1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.273 2005/12/12 11:41:50 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 The code for finding all the local interface addresses on a FreeBSD
10 system running IPv6 was broken. This may well have applied to all BSD
11 systems, as well as to others that have similar system calls. The broken
12 code found IPv4 interfaces correctly, but gave incorrect values for the
13 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
14 that it would not match correctly against @[] and not recognize the IPv6
17 PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
20 PH/03 Added disable_ipv6.
22 PH/04 Changed $reply_address to use the raw form of the headers instead of the
23 decoded form, because it is most often used to construct To: headers
24 lines in autoreplies, and the decoded form may well be syntactically
25 invalid. However, $reply_address has leading white space removed, and all
26 newlines turned into spaces so that the autoreply transport does not
29 PH/05 If group was specified without a user on a router, and no group or user
30 was specified on a transport, the group from the router was ignored.
36 PH/01 Two changes to the default runtime configuration:
38 (1) Move the checks for relay_from_hosts and authenticated clients from
39 after to before the (commented out) DNS black list checks.
41 (2) Add control=submission to the relay_from_hosts and authenticated
42 clients checks, on the grounds that messages accepted by these
43 statements are most likely to be submissions.
45 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
47 (1) Generate an error if the third argument for the ${prvs expansion is
50 (2) Treat a missing third argument of ${prvscheck as if it were an empty
53 (3) Reset the variables that are obtained from the first argument of
54 ${prvscheck and used in the second argument before leaving the code,
55 because their memory is reclaimed, so using them afterwards may do
58 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
59 one (it's much easier than Tom thought :-).
61 (5) Because of (4), we can now allow for the use of $prvscheck_result
62 inside the third argument.
64 PH/03 For some reason, the default setting of PATH when running a command from
65 a pipe transport was just "/usr/bin". I have changed it to
68 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
69 anything to be listed in the output from -bV.
71 PH/05 When a filter generated an autoreply, the entire To: header line was
72 quoted in the delivery log line, like this:
74 => >A.N.Other <ano@some.domain> <original@ddress> ...
76 This has been changed so that it extracts the operative address. There
77 may be more than one such address. If so, they are comma-separated, like
80 => >ano@some.domain,ona@other.domain <original@ddress> ...
82 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
83 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
84 not being looked up in the rDNS to get a host name, Exim was showing the
85 IP address twice in Received: lines, even though the IP addresses were
86 identical. For example:
88 Received: from [1.2.3.4] (helo=[1.2.3.4])
90 However, if the real host name was known, it was omitting the HELO data
91 if it matched the actual IP address. This has been tidied up so that it
92 doesn't show the same IP address twice.
94 PH/07 When both +timestamp and +memory debugging was on, the value given by
95 $tod_xxx expansions could be wrong, because the tod_stamp() function was
96 called by the debug printing, thereby overwriting the timestamp buffer.
97 Debugging no longer uses the tod_stamp() function when +timestamp is set.
99 PH/08 When the original message was included in an autoreply transport, it
100 always said "this is a copy of the message, including all the headers",
101 even if body_only or headers_only was set. It now gives an appropriate
104 PH/09 Applied a patch from the Sieve maintainer which:
106 o fixes some comments
107 o adds the (disabled) notify extension core
108 o adds some debug output for the result of if/elsif tests
109 o points to the current vacation draft in the documentation
110 and documents the missing references header update
114 o fixes a bug in processing the envelope test (when testing
115 multiple envelope elements, the last element determinted the
118 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
119 Electronic Mail") by including:
121 Auto-submitted: auto-generated
123 in the messages that it generates (bounce messages and others, such as
124 warnings). In the case of bounce messages for non-SMTP mesages, there was
125 also a typo: it was using "Auto_submitted" (underscore instead of
126 hyphen). Since every message generated by Exim is necessarily in response
127 to another message, thes have all been changed to:
129 Auto-Submitted: auto-replied
131 in accordance with these statements in the RFC:
133 The auto-replied keyword:
135 - SHOULD be used on messages sent in direct response to another
136 message by an automatic process,
138 - MUST NOT be used on manually-generated messages,
140 - MAY be used on Delivery Status Notifications (DSNs) and Message
141 Disposition Notifications (MDNs),
143 - MUST NOT be used on messages generated by automatic or periodic
144 processes, except for messages which are automatic responses to
147 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
148 to the default Received: header definition.
