challenge and reading
1578 a response while authentication was assuming a zero-terminated challenge
1579 string. It's now changed to take a pointer and a length, to allow for
1580 binary data in such strings.
1582 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1584 23. Exim was not respecting finduser_retries when seeking the login of the
1585 uid under which it was called; it was always trying 10 times. (The default
1586 setting of finduser_retries is zero.) Also, it was sleeping after the final
1587 failure, which is pointless.
1589 24. Implemented tls_on_connect_ports.
1591 25. Implemented acl_smtp_predata.
1593 26. If the domain in control=submission is set empty, Exim assumes that the
1594 authenticated id is a complete email address when it generates From: or
1595 Sender: header lines.
1597 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1598 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1599 chown and chgrp in /bin and hostname in /usr/bin.
1601 28. Exim was keeping the "process log" file open after each use, just as it
1602 does for the main log. This opens the possibility of it remaining open for
1603 long periods when the USR1 signal hits a daemon. Occasional processlog
1604 errors were reported, that could have been caused by this. Anyway, it seems
1605 much more sensible not to leave this file open at all, so that is what now
1608 29. The long-running daemon process does not normally write to the log once it
1609 has entered its main loop, and it closes the log before doing so. This is
1610 so that log files can straightforwardly be renamed and moved. However,
1611 there are a couple of unusual error situations where the daemon does write
1612 log entries, and I had neglected to close the log afterwards.
1614 30. The text of an SMTP error response that was received during a remote
1615 delivery was being truncated at 512 bytes. This is too short for some of
1616 the long messages that one sometimes sees. I've increased the limit to
1619 31. It is now possible to make retry rules that apply only when a message has a
1620 specific sender, in particular, an empty sender.
1622 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1623 it possible to be selective about when SMTP synchronization is enforced.
1625 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1627 32. Implemented hosts_connection_nolog.
1629 33. Added an ACL for QUIT.
1631 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1634 35. Added mailbox_size and mailbox_filecount to appendfile.
1636 36. Added control = no_multiline_responses to ACLs.
1638 37. There was a bug in the logic of the code that waits for the clock to tick
1639 in the case where the clock went backwards by a substantial amount such
1640 that the microsecond fraction of "now" was more than the microsecond
1641 fraction of "then" (but the whole seconds number was less).
1643 38. Added support for the libradius Radius client library this is found on
1644 FreeBSD (previously only the radiusclient library was supported).
1650 1. When certain lookups returned multiple values in the form name=value, the
1651 quoting of the values was not always being done properly. Specifically:
1652 (a) If the value started with a double quote, but contained no whitespace,
1654 (b) If the value contained whitespace other than a space character (i.e.
1655 tabs or newlines or carriage returns) it was not quoted.
1656 This fix has been applied to the mysql and pgsql lookups by writing a
1657 separate quoting function and calling it from the lookup code. The fix
1658 should probably also be applied to nisplus, ibase and oracle lookups, but
1659 since I cannot test any of those, I have not disturbed their existing code.
1661 2. A hit in the callout cache for a specific address caused a log line with no
1662 reason for rejecting RCPT. Now it says "Previous (cached) callout
1663 verification failure".
1665 3. There was an off-by-one bug in the queryprogram router. An over-long
1666 return line was truncated at 256 instead of 255 characters, thereby
1667 overflowing its buffer with the terminating zero. As well as fixing this, I
1668 have increased the buffer size to 1024 (and made a note to document this).
1670 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1671 when Exim is waiting for an SMTP response from a remote server, Exim
1672 restarts its select() call on the socket, thereby resetting its timeout.
1673 This is not a problem when such interrupts are rare. Somebody set up a cron
1674 job to run exiwhat every 2 minutes, which is less than the normal select()
1675 timeout (5 or 10 minutes). This meant that the select() timeout never
1676 kicked in because it was always reset. I have fixed this by comparing the
1677 time when an interrupt arrives with the time at the start of the first call
1678 to select(). If more time than the timeout has elapsed, the interrupt is
1679 treated as a timeout.
1681 5. Some internal re-factoring in preparation for the addition of Sieve
1682 extensions (by MH). In particular, the "personal" test is moved to a
1683 separate function, and given an option for scanning Cc: and Bcc: (which is
1684 not set for Exim filters).
1686 6. When Exim created an email address using the login of the caller as the
1687 local part (e.g. when creating a From: or Sender: header line), it was not
1688 quoting the local part when it contained special characters such as @.
1690 7. Installed new OpenBSD configuration files.
1692 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1693 try to make them clearer.
1695 9. Callout options, other than the timeout value, were being ignored when
1696 verifying sender addresses in header lines. For example, when using
1698 verify = header_sender/callout=no_cache
1700 the cache was (incorrectly) being used.
1702 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1703 only the Cygwin environment.
1705 11. When return_path_on_delivery was set as a log selector, if different remote
1706 addresses in the same message used different return paths and parallel
1707 remote delivery occurred, the wrong values would sometimes be logged.
1708 (Whenever a remote delivery process finished, the return path value from
1709 the most recently started remote delivery process was logged.)
1711 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1712 lines when AUTH and/or TLS are in use. This is the "received protocol"
1713 field. Exim used to use "asmtp" for authenticated SMTP, without any
1714 indication (in the protocol name) for TLS use. Now it follows the RFC and
1715 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1716 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1717 names appear in log lines as well as in Received: header lines.
1719 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1720 extensions, and comparison tests, and to fix some bugs.
1722 14. Changes to the "personal" filter test:
1724 (1) The test was buggy in that it was just doing the equivalent of
1725 "contains" tests on header lines. For example, if a user's address was
1726 anne@some.where, the "personal" test would incorrectly be true for
1728 To: susanne@some.where
1730 This test is now done by extracting each address from the header in turn,
1731 and checking the entire address. Other tests that are part of "personal"
1732 are now done using regular expressions (for example, to check local parts
1733 of addresses in From: header lines).
1735 (2) The list of non-personal local parts in From: addresses has been
1736 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1737 taken from the Sieve specification recommendations.
1739 (3) If the message contains any header line starting with "List-" it is
1740 treated as non-personal.
1742 (4) The test for "circular" in the Subject: header line has been removed
1743 because it now seems ill-conceived.
1745 15. Minor typos in src/EDITME comments corrected.
1747 16. Installed latest exipick from John Jetmore.
1749 17. If headers_add on a router specified a text string that was too long for
1750 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1751 of string_sprintf() is now avoided.
1753 18. $message_body_size was not set (it was always zero) when running the DATA
1754 ACL and the local_scan() function.
1756 19. For the "mail" command in an Exim filter, no default was being set for
1757 the once_repeat time, causing a random time value to be used if "once" was
1758 specified. (If the value happened to be <= 0, no repeat happened.) The
1759 default is now 0s, meaning "never repeat". The "vacation" command was OK
1760 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1761 (I found it when inspecting the code).
1763 20. There is now an overall timeout for performing a callout verification. It
1764 defaults to 4 times the callout timeout, which applies to individual SMTP
1765 commands during the callout. The overall timeout applies when there is more
1766 than one host that can be tried. The timeout is checked before trying the
1767 next host. This prevents very long delays if there are a large number of
1768 hosts and all are timing out (e.g. when the network connections are timing
1769 out). The value of the overall timeout can be changed by specifying an
1770 additional sub-option for "callout", called "maxwait". For example:
1772 verify = sender/callout=5s,maxwait=20s
1774 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1775 to the end before writing, but this should make it even safer).
1777 22. Exim was forgetting that it had advertised PIPELINING for the second and
1778 subsequent messages on an SMTP connection. It was also not resetting its
1779 memory on STARTTLS and an internal HELO.
1781 23. When Exim logs an SMTP synchronization error within a session, it now
1782 records whether PIPELINING has been advertised or not.
1784 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1785 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1786 rather than long int.
1788 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1790 26. Added the never_mail option to autoreply.
1796 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1797 crash if the getsockname() call failed; this can happen if a connection is
1798 closed very soon after it is established. The problem was simply in the
1799 order in which certain operations were done, causing Exim to try to write
1800 to the SMTP stream before it had set up the file descriptor. The bug has
1801 been fixed by making things happen in the correct order.
1807 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1808 before the connection was closed, thus losing the rejection response.
1810 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1811 some early Solaris releases, but causes trouble in current releases where
1812 socklen_t is defined.
1814 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1817 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1820 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1821 that releasing the top of it at the end releases what was used for sub-
1822 expansions (unless the block got too big). However, discard this block if
1823 the first thing is a variable or header, so that we can use its block when
1824 it is dynamic (useful for very large $message_headers, for example).