150 PH/12 Added log selector acl_warn_skipped (default on).
152 PH/13 After a successful wildlsearch lookup, discard the values of numeric
153 variables because (a) they are in the wrong storage pool and (b) even if
154 they were copied, it wouldn't work properly because of the caching.
156 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
157 checking when decoding. Apparently there are clients that generate
158 overlong encoded strings. Why am I not surprised?
160 PH/15 If the first argument of "${if match_address" was not empty, but did not
161 contain an "@" character, Exim crashed. Now it writes a panic log message
162 and treats the condition as false.
164 PH/16 In autoreply, treat an empty string for "once" the same as unset.
166 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
167 extension "envelope-auth". The code is finished and in agreement with
168 other implementations, but there is no documentation so far and in fact,
169 nobody wrote the draft yet. This extension is currently #undef'ed, thus
170 not changing the active code.
172 Print executed "if" and "elsif" statements when debugging is used. This
173 helps a great deal to understand what a filter does.
175 Document more things not specified clearly in RFC3028. I had all this
176 sorted out, when out of a sudden new issues came to my mind. Oops."
178 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
181 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
182 canonical form (as documented). However, after a host name lookup from
183 the IP address, check_host() was doing a simple string comparison with
184 addresses acquired from the DNS when checking that the found name did
185 have the original IP as one of its addresses. Since any found IPv6
186 addresses are likely to be in abbreviated form, the comparison could
187 fail. Luckily, there already exists a function for doing the comparison
188 by converting both addresses to binary, so now that is used instead of
191 PH/20 There was another similar case to PH/19, when a complete host name was
192 given in a host list; looking up its IP address could give an abbreviated
193 form, whereas the current host's name might or might not be abbreviated.
194 The same fix has been applied.
200 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
201 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
204 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
205 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
207 PH/03 Typo: missing ".o" in src/pcre/Makefile.
209 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
210 header line, restrict the check to what is listed in RFCs 2369 and 2929.
211 Also, for "Auto-Submitted", treat anything other than "no" as
212 non-personal, in accordance with RFC 3834. (Previously it treated
213 anything starting "auto-" as non-personal.)
215 TF/01 The control=submission/name=... option had a problem with syntax
216 errors if the name included a slash character. The /name= option
217 now slurps the rest of the string, so it can include any characters
218 but it must come last in the list of options (after /sender_retain
221 PH/05 Some modifications to the interface to the fake nameserver for the new
224 PH/06 Add ${if def:sender_address {(envelope-from <$sender_address>)\n\t}}
225 into the default Received: header string.
232 TK/01 Added the "success_on_redirect" address verification option. See
233 NewStuff for rationale and an example.
235 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
237 PH/02 Patch to exigrep to allow it to work on syslog lines.
239 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
240 fread() to skip over the body file's header line, because in Cygwin the
241 header line is locked and is inaccessible.
243 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
244 co-exist for some time) to make it clear that it is the Exim ID that is
245 referenced, not the Message-ID: header line.
247 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
248 string_format() function, because snprintf() does not exist on all
251 PH/06 The use of forbid_filter_existstest now also locks out the use of the
252 ${stat: expansion item.
254 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
255 protocol synchronization error", to keep the pedants happy.
257 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
258 well as for IRIX systems, when gcc is being used. See the host.c source
261 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
263 PH/10 Named domain lists were not working if used in a queue_smtp_domains
266 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
267 transport and to the smtp transport in LMTP mode.
269 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
271 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
272 run a filter in a subprocess. This could lead to confusion in subsequent
273 lookups in the parent process. There should also be a search_tidyup() at
274 the end of the subprocess.
276 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
277 only if the host matched helo_try_verify_hosts, which caused the
278 verification to occur when the EHLO/HELO command was issued. The ACL just
279 tested the remembered result. Now, if a previous verification attempt has
280 not happened, "verify = helo" does it there and then.
282 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
284 TK/03 Fix log output including CR from clamd.
286 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
287 exist provoked a memory error which could cause a segfault.
289 PH/15 Installed PCRE 6.2
291 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
293 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
294 of the problem. Specifically, suggested +O2 rather than +O1 for the
297 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
299 PH/20 If a delivery was routed to a non-standard port by means of an SRV
300 record, the port was not correctly logged when the outgoing_port log
301 selector was set (it logged the transort's default port).