1826 6. Lookups now cache *every* query, not just the most recent. A new, separate
1827 store pool is used for this. It can be recovered when all lookup caches are
1828 flushed. Lookups now release memory at the end of their result strings.
1829 This has involved some general refactoring of the lookup sources.
1831 7. Some code has been added to the store_xxx() functions to reduce the amount
1832 of flapping under certain conditions.
1834 8. log_incoming_interface used to affect only the <= reception log lines. Now
1835 it causes the local interface and port to be added to several more SMTP log
1836 lines, for example "SMTP connection from", and rejection lines.
1838 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1840 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1842 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1843 could be overwritten at the end of the current message (or the start of a
1844 new message if it was set in a HELO ACL). The value is now preserved for
1845 the duration of the SMTP connection.
1847 12. If a transport had a headers_rewrite setting, and a matching header line
1848 contained an unqualified address, that address was qualified, even if it
1849 did not match any rewriting rules. The underlying bug was that the values
1850 of the flags that permit the existence of unqualified sender and recipient
1851 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1852 non-local messages, and by -bnq for local messages) were not being
1853 preserved with the message after it was received.
1855 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1856 "next input=" as part of the text comprising the host identity instead of
1857 the correct text. The code was using the same buffer for two different
1858 strings. However, depending on which order the printing function evaluated
1859 its arguments, the bug did not always show up. Under Linux, for example, my
1860 test suite worked just fine.
1862 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1863 doesn't work with some older versions of Perl. It has been changed to "my",
1864 which in any case is probably the better facility to use.
1866 15. A really picky compiler found some instances of statements for creating
1867 error messages that either had too many or two few arguments for the format
1870 16. The size of the buffer for calls to the DNS resolver has been increased
1871 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1872 for addresses that have a lot of PTR records. This alleviates a problem; it
1873 does not fully solve it.
1875 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1876 buffer now truncates the list and logs the incident, which is the same
1877 action as happens when Exim is looking up a host name and its aliases.
1878 Previously in this situation something unpredictable would happen;
1879 sometimes it was "internal error: store_reset failed".
1881 18. If a server dropped the connection unexpectedly when an Exim client was
1882 using GnuTLS and trying to read a response, the client delivery process
1883 crashed while trying to generate an error log message.
1885 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1886 string, for example:
1888 warn message = H1: something\nH2: something
1890 the text was added as a single header line from Exim's point of view
1891 though it ended up OK in the delivered message. However, searching for the
1892 second and subsequent header lines using $h_h2: did not work. This has been
1893 fixed. Similarly, if a system filter added multiple headers in this way,
1894 the routers could not see them.
1896 20. Expanded the error message when iplsearch is called with an invalid key to
1897 suggest using net-iplsearch in a host list.
1899 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1900 ACLs are no longer actually imposed (and a message to that effect is
1903 22. If a "gecos" field in a passwd entry contained escaped characters, in
1904 particular, if it contained a \" sequence, Exim got it wrong when building
1905 a From: or a Sender: header from that name. A second bug also caused
1906 incorrect handling when an unquoted " was present following a character
1907 that needed quoting.
1909 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1910 was not being matched caselessly.
1912 24. Arranged for all hyphens in the exim.8 source to be escaped with
1915 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1916 properly. Recipient callout cache records were still being keyed to include
1917 the sender, even when use_sender was set false. This led to far more
1918 callouts that were necessary. The sender is no longer included in the key
1919 when use_sender is false.
1921 26. Added "control = submission" modifier to ACLs.
1923 27. Added the ${base62d: operator to decode base 62 numbers.
1925 28. dnsdb lookups can now access SRV records.
1927 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1928 the configuration file.
1930 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1931 (-v) mode. This makes the output for a verbose queue run more intelligible.
1933 31. Added a use_postmaster feature to recipient callouts.
1935 32. Added the $body_zerocount variable, containing the number of binary zero
1936 bytes in the message body.
1938 33. The time of last modification of the "new" subdirectory is now used as the
1939 "mailbox time last read" when there is a quota error for a maildir
1942 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1944 35. Added +ignore_unknown as a special item in host lists.
1946 36. Code for decoding IPv6 addresses in host lists is now included, even if
1947 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1948 address was recognized as an IP address, but was then not correctly decoded
1949 into binary, causing unexpected and incorrect effects when compared with
1956 1. Very minor rewording of debugging text in manualroute to say "list of
1957 hosts" instead of "hostlist".
1959 2. If verify=header_syntax was set, and a header line with an unqualified
1960 address (no domain) and a large number of spaces between the end of the
1961 name and the colon was received, the reception process suffered a buffer
1962 overflow, and (when I tested it) crashed. This was caused by some obsolete
1963 code that should have been removed. The fix is to remove it!
1965 3. When running in the test harness, delay a bit after writing a bounce
1966 message to get a bit more predictability in the log output.
1968 4. Added a call to search_tidyup() just before forking a reception process. In
1969 theory, someone could use a lookup in the expansion of smtp_accept_max_
1970 per_host which, without the tidyup, could leave open a database connection.
1972 5. Added the variables $recipient_data and $sender_data which get set from a
1973 lookup success in an ACL "recipients" or "senders" condition, or a router
1974 "senders" option, similar to $domain_data and $local_part_data.
1976 6. Moved the writing of debug_print from before to after the "senders" test
1979 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1980 problems for message scanning, either using a data ACL, or using
1981 local_scan() because the Received: header was not generated till after they
1982 were called (in order to set the time as the time of reception completion).
1983 I have revised the way this works. The header is now generated after the
1984 body is received, but before the ACL or local_scan() are called. After they
1985 are run, the timestamp in the header is updated.
1991 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1992 before starting a queue runner without re-exec. This happened only when
1993 deliver_drop_privilege was set or when the Exim user was set to root. The
1994 effect of the bug was that timeouts during subsequent deliveries caused
1995 crashes instead of being properly handled. The handler is now left at its
1996 default (and expected) setting.
1998 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1999 message, again when deliver_drop_privilege is set or Exim is run as root.
2000 The bug described in (1) was not present in this case, but the tidying up
2001 of the other signals was missing. I have made the two cases consistent.
2003 3. The ignore_target_hosts setting on a manualroute router was being ignored
2004 for hosts that were looked up using the /MX notation.
2006 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2009 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2010 operated on the sender address. After changing the $sender_address to <>
2011 for the sender address verify, Exim was re-instated it as the original
2012 (before rewriting) address, but remembering that it had rewritten it, so it
2013 wasn't rewriting it again. This bug also had the effect of breaking the
2014 sender address verification caching when the sender address was rewritten.
2016 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2017 This has been changed so that if the ip literal address matches
2018 ignore_target_hosts, the router declines.
2020 7. Added expansion conditions match_domain, match_address, and match_local_
2021 part (NOT match_host).
2023 8. The placeholder for the Received: header didn't have a length field set.
2025 9. Added code to Exim itself and to exim_lock to test for a specific race
2026 condition that could lead to file corruption when using MBX delivery. The
2027 issue is with the lockfile that is created in /tmp. If this file is removed
2028 after a process has opened it but before that process has acquired a lock,
2029 there is the potential for a second process to recreate the file and also
2030 acquire a lock. This could lead to two Exim processes writing to the file
2031 at the same time. The added code performs the same test as UW imapd; it
2032 checks after acquiring the lock that its file descriptor still refers to
2033 the same named file.
2035 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2036 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2039 11. Added the smtp_active_hostname option. If used, this will typically be made
2040 to depend on the incoming interface address. Because $interface_address is
2041 not set up until the daemon has forked a reception process, error responses
2042 that can happen earlier (such as "too many connections") no longer contain
2045 12. If an expansion in a condition on a "warn" statement fails because a lookup
2046 defers, the "warn" statement is abandoned, and the next ACL statement is
2047 processed. Previously this caused the whole ACL to be aborted.
2049 13. Added the iplsearch lookup type.
2051 14. Added ident_timeout as a log selector.
2053 15. Added tls_certificate_verified as a log selector.
2055 16. Added a global option tls_require_ciphers (compare the smtp transport
2056 option of the same name). This controls incoming TLS connections.
2058 17. I finally figured out how to make tls_require_ciphers do a similar thing
2059 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2060 before starting the TLS session.
2062 18. Tabs are now shown as \t in -bP output.
2064 19. If the log selector return_path_on_delivery was set, Exim crashed when
2065 bouncing a message because it had too many Received: header lines.
2067 20. If two routers both had headers_remove settings, and the first one included
2068 a superfluous trailing colon, the final name in the first list and the
2069 first name in the second list were incorrectly joined into one item (with a
2070 colon in the middle).
2076 1. Added -C and -D options to the exinext utility, mainly to make it easier
2077 to include in the automated testing, but these could be helpful when
2078 multiple configurations are in use.
2080 2. The exinext utility was not formatting the output nicely when there was
2081 an alternate port involved in the retry record key, nor when there was a
2082 message id as well (for retries that were specific to a specific message
2083 and a specific host). It was also confused by IPv6 addresses, because of
2084 the additional colons they contain. I have fixed the IPv4 problem, and
2085 patched it up to do a reasonable job for IPv6.
2087 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2088 delivery, the log line now contains "pipelined" if PIPELINING was used.
2090 4. An SMTP transport process used to panic and die if the bind() call to set
2091 an explicit outgoing interface failed. This has been changed; it is now
2092 treated in the same way as a connect() failure.
2094 5. A reference to $sender_host_name in the part of a conditional expansion
2095 that was being skipped was still causing a DNS lookup. This no longer
2098 6. The def: expansion condition was not recognizing references to header lines
2099 that used bh_ and bheader_.
2101 7. Added the _cache feature to named lists.
2103 8. The code for checking quota_filecount in the appendfile transport was
2104 allowing one more file than it should have been.
2106 9. For compatibility with Sendmail, the command line option
2114 and sets the incoming protocol and host name (for trusted callers). The
2115 host name and its colon can be omitted when only the protocol is to be set.
2116 Note the Exim already has two private options, -pd and -ps, that refer to
2117 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2118 "s", but I don't think that's a major issue.
2120 10. A number of refactoring changes to the code, none of which should affect
2123 (a) The number of logging options was getting close to filling up the
2124 32-bit word that was used as a bit map. I have split them into two classes:
2125 those that are passed in the argument to log_write(), and those that are
2126 only ever tested independently outside of that function. These are now in
2127 separate 32-bit words, so there is plenty of room for expansion again.
2128 There is no change in the user interface or the logging behaviour.
2130 (b) When building, for example, log lines, the code previously used a
2131 macro that called string_cat() twice, in order to add two strings. This is
2132 not really sufficiently general. Furthermore, there was one instance where
2133 it was actually wrong because one of the argument was used twice, and in
2134 one call a function was used. (As it happened, calling the function twice
2135 did not affect the overall behaviour.) The macro has been replaced by a
2136 function that can join an arbitrary number of extra strings onto a growing
2139 (c) The code for expansion conditions now uses a table and a binary chop
2140 instead of a serial search (which was left over from when there were very
2141 few conditions). Also, it now recognizes conditions like "pam" even when
2142 the relevant support is not compiled in: a suitably worded error message is
2143 given if an attempt is made to use such a condition.
2145 11. Added ${time_interval:xxxxx}.
2147 12. A bug was causing one of the ddress fields not to be passed back correctly
2148 from remote delivery subprocesses. The field in question was not being
2149 subsequently used, so this caused to problems in practice.
2151 13. Added new log selectors queue_time and deliver_time.
2153 14. Might have fixed a bug in maildirsizefile handling that threw up
2154 "unexpected character" debug warnings, and recalculated the data
2155 unnecessarily. In any case, I expanded the warning message to give more
2158 15. Added the message "Restricted characters in address" to the statements in
2159 the default ACL that block characters like @ and % in local parts.
2161 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2162 Three changes have been made:
2164 (a) There was a serious bug; a negative response to MAIL caused the whole
2165 recipient domain to be cached as invalid, thereby blocking all messages
2166 to all local parts at the same domain, from all senders. This bug has
2167 been fixed. The domain is no longer cached after a negative response to
2168 MAIL if the sender used is not empty.
2170 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2173 (c) A new callout option, "use_sender" has been added for people who want
2174 the modified behaviour.
2180 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2183 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2184 indeed breaks things for older releases.
2186 3. Added additional logging to the case where there is a problem reading data
2187 from a filter that is running in a subprocess using a pipe, in order to
2188 try to track down a specific problem.
2190 4. Testing facility fudge: when running in the test harness and attempting
2191 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2192 getting "No route to host". Convert this to a timeout.
2194 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2197 6. Some OS don't have socklen_t but use size_t instead. This affects the
2198 fifth argument of getsockopt() amongst other things. This is now
2199 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2200 can be set for individual OS. I have set it for SunOS5, OSF1, and
2201 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2202 some earlier ones do not.
2204 7. Change 4.30/15 was not doing the test caselessly.
2206 8. The standard form for an IPv6 address literal was being rejected by address
2207 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2208 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2209 this, as well as the form without the "IPv6" on the front (but only when
2210 address literals are enabled, of course).
2212 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2214 10. Exim crashed if a message with an empty sender address specified by -f
2215 encountered a router with an errors_to setting. This could be provoked only
2216 by a command such as
2220 where an empty string was supplied; "<>" did not hit this bug.
2222 11. Installed PCRE release 4.5.
2224 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2225 remained set. It is now erased.
2227 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2228 times from message ids (which are base 36 rather than the normal 62).
2230 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2231 were being counted as actual protocol errors, and logged if the log
2232 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2233 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2234 and DATA following a set of rejected RCPTs do not count as protocol errors.
2235 In other words, Exim assumes they were pipelined, though this may not
2236 actually be the case. Of course, in all cases the client gets an
2237 appropriate error code.
2239 15. If a lookup fails in an ACL condition, a message about the failure may
2240 be available; it is used if testing the ACL cannot continue, because most
2241 such messages specify what the cause of the deferral is. However, some
2242 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2243 that caused an old message to be retained and used if a later statement
2244 caused a defer, replacing the real cause of the deferral.
2246 16. If an IP address had so many PTR records that the DNS lookup buffer
2247 was not large enough to hold them, Exim could crash while trying to process
2248 the truncated data. It now detects and logs this case.
2250 17. Further to 4.21/58, another change has been made: if (and only if) the
2251 first line of a message (the first header line) ends with CRLF, a bare LF
2252 in a subsequent header line has a space inserted after it, so as not to
2253 terminate the header.
2255 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2256 unnecessarily, used atoi() instead of strtol(), and didn't check the
2257 termination when getting file sizes from file names by regex.
2259 19. Completely re-implemented the support for maildirsize files, in the light
2260 of a number of problems with the previous contributed implementation
2261 (4.30/29). In particular:
2263 . If the quota is zero, the maildirsize file is maintained, but no quota is
2266 . If the maildir directory does not exist, it is created before any attempt
2267 to write a maildirsize file.
2269 . The quota value in the file is just a cache; if the quota is changed in
2270 the transport, the new value overrides.
2272 . A regular expression is available for excluding directories from the
2275 20. The autoreply transport checks the characters in options that define the
2276 message's headers; it allows continued headers, but it was checking with
2277 isspace() after an embedded newline instead of explicitly looking for a
2280 21. If all the "regular" hosts to which an address was routed had passed their
2281 expiry times, and had not reached their retry times, the address was
2282 bounced, even if fallback hosts were defined. Now Exim should go on to try
2285 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2286 equivalent code in the SMTP transport. Some hosts send humungous responses
2287 to HELO/EHLO, more than 1024 it seems.
2289 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2290 gives compiler warnings in some environments. I've now done it "properly",
2293 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2294 (because of problems with the built-in one) was declared to return uschar *
2295 instead of char *, causing compiler failure.
2297 25. Fixed a file descriptor leak when processing alias/forward files.
2299 26. Fixed a minor format string issue in dbfn.c.
2301 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2303 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2304 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2307 29. When a sender address is verified, it is cached, to save repeating the test
2308 when there is more than one recipient in a message. However, when the
2309 verification involves a callout, it is possible for different callout
2310 options to be set for different recipients. It is too complicated to keep
2311 track of this in the cache, so now Exim always runs a verification when a
2312 callout is required, relying on the callout cache for the optimization.
2313 The overhead is duplication of the address routing, but this should not be
2316 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2317 to be verified with callout=postmaster, and the main callout worked but the
2318 postmaster check failed, the verification correctly failed. However, if a
2319 subsequent RCPT command asked for sender verification *without* the
2320 postmaster check, incorrect caching caused this verification also to fail,
2323 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2324 it was not caching the DNS options (qualify_single, search_parents) that
2325 were used when the lookup failed. A subsequent lookup with different
2326 options therefore always gave the same answer, though there were cases
2327 where it should not have. (Example: a "domains = !$mx_any" option on a
2328 dnslookup router: the "domains" option is always processed without any
2329 widening, but the router might have qualify_single set.) Now Exim uses the
2330 cached value only when the same options are set.
2332 32. Added John Jetmore's "exipick" utility to the distribution.
2334 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2335 than a timeout (e.g. a certificate is required, and is not provided), an
2336 Exim server now closes the connection immediately. Previously it waited for
2337 the client to close - but if the client is SSL, it seems that they each
2338 wait for each other, leading to a delay before one of them times out.
2340 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2341 maintained 0.8.x compatibility because I don't think many are using it, and
2342 it is clearly obsolete.
2344 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2347 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2348 client certificate was expired. A simple patch fixes this, though I don't
2349 understand the full logic of why the verify callback is called multiple
2352 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2353 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2354 which causes problems with some clients (such as the Certicom SSL Plus
2355 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2356 disables the coutermeasure allowing Eudora to connect."
2358 38. Exim was not checking that a write() to a log file succeeded. This could
2359 lead to Bad Things if a log got too big, in particular if it hit a file
2360 size limit. Exim now panics and dies if it cannot write to a log file, just
2361 as it does if it cannot open a log file.
2363 39. Modified OS/Makefile-Linux so that it now contains
2365 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2367 The two -D definitions ensure that Exim is compiled with large file
2368 support, which makes it possible to handle log files that are bigger than
2371 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2372 instance) a domain was checked against a named list that involved a lookup,
2373 causing $domain_data to be set, then another domain was checked against the
2374 same list, then the first domain was re-checked, the value of $domain_data
2375 after the final check could be wrong. In particular, if the second check
2376 failed, it could be set empty. This bug probably also applied to
2379 41. The strip_trailing_dot option was not being applied to the address given
2380 with the -f command-line option.
2382 42. The code for reading a message's header from the spool was incrementing
2383 $received_count, but never initializing it. This meant that the value was
2384 incorrect (doubled) while delivering a message in the same process in which
2385 it was received. In the most common configuration of Exim, this never
2386 happens - a fresh exec is done - but it can happen when
2387 deliver_drop_privilege is set.
2389 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2390 it now includes up to 150 characters of the unexpected data in the log
2393 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2394 and building data strings. The size of both of these buffers was 10 000
2395 bytes - far larger than anybody would *ever* want, thought I. Needless to
2396 say, somebody hit the limit. I have increased the maximum line length to
2397 20 000 and the maximum data length of concatenated lines to 100 000. I have
2398 also fixed two bugs, because there was no checking on these buffers. Tsk,
2399 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2400 buffer is too small.
2402 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2403 lsearch lookups. Now it does.
2405 46. When parsing a route_list item in a manualroute router, a fixed-length
2406 buffer was used for the list of hosts. I made this 1024 bytes long,
2407 thinking that nobody would ever have a list of hosts that long. Wrong.
2408 Somebody had a whole pile of complicated expansion conditions, and the
2409 string was silently truncated, leading to an expansion error. It turns out
2410 that it is easier to change to an unlimited length (owing to other changes
2411 that have happened since this code was originally written) than to build
2412 structure for giving a limitation error. The length of the item that
2413 expands into the list of hosts is now unlimited.
2415 47. The lsearch lookup could not handle data where the length of text line was
2416 more than 4095 characters. Such lines were truncated, leading to shortened
2417 data being returned. It should now handle lines of any length.
2419 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2420 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2423 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2424 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2425 no longer happens in comment lines. A list of replacements is now placed
2426 at the head of all of the source files, except those whose only change is
2427 to replace PERL_COMMAND in the very first #! line.
2429 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2430 messages on the queue, with a bottom-up merge sort, using code contributed
2431 by Michael Haardt. This should make operations like -bp somewhat faster on
2432 large queues. It won't affect queue runners, except when queue_run_in_order
2435 51. Installed eximstats 1.31 in the distribution.
2437 52. Added support for SRV lookups to the dnslookup router.
2439 53. If an ACL referred to $message_body or $message_body_end, the value was not
2440 reset for any messages that followed in the same SMTP session.
2442 54. The store-handling optimization for building very long strings was not
2443 differentiating between the different store pools. I don't think this
2444 actually made any difference in practice, but I've tidied it.
2446 55. While running the routers to verify a sender address, $sender_address
2447 was still set to the sender address. This is wrong, because when routing to
2448 send a bounce to the sender, it would be empty. Therefore, I have changed
2449 it so that, while verifying a sender address, $sender_address is set to <>.
2450 (There is no change to what happens when verifying a recipient address.)
2452 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2453 target A or AAAA records (if not already returned) without resetting the
2454 qualify_single or search_parents options of the DNS resolver. These are
2455 inappropriate in this case because the targets of MX and SRV records must
2456 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2457 target that, when qualified, matched something in the local domain. These
2458 two options are now turned off when doing these lookups.
2460 57. It seems that at least some releases of Reiserfs (which does not have the
2461 concept of a fixed number of inodes) returns zero and not -1 for the
2462 number of available inodes. This interacted badly with check_spool_inodes,
2463 which assumed that -1 was the "no such thing" setting. What I have done is
2464 to check that the total number of inodes is greater than zero before doing
2465 the test of how many are available.
2467 58. When a "warn" ACL statement has a log_message modifier, the message is
2468 remembered, and not repeated. This is to avoid a lot of repetition when a
2469 message has many recipients that cause the same warning to be written.
2470 Howewer, Exim was preserving the list of already written lines for an
2471 entire SMTP session, which doesn't seem right. The memory is now reset if a
2472 new message is started.
2474 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2475 debugging output unless log_rewrite was also set.
2477 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2478 of (int)(handle) when we know that handle contains (void *)(-1).
2480 61. The Exim daemon panic-logs an error return when it closes the incoming
2481 connection. However "connection reset by peer" seems to be common, and
2482 isn't really an error worthy of noting specially, so that particular error
2485 62. When Exim is trying to find all the local interfaces, it used to panic and
2486 die if the ioctl to get the interface flags failed. However, it seems that
2487 on at least one OS (Solaris 9) it is possible to have an interface that is
2488 included in the list of interfaces, but for which you get a failure error
2489 for this call. This happens when the interface is not "plumbed" into a
2490 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2491 failure of the "get flags" call assumes that the interface is down.
2493 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2494 makes life easier for people who are doing arithmetic on fields extracted
2495 from dates, where you often get leading zeros that should not be
2496 interpreted as octal.
2498 64. Added qualify_domain to the redirect router, to override the global
2501 65. If a pathologically long header line contained very many addresses (the
2502 report of this problem mentioned 10 000) and each of them was rewritten,
2503 Exim could use up a very large amount of memory. (It kept on making new
2504 copies of the header line as it rewrote, and never released the old ones.)
2505 At the expense of a bit more processing, the header rewriting function has
2506 been changed so that it no longer eats memory in this way.
2508 66. The generation of the Received: header has been moved from the time that a
2509 message starts to be received, to the time that it finishes. The timestamp
2510 in the Received: header should now be very close to that of the <= log
2511 line. There are two side-effects of this change:
2513 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2514 logged header lines no longer include the local Received: line, because
2515 it has not yet been created. The same applies to a copy of the message
2516 that is returned to a non-SMTP sender when a message is rejected.
2518 (b) When a filter file is tested using -bf, no additional Received: header
2519 is added to the test message. After some thought, I decided that this
2522 This change does not affect the value of $received_for. It is still set
2523 after address rewriting, but before local_scan() is called.
2525 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2527 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2528 gave an unhelpful panic error message, and a defer error. I have managed to
2529 change this behaviour so that it now rejects any supplied certificate,
2530 which seems right, as the list of acceptable certificates is empty.
2532 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2533 gave an unhelpful defer error. I have not managed to make this reject any
2534 supplied certificates, but the error message it gives is "no certificate
2535 supplied", which is not helpful.
2537 70. exigrep's output now also includes lines that are not associated with any
2538 message, but which match the given pattern. Implemented by a patch from
2539 Martin Sluka, which also tidied up the Perl a bit.
2541 71. Recipient callout verification, like sender verification, was using <> in
2542 the MAIL FROM command. This isn't really the right thing, since the actual
2543 sender may affect whether the remote host accepts the recipient or not. I
2544 have changed it to use the actual sender in the callout; this means that
2545 the cache record is now keyed on a recipient/sender pair, not just the
2546 recipient address. There doesn't seem to be a real danger of callout loops,
2547 since a callout by the remote host to check the sender would use <>.
2548 [SEE ABOVE: changed after hitting problems.]
2550 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2551 temporary errors. However, in the case of such a code being given after
2552 the end of a data transmission (i.e. after ".") Exim was failing to write
2553 a retry record for the message. (Yes, there was some broken host that was
2554 actually sending 8xx at this point.)
2556 73. An unknown lookup type in a host list could cause Exim to panic-die when
2557 the list was checked. (An example that provoked this was putting <; in the
2558 middle of a list instead of at the start.) If this happened during a DATA
2559 ACL check, a -D file could be left lying around. This kind of configuration
2560 error no longer causes Exim to die; instead it causes a defer errror. The
2561 incident is still logged to the main and panic logs.
2563 74. Buglet left over from Exim 3 conversion. The message "too many messages
2564 in one connection" was written to the rejectlog but not the mainlog, except
2565 when address rewriting (yes!) was being logged.
2567 75. Added write_rejectlog option.
2569 76. When a system filter was run not as root (that is, when system_filter_user
2570 was set), the values of the $n variables were not being returned to the
2571 main process; thus, they were not subsequently available in the $sn
2574 77. Added +return_path_on_delivery log selector.
2576 78. A connection timeout was being treated differently from recipients deferred
2577 when testing hosts_max_try with a message that was older than the host's
2578 retry timeout. (The host should not be counted, thus allowing all hosts to
2579 be tried at least once before bouncing.) This may have been the cause of an
2580 occasionally reported bug whereby a message would remain on the queue
2581 longer than the retry timeout, but would be bounced if a delivery was
2582 forced. I say "may" because I never totally pinned down the problem;
2583 setting up timeout/retry tests is difficult. See also the next item.
2585 79. The ultimate address timeout was not being applied to errors that involved
2586 a combination of host plus message (for example, a timeout on a MAIL
2587 command). When an address resolved to a number of possible hosts, and they
2588 were not all tried for each delivery (e.g. because of hosts_max_try), a
2589 message could remain on the queue longer than the retry timeout.
2591 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2594 81. Fixed an obscure SMTP outgoing bug which required at least the following
2595 conditions: (a) there was another message waiting for the same server;
2596 (b) the server returned 5xx to all RCPT commands in the first message so
2597 that the message was not completed; (c) the server dropped the connection
2598 or gave a negative response to the RSET that Exim sends to abort the
2599 transaction. The observed case was a dropped connection after DATA that had
2600 been sent in pipelining mode. That is, the server had advertised PIPELINING
2601 but was not implementing it correctly. The effect of the bug was incorrect
2602 behaviour, such as trying another host, and this could lead to a crash.
2608 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2609 and daemon.c were passed as pointers to ints; they should have been
2610 pointers to socklen_t variables (which are typically unsigned ints).
2612 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2615 3. Fixed a really odd bug that affected only the testing scheme; patching a
2616 certain fixed string in the binary changed the value of another string that
2617 happened to be identical to the end of the original first string.
2619 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2620 name", it returns that address as the IP address. On some operating
2621 systems (e.g. Solaris), it also passes back the IP address string as the
2622 "host name". However, on others (e.g. Linux), it passes back an empty
2623 string. Exim wasn't checking for this, and was changing the host name to an
2624 empty string, assuming it had been canonicized.
2626 5. Although rare, it is permitted to have more than one PTR record for a given
2627 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2628 all the names associated with an address, because they do in Solaris.
2629 However, it seems that they do not in Linux for data that comes from the
2630 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2631 I found this out when I moved to a new Linux workstation and tried to run
2632 the Exim test suite.
2634 To get round this problem I have changed the code so that it now does its
2635 own call to the DNS to look up PTR records when searching for a host name.
2636 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2637 addresses that are only in /etc/hosts are still found.
2639 This behaviour is, however, controlled by an option called host_lookup_
2640 order, which defaults to "bydns:byaddr". If people want to use the other
2641 order, or indeed, just use one or the other means of lookup, they can
2642 specify it in this variable.
2644 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2645 some operating systems, this comes back from gethostbyaddr() as an empty
2646 string, and this is what Exim used to test for. However, it seems that in
2647 other systems, "." is yielded. Exim now tests for this case too.
2649 7. The values of check_spool_space and check_log_space are now held internally
2650 as a number of kilobytes instead of an absolute number of bytes. If a
2651 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2652 kilobyte. This means that much larger values can be stored.
2654 8. Exim monitor: an attempt to get the action menu when not actually pointing
2655 at a message produces an empty menu entitled "No message selected". This
2656 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2657 no entries in it ("Shell widget menu has zero width and/or height"). So I
2658 have added a single, blank menu entry in this case.
2660 9. Added ${quote_local_part.
2662 10. MIME decoding is now applied to the contents of Subject: header lines when
2665 11. Now that a reference to $sender_host_address automatically causes a reverse
2666 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2667 host lookup before query-style lookups in lists that might use this
2668 variable. This has therefore been abolished, and the "net-" prefix is no
2669 longer necessary for query-style lookups.
2671 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2672 have been a typo for LFLAGS, so it has been changed.
2674 13. The install script calls Exim with "-C /dev/null" in order to find the
2675 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2676 to be output. Howeve, since Exim outputs its version number before the
2677 error, it didn't break the script. It just looked ugly. I fixed this by
2678 always allowing "-C /dev/null" if the caller is root.
2680 14. Ignore overlarge ACL variable number when reading spool file - insurance
2681 against a later release with more variables having written the file.
2683 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2684 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2685 this, as well as the form without the "IPv6" on the front.
2687 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2688 OS/Makefile-Darwin file.
2690 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2691 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2693 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2694 variable. However, some OS use ints for inodes. I've added cast to long int
2695 to get rid of the compiler warning.
2697 19. I had forgotten to lock out "/../" in configuration file names when
2698 ALT_CONFIG_PREFIX was set.
2700 20. Routers used for verification do not need to specify transports. However,
2701 if such a router generated a host list, and callout was configured, Exim
2702 crashed, because it could not find a port number from the (non-existent)
2703 transport. It now assumes port 25 in this circumstance.
2705 21. Added the -t option to exigrep.
2707 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2708 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2709 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2710 them is set, LOOKUP_LSEARCH is forced.
2712 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2714 24. Added sender and host information to the "rejected by local_scan()" log
2715 line; previously there was no indication of these.
2717 25. Added .include_if_exists.
2719 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2720 Linux. It turns out that not all file systems support this. Apparently some
2721 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2722 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2723 not supported on the file descriptor, is now ignored when Exim is trying to
2724 sync a directory. This applies only to Linux.
2726 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2728 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2729 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2732 29. Added support for maildirsize files from supplied patch (modified a bit).
2734 30. The use of :fail: followed by an empty string could lead Exim to respond to
2735 sender verification failures with (e.g.):
2737 550 Verification failed for <xxx>
2738 550 Sender verify failed
2740 where the first response line was missing the '-' that indicates it is not
2741 the final line of the response.
2743 31. The loop for finding the name of the user that called Exim had a hardwired
2744 limit of 10; it now uses the value of finduser_retries, which is used for
2745 all other user lookups.
2747 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2750 33. Exim was neglecting to zero errno before one call of strtol() when
2751 expanding a string and expecting an integer value. On some systems this
2752 resulted in spurious "integer overflow" errors. Also, it was casting the
2753 result into an int without checking.
2755 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2756 did not work. The code looks as if it has *never* worked, though it appears
2757 to have been documented since at least releast 1.62. I have made it work.
2759 35. The "timeout_DNS" error in retry rules, also documented since at least
2760 1.62, also never worked. As it isn't clear exactly what this means, and
2761 clearly it isn't a major issue, I have abolished the feature by treating it
2762 as "timeout", and writing a warning to the main and panic logs.
2764 36. The display of retry rules for -brt wasn't always showing the error code
2767 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2768 timeout_connect_A, timeout_connect_MX.
2770 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2771 to the empty sender.
2773 39. The daemon was not analysing the content of -oX till after it had closed
2774 stderr and disconnected from the controlling terminal. This meant that any
2775 syntax errors were only noted on the panic log, and the return code from
2776 the command was 0. By re-arranging the code a little, I've made the
2777 decoding happen first, so such errors now appear on stderr, and the return
2778 code is 1. However, the actual setting up of the sockets still happens in
2779 the disconnected process, so errors there are still only recorded on the
2782 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2783 connections (as happens on some IP stacks) was logged at start up time as
2784 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2785 it from "IPv6 and IPv4", which means that two separate sockets are being
2788 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2789 says whether AF_INET or AF_INET6 was passed as an argument.
2791 42. Exiwhat output was messed up when time zones were included in log
2794 43. Exiwhat now gives more information about the daemon's listening ports,
2795 and whether -tls-on-connect was used.
2797 44. The "port" option of the smtp transport is now expanded.
2799 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2800 silently ignored. Now an error message is written to the main and panic
2803 46. There's a new ACL modifier called "logwrite" which writes to a log file
2804 as soon as it is encountered.
2806 47. Added $local_user_uid and $local_user_gid at routing time.
2808 48. Exim crashed when trying to verify a sender address that was being
2811 49. Exim was recognizing only a space character after ".include". It now also
2812 recognizes a tab character.
2814 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2815 extracting the relevant information from the specification. The man page no
2816 longer contains scrambled data for the -d option, and I've added a section
2817 at the front about calling Exim under different names.
2819 51. Added "extra_headers" argument to the "mail" command in filter files.
2821 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2824 53. Installed eximstats 1.29.
2826 54. Added transport_filter_timeout as a generic transport option.
2828 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2829 Cc: header lines. This was required by RFC 822, but it not required by RFC
2832 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2833 incoming messages that did not have them. Now it does so only if the
2834 message originates locally, that is, if there is no associated remote host
2835 address. When Resent- header lines are present, this applies to the Resent-
2836 lines rather than the non-Resent- lines.
2838 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2839 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2841 58. Messages for configuration errors now include the name of the main
2842 configuration file - useful now that there may be more than one file in a
2843 list (.included file names were always shown).
2845 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2846 for those rare installations that do not start the daemon as root or run it
2847 setuid root. I've cut out the call to initgroups() if the daemon is not
2850 60. The Exim user and group can now be bound into the binary as text strings
2851 that are looked up at the start of Exim's processing.
2853 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2855 62. Added $mailstore_basename variable.
2857 63. Installed patch to sieve.c from Michael Haardt.
2859 64. When Exim failed to open the panic log after failing to open the main log,
2860 the original message it was trying to log was written to stderr and debug
2861 output, but if they were not available (the usual case in production), it
2862 was lost. Now it is written to syslog before the two lines that record the
2863 failures to open the logs.
2865 65. Users' Exim filters run in subprocesses under the user's uid. It is
2866 possible for a "deliver" command or an alias in a "personal" command to
2867 provoke an address rewrite. If logging of address rewriting is configured,
2868 this fails because the process is not running as root or exim. There may be
2869 a better way of dealing with this, but for the moment (because 4.30 needs
2870 to be released), I have disabled address rewrite logging when running a
2871 filter in a non-root, non-exim process.
2877 1. The buildconfig auxiliary program wasn't quoting the value set for
2878 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2879 not defined. This bug was masked in 4.22 by the effect that was fixed in
2882 2. Some messages that were rejected after a message id was allocated were
2883 shown as "incomplete" by exigrep. It no longer does this for messages that
2884 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2886 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2887 have allow_domain_literals set, the ID did not get logged in the <= line.
2888 Domain literals are now always recognized in Message-ID: header lines.
2890 4. The first argument for a ${extract expansion item is the key name or field
2891 number. Leading and trailing spaces in this item were not being ignored,
2892 causing some misleading effects.
2894 5. When deliver_drop_privilege was set, single queue runner processes started
2895 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2896 same command in the process it spins off) were not dropping privilege.
2898 6. When the daemon running as "exim" started a queue runner, it always
2899 re-executed Exim in the spun-off process. This is a waste of effort when
2900 deliver_drop_privilege is set. The new process now just calls the
2901 queue-runner function directly.
2907 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2910 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2911 ignored. Though the use of -oP was forcing the writing of a pid file, it
2912 was always written to the default place.
2914 3. If the message "no IP address found for host xxxx" is generated during
2915 incoming verification, it is now followed by identification of the incoming
2916 connection (so you can more easily find what provoked it).
2918 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2920 5. Added some features to "harden" Exim a bit more against certain attacks:
2922 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2923 be put in Local/Makefile. This is like the never_users runtime option,
2924 but it cannot be overridden. The default setting is "root".
2926 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2927 prefix string with which any file named in a -C command line option
2930 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2931 is retained for -C and -D only if the caller of Exim is root. Without
2932 it, the exim user may also use -C and -D and retain privilege.
2934 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2935 command line option is disabled.
2937 6. Macro names set by the -D option must start with an upper case letter, just
2938 like macro names defined in the configuration file.
2940 7. Added "dereference=" facility to LDAP.
2942 8. Two instances of the typo "uknown" in the source files are fixed.
2944 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2945 the Configure-Makefile script screwed up while processing it.
2947 10. Incorporated PCRE 4.4.
2949 11. The SMTP synchronization check was not operating right at the start of an
2950 SMTP session. For example, it could not catch a HELO sent before the client
2951 waited for the greeting. There is now a check for outstanding input at the
2952 point when the greeting is written. Because of the duplex, asynchronous
2953 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2954 way, but not yet received, when the check is performed.
2956 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2957 on TCP/IP sockets, because this apparently causes some broken clients to
2960 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2961 unchanged) from the Cygwin maintainer.
2963 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2964 was supported instead of testing for mbx. Effectively a typo.
2966 15. The spa authenticator server code was not checking that the input it
2967 received was valid base64.
2969 16. The debug output line for the "set" modifier in ACLs was not showing the
2970 name of the variable that was being set.
2972 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2974 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2975 lookup on its own. Something else was needed to trigger the lookup. For
2976 example, a match in host_lookup or the need for a host name in a host list.
2977 Now, if $sender_host_name is referenced and the host name has not yet been
2978 looked up, a lookup is performed. If the lookup fails, the variable remains
2979 empty, and $host_lookup_failed is set to "1".
2981 19. Added "eqi" as a case-independent comparison operator.
2983 20. The saslauthd authentication condition could segfault if neither service
2984 nor realm was specified.
2986 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2987 error message that was logged was misleading, and incoming SMTP
2988 connections were dropped. The message is now more accurate, and temporary
2989 errors are given to SMTP connections.
2991 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2992 (except RSET) with a 421 error, until QUIT is received. However, it was
2993 failing to send a response to QUIT.
2995 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2996 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2997 ACL is now run after the helo_try_verify_hosts code.
2999 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3000 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3001 case-independent, so other case variants are also recognized). Apparently
3002 some systems use these upper case variants.
3004 25. If more than two messages were waiting for the same host, and a transport
3005 filter was specified for the transport, Exim sent two messages over the
3006 same TCP/IP connection, and then failed with "socket operation on non-
3007 socket" when it tried to send the third.
3009 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3011 27. The extern definition of crypt16() in expand.c was not being excluded when
3012 the OS had its own crypt16() function.
3014 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3015 as a preferred synonym for return_size_limit, both as an option and as an
3018 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3020 30. Changed the default configuration ACL to relax the local part checking rule
3021 for addresses that are not in any local domains. For these addresses,
3022 slashes and pipe symbols are allowed within local parts, but the sequence
3023 /../ is explicitly forbidden.
3025 31. SPA server authentication was not clearing the challenge buffer before
3028 32. log_message in a "warn" ACL statement was writing to the reject log as
3029 well as to the main log, which contradicts the documentation and doesn't
3030 seem right (because no rejection is happening). So I have stopped it.
3032 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3033 However, I am unable to do any testing of this.
3035 34. Fixed an infelicity in the appendfile transport. When checking directories
3036 for a mailbox, to see if any needed to be created, it was accidentally
3037 using path names with one or more superfluous leading slashes; tracing
3038 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3040 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3041 contents are added to the log line that is written for every discarded
3042 recipient. (Previously a log_message setting was ignored.)
3044 36. The ${quote: operator now quotes the string if it is empty.
3046 37. The install script runs exim in order to find its version number. If for
3047 some reason other than non-existence or emptiness, which it checks, it
3048 could not run './exim', it was installing it with an empty version number,
3049 i.e. as "exim-". This error state is now caught, and the installation is
3052 38. An argument was missing from the function that creates an error message
3053 when Exim fails to connect to the socket for saslauthd authentication.
3054 This could cause Exim to crash, or give a corrupted message.
3056 39. Added isip, isip4, and isip6 to ${if conditions.
3058 40. The ACL variables $acl_xx are now saved with the message, and can be
3059 accessed later in routers, transports, and filters.
3061 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3062 strings in the file are not string-expanded.
3064 42. If a MAIL command specified a SIZE value that was too large to fit into an
3065 int variable, the check against message_size_limit failed. Such values are
3066 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3067 day this will have to be increased, but I don't think I want to be around
3068 when emails are that large.
3075 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3076 iconv() is not standard in FreeBSD.
3078 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3079 IPv6 enabled. The observed symptom was a segmentation fault on return from
3080 the function os_common_find_running_interfaces() in src/os.c.
3082 3. In the check_special_case() function in daemon.c I had used "errno" as an
3083 argument name, which causes warnings on some systems. This was basically a
3084 typo, since it was named "eno" in the comments!
3086 4. The code that waits for the clock to tick (at a resolution of some fraction
3087 of a second) so as to ensure message-id uniqueness was always waiting for
3088 at least one whole tick, when it could have waited for less. [This is
3089 almost certainly not relevant at current processor speeds, where it is
3090 unlikely to ever wait at all. But we try to future-proof.]
3092 5. The function that sleeps for a time interval that includes fractions of a
3093 second contained a race. It did not block SIGALRM between setting the
3094 timer, and suspending (a couple of lines later). If the interval was short
3095 and the sigsuspend() was delayed until after it had expired, the suspension
3096 never ended. On busy systems this could lead to processes getting stuck for
3099 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3100 process, before it forks any delivery processes. The open lookup caching
3101 mechanism meant that the open file or database connection was passed into
3102 the delivery process. The problem was that delivery processes always tidy
3103 up cached lookup data. This could cause a problem for the next delivery
3104 process started by the queue runner, because the external queue runner
3105 process does not know about the closure. So the next delivery process
3106 still has data in the lookup cache. In the case of a file lookup, there was
3107 no problem because closing a file descriptor in a subprocess doesn't affect
3108 the parent. However, if the lookup was caching a connection to a database,
3109 the connection was closed, and the second delivery process was likely to
3110 see errors such as "PGSQL: query failed: server closed the connection
3111 unexpectedly". The problem has been fixed by closing all cached lookups
3112 in a queue runner before running a delivery process.
3114 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3115 seem to have the "const" qualifier which it has on other OS. I've
3118 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3119 *of the same type* (client or server) with the same public name that an
3120 error should be diagnosed.
3122 9. When Exim looked up a host name for an IP address, but failed to find the
3123 original IP address when looking up the host name (a safety check), it
3124 output the message "<ip address> does not match any IP for NULL", which was
3125 confusing, to say the least. The bug was that the host name should have
3126 appeared instead of "NULL".
3128 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3129 user that is built into the binary, and the -C or -D options is used, root
3130 privilege is dropped before the configuration file is read. In addition,
3131 logging is switched to stderr instead of the normal log files. If the
3132 configuration then re-defines the Exim user, the unprivileged environment
3133 is probably not what is expected, so Exim logs a panic warning message (but
3136 However, if deliver_drop_privilege is set, the unprivileged state may well
3137 be exactly what is intended, so the warning has been cut out in that case,
3138 and Exim is allowed to try to write to its normal log files.
3144 1. smtp_return_error_details was not giving details for temporary sender
3145 or receiver verification errors.
3147 2. Diagnose a configuration error if two authenticators have the same public
3150 3. Exim used not to create the message log file for a message until the first
3151 delivery attempt. This could be confusing when incoming messages were held
3152 for policy or load reasons. The message log file is now created at the time
3153 the message is received, and an initial "Received" line is written to it.
3155 4. The automatically generated man page for command line options had a minor
3156 bug that caused no ill effects; however, a more serious problem was that
3157 the procedure for building the man page automatically didn't always
3158 operate. Consequently, release 4.20 contains an out-of-date version. This
3159 shouldn't happen again.
3161 5. When building Exim with embedded Perl support, the script that builds the
3162 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3163 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3165 6. The freeze_tell option was not being used for messages that were frozen on
3166 arrival, either by an ACL or by local_scan().
3168 7. Added the smtp_incomplete_transaction log selector.
3170 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3171 was accepting AUTH without a new EHLO.
3173 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3174 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3177 10. Logging of TCP/IP connections (when configured) now happens in the main
3178 daemon process instead of the child process, so that the TCP/IP connection
3179 count is more accurate (but it can never be perfect).
3181 11. The use of "drop" in a nested ACL was not being handled correctly in the
3182 outer ACL. Now, if condition failure induced by the nested "drop" causes
3183 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3184 or "require"), the connection is dropped.
3186 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3187 that yield "discard" can now be used with an "accept" or a "discard" verb,
3188 but an error is generated for any others (because I can't see a useful way
3189 to define what should happen).
3191 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3192 are now processed in the same way as lines in the Exim configuration file.
3193 In particular, continuation lines are supported.
3195 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3197 15. Added -ti meaning -t -i.
3199 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3200 domains, and warn by logging if others are found.
3202 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3203 returned from GIFCONF when Exim is trying to find the list of interfaces on
3204 a host. The code in os.c has been modified to copy each ifreq to an aligned
3205 structure in all cases.
3207 Also, in some cases, the returned ifreq's were being copied to a 'struct
3208 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3209 means the last couple of bytes of an IPv6 address could be chopped if the
3210 ifreq contained only a normal sockaddr (14 bytes storage).
3212 18. Named domain lists were not supported in the hosts_treat_as_local option.
3213 An entry such as +xxxx was not recognized, and was treated as a literal
3216 19. Ensure that header lines added by a DATA ACL are included in the reject log
3217 if the ACL subsequently rejects the message.
3219 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3220 MD5 (which is deprecated).
3222 21. When testing a filter file using -bf, Exim was writing a message when it
3223 took the sender from a "From " line in the message, but it was not doing so
3224 when it took $return_path from a Return-Path: header line. It now does.
3226 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3227 with a valid header line field name (a series of printing characters
3228 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3230 23. Changed "disc" in the source to "disk" to conform to the documentation and
3231 the book and for uniformity.
3233 24. Ignore Sendmail's -Ooption=value command line item.
3235 25. When execve() failed while trying to run a command in a pipe transport,
3236 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3237 could be confused with a return value of 69 from the command itself. This
3238 has been changed to 127, the value the shell returns if it is asked to run
3239 a non-existent command. The wording for the related log line suggests a
3240 non-existent command as the problem.
3242 26. If received_header_text expands to an empty string, do not add a Received:
3243 header line to the message. (Well, it adds a token one on the spool, but
3244 marks it "old" so that it doesn't get used or transmitted.)
3246 27. Installed eximstats 1.28 (addition of -nt option).
3248 28. There was no check for failure on the call to getsockname() in the daemon
3249 code. This can fail if there is a shortage of resources on the system, with
3250 ENOMEM, for example. A temporary error is now given on failure.
3252 29. Contrary to the C standard, it seems that in some environments, the
3253 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3254 program. Exim now does this explicitly; it affects the formatting of
3255 timestamps using strftime().
3257 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3258 complaints. I've now initialized all the variables, to avoid this.
3260 32. Header lines added by a system filter were not being "seen" during
3261 transport-time rewrites.
3263 33. The info_callback() function passed to OpenSSL is set up with type void
3264 (*)(SSL *, int, int), as described somewhere. However, when calling the
3265 function (actually a macro) that sets it up, the type void(*)() is
3266 expected. I've put in a cast to prevent warnings from picky compilers.
3268 34. If a DNS black list lookup found a CNAME record, but there were no A
3269 records associated with the domain it pointed at, Exim crashed.
3271 35. If a DNS black list lookup returned more than one A record, Exim ignored
3272 all but the first. It now scans all returned addresses if a particular IP
3273 value is being sought. In this situation, the contents of the
3274 $dnslist_value variable are a list of all the addresses, separated by a
3277 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3278 to accept a host name and all its aliases if the forward lookup for any of
3279 them yielded the IP address of the incoming connection. Now it accepts only
3280 those names whose forward lookup yields the correct IP address. Any other
3281 names are discarded. This closes a loophole whereby a rogue DNS
3282 administrator could create reverse DNS records to break through a
3283 wildcarded host restriction in an ACL.
3285 37. If a user filter or a system filter that ran in a subprocess used any of
3286 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3287 the wrong values were passed to the pipe command ($thisaddress had the
3288 value of $0, $0 had the value of $1, etc). This bug was introduced by
3289 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3291 38. Improved the line breaking for long SMTP error messages from ACLs.
3292 Previously, if there was no break point between 40 and 75 characters, Exim
3293 left the rest of the message alone. Two changes have been made: (a) I've
3294 reduced the minimum length to 35 characters; (b) if it can't find a break
3295 point between 35 and 75 characters, it looks ahead and uses the first one
3296 that it finds. This may give the occasional overlong line, but at least the
3297 remaining text gets split now.
3299 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3300 file descriptors might be low, and that setting 1000 would always raise it.
3301 It turns out that in some environments, the limit is already over 1000 and
3302 that lowering it causes trouble. So now Exim takes care not to decrease it.
3304 40. When delivering a message, the value of $return_path is set to $sender_
3305 address at the start of routing (routers may change the value). By an
3306 oversight, this default was not being set up when an address was tested by
3307 -bt or -bv, which affected the outcome if any router or filter referred to
3310 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3311 log only when "message" or "log_message" are set. However, if one of the
3312 conditions was an address verification, or a call to a nested ACL, the
3313 messages generated by the underlying test were being passed through. This
3314 no longer happens. The underlying message is available in $acl_verify_
3315 message for both "message" and "log_message" expansions, so it can be
3316 passed through if needed.
3318 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3319 new expansion $bh_ to give the encoded byte string without charset
3320 translation. Translation happens only if iconv() is available; HAVE_ICONV
3321 indicates this at build time. HEADERS_CHARSET gives the charset to
3322 translate to; headers_charset can change it in the configuration, and
3323 "headers charset" can change it in an individual filter file.
3325 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3326 that creates RFC 2047 encoded "words" labels them as that charset instead
3327 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3328 expansion operator; (ii) when Exim creates a From: line for a local
3329 message; (iii) when a header line is rewritten to include a "phrase" part.
3331 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3332 buggy, causing it to skip the first lines of messages whose message ID
3333 ended in 'D'. This would not have bitten before Exim release 4.14, because
3334 message IDs were unlikely to end in 'D' before then. The effect was to have
3335 incorrect size information for certain domains.
3337 45. #include "config.h" was missing at the start of the crypt16.c module. This
3338 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3341 46. If there was a timeout during a "random" callout check, Exim treated it as
3342 a failure of the random address, and carried on sending RSET and the real
3343 address. If the delay was just some slowness somewhere, the response to the
3344 original RCPT would be taken as a response to RSET and so on, causing
3345 mayhem of various kinds.
3347 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3348 when I implemented it. It didn't allow for the fact that some option values
3349 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3350 the right test for positive values.
3352 48. Domain names in DNS records are case-independent. Exim always looks them up
3353 in lower case. Some resolvers return domain names in exactly the case they
3354 appear in the zone file, that is, they may contain uppercase letters. Not
3355 all resolvers do this - some return always lower case. Exim was treating a
3356 change of case by a resolver as a change of domain, similar to a widening
3357 of a domain abbreviation. This triggered its re-routing code and so it was
3358 trying to route what was effectively the same domain again. This normally
3359 caused routing to fail (because the router wouldn't handle the domain
3360 twice). Now Exim checks for this case specially, and just changes the
3361 casing of the domain that it ultimately uses when it transmits the message
3364 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3367 50. If a filter generated a file delivery with a non-absolute name (possible if
3368 no home directory exists for the router), the forbid_file option was not
3371 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3372 the existing equality matching.
3374 52. Exim was using ints instead of ino_t variables in some places where it was
3375 dealing with inode numbers.
3377 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3378 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3379 called TMPDIR, and if it finds it is different, it changes its value.
3381 54. The smtp_printf() function is now made available to local_scan() so
3382 additional output lines can be written before returning. There is also an
3383 smtp_fflush() function to enable the detection of a dropped connection.
3384 The variables smtp_input and smtp_batched_input are exported to
3387 55. Changed the default runtime configuration: the message "Unknown user"
3388 has been removed from the ACL, and instead placed on the localuser router,
3389 using the cannot_route_message feature. This means that any verification
3390 failures that generate their own messages won't get overridden. Similarly,
3391 the "Unrouteable address" message that was in the ACL for unverifiable
3392 relay addresses has also been removed.
3394 56. Added hosts_avoid_esmtp to the smtp transport.
3396 57. The exicyclog script was not checking for the esoteric option
3397 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3398 will work only if exicyclog is run under the appropriate euid.
3400 58. Following a discussion on the list, the rules by which Exim recognises line
3401 endings on incoming messages have been changed. The -dropcr and drop_cr
3402 options are now no-ops, retained only for backwards compatibility. The
3403 following line terminators are recognized: LF CRLF CR. However, special
3404 processing applies to CR:
3406 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3407 nor a local message in the state where . is a terminator.
3409 (ii) If a bare CR is encountered in a header line, an extra space is added
3410 after the line terminator so as not to end the header. The reasoning
3411 behind this is that bare CRs in header lines are most likely either
3412 to be mistakes, or people trying to play silly games.
3414 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3415 was being incorrectly given as 18 bytes larger than it should have been.
3416 This is a VOB (very old bug).
3418 60. This may never have affected anything current, but just in case it has:
3419 When the local host is found other than at the start of a list of hosts,
3420 the local host, those with the same MX, and any that follow, are discarded.
3421 When the list in question was part of a longer list of hosts, the following
3422 hosts (not currently being processed) were also being discarded. This no
3423 longer happens. I'm not sure if this situation could ever has previously
3426 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3429 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3430 header. This is something that is recommended in a new Internet Draft, and
3431 is something that is documented as being done by Sendmail. There are two
3432 possible values. For messages generated by the autoreply transport, Exim
3435 Auto-Submitted: auto-replied
3437 whereas for all other generated messages (e.g. bounces) it adds
3439 Auto-Submitted: auto-generated
3441 63. The "personal" condition in filters now includes a test for the
3442 Auto-Submitted: header. If it contains the string "auto-" the message it
3443 not considered personal.
3445 64. Added rcpt_include_affixes as a generic transport option.
3447 65. Added queue_only_override (default true).
3449 66. Added the syslog_duplication option.
3451 67. If what should have been the first header line of a message consisted of
3452 a space followed by a colon, Exim was mis-interpreting it as a header line.
3453 It isn't of course - it is syntactically invalid and should therefore be
3454 treated as the start of the message body. The misbehaviour could have
3455 caused a number of strange effects, including loss of data in subsequent
3456 header lines, and spool format errors.
3458 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3459 client host had authenticated. This control can now be exercised by an ACL
3460 for more flexibility.
3462 69. By default, callouts do not happen when testing with -bh. There is now a
3463 variant, -bhc, which does actually run the callout code, including
3464 consulting and updating the callout cache.
3466 70. Added support for saslauthd authentication, courtesy of Alexander
3469 71. If statvfs() failed on the spool or log directories while checking their
3470 size for availability, Exim confusingly gave the error "space shortage".
3471 Furthermore, in debugging mode it crashed with a floating point exception.
3472 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3473 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3474 serious problem, Exim now writes to the main and panic logs when this
3475 happens, with details of the failure. It then refuses to accept the
3476 incoming message, giving the message "spool directory problem" or "log
3477 directory problem" with a 421 code for SMTP messages.
3479 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3480 0, 1, and 2 exist, because some OS complain for execs without them (see
3481 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3482 descriptors. However, the code omitted to check that the open succeeded,
3483 causing mysterious errors if for some reason the permissions on /dev/null
3484 got screwed. Now Exim writes a message to the main and panic logs, and
3485 bombs out if it can't open /dev/null.
3487 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3488 interact so that it is all more flexible. It is supposed to remain
3489 backwards compatible. Also added extra_local_interfaces.
3491 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3492 to bomb out with an assertion failure - to the client this appears as a
3493 connection drop. This problem occurs in the part of the code that was taken
3494 from the Samba project. Fortunately, the assertion is in a very simple
3495 function, so I have fixed this by reproducing the function inline in the
3496 one place where it is called, and arranging for authentication to fail
3497 instead of killing the process with assert().
3499 75. The SPA client code was not working when the server requested OEM rather
3500 than Unicode encoding.
3502 76. Added code to make require_files with a specific uid setting more usable in
3503 the case where statting the file as root fails - usually a non-root-mounted
3504 NFS file system. When this happens and the failure is EACCES, Exim now
3505 forks a subprocess and does the per-uid checking as the relevant uid.
3507 77. Added process_log_path.
3509 78. If log_file_path was not explicitly set, a setting of check_log_space or
3510 check_log_inodes was ignored.
3512 79. If a space check for the spool or log partitions fails, the incident is now
3513 logged. Of course, in the latter case the data may get lost...
3515 80. Added the %p formatting code to string_format() so that it can be used to
3516 print addresses in debug_print(). Adjusted all the address printing in the
3517 debugging in store.c to use %p rather than %d.
3519 81. There was a concern that a line of code in smtp_in.c could overflow a
3520 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3521 initially expressed, the concern was not well-founded, because trailing
3522 spaces are removed early. However, if the trailing spaces were followed by
3523 a NULL, they did not get removed, so the overflow was possible. Two fixes
3526 (a) I re-wrote the offending code in a cleaner fashion.
3527 (b) If an incoming SMTP command contains a NULL character, it is rejected
3530 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3531 runs initgroups(), so that if the Exim user is in any additional groups,
3532 they will be used during message reception.
3538 The change log for 4.20 and earlier releases has been archived.