303 PH/21 Added support for host-specific ports to manualroute, queryprogram,
304 fallback_hosts, and "hosts" in the smtp transport.
306 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
307 host errors such as "Connection refused".
309 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
310 authentication with radiusclient 0.4.9:
312 - Error returned from rc_read_config was caught wrongly
313 - Username/password not passed on to radius server due to wrong length.
315 The presumption is that some radiusclient API changes for 4.51/PH/17
316 were not taken care of correctly. The code is still untested by me (my
317 Linux distribution still has 0.3.2 of radiusclient), but it was
318 contributed by a Radius user.
320 PH/24 When doing a callout, the value of $domain wasn't set correctly when
321 expanding the "port" option of the smtp transport.
323 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
324 while reading a MIME header. Thanks to Tom Hughes for a patch.
326 PH/24 Include config.h inside local_scan.h so that configuration settings are
329 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
330 that in an ACL for RCPT (for example), you can examine exactly what was
333 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
334 commands, but it was not correctly comparing the address with the actual
335 client host address. Thus, it would show the EHLO address in Received:
336 header lines when this was not necessary.
338 PH/27 Added the % operator to ${eval:}.
340 PH/28 Exim tries to create and chdir to its spool directory when it starts;
341 it should be ignoring failures (because with -C, for example, it has lost
342 privilege). It wasn't ignoring creation failures other than "already
345 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
348 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
349 because an input line was too long, either on its own, or by virtue of
350 too many continuations, the temporary file was not being removed, and the
351 return code was incorrect.
353 PH/31 Missing "BOOL" in function definition in filtertest.c.
355 PH/32 Applied Sieve patches from the maintainer.
357 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
359 PH/33 Added "verify = not_blind".
361 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
362 Local/Makefile (with some defaults set). These are used in built scripts
363 such as exicyclog, but they have never been used in the exim_install
364 script (though there are many overriding facilities there). I have
365 arranged that the exim_install script now takes note of these two
368 PH/35 Installed configuration files for Dragonfly.
370 PH/36 When a locally submitted message by a trusted user did not contain a
371 From: header, and the sender address was obtained from -f or from an SMTP
372 MAIL command, and the trusted user did not use -F to supply a sender
373 name, $originator_name was incorrectly used when constructing a From:
374 header. Furthermore, $originator_name was used for submission mode
375 messages from external hosts without From: headers in a similar way,
376 which is clearly wrong.
378 PH/37 Added control=suppress_local_fixups.
380 PH/38 When log_selector = +received_sender was set, and the addition of the
381 sender made the log line's construction buffer exactly full, or one byte
382 less than full, an overflow happened when the terminating "\n" was
385 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
386 when the result of a list match is failure because a DNS lookup failed.
388 PH/40 RM_COMMAND is now used in the building process.
390 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
391 the "build-* directories that it finds.
393 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
394 address in a domain literal was a prefix of an interface address.
396 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
397 when verifying a sender address, unless rewrite_headers is false.
399 PH/44 Wrote a long comment about why errors_to addresses are verified as
400 recipients, not senders.
402 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
403 the ratelimit ACL was added.
405 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
407 PH/46 Added extra information about PostgreSQL errors to the error string.
409 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
410 avoiding the need to install special zones in a real server. This is
411 backwards compatible; if it can't find the fake resolver, it drops back.
412 Thus, both old and new test suites can be run.
414 TF/02 Added util/ratelimit.pl
416 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
417 clock is set back in time.
419 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
420 Candler <B.Candler@pobox.com>.
422 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
423 OK for addresses that are the result of redirections.
425 PH/48 A number of further additions for the benefit of the new test suite,
426 including a fake gethostbyname() that interfaces to the fake DNS resolver
429 TF/06 The fix for widen_domains has also been applied to qualify_single and
430 search_parents which are the other dnslookup options that can cause
433 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
436 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
438 TF/07 Exim produced the error message "an SRV record indicated no SMTP
439 service" if it encountered an MX record with an empty target hostname.
440 The message is now "an MX or SRV record indicated no SMTP service".
442 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
443 if the DNS of the sending site is misconfigured. This is quite a
444 common situation. This change restores the behaviour of treating a
445 helo verification defer as a failure.
447 PH/51 If self=fail was set on a router, the bounce message did not include the
448 actual error message.
454 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
456 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
457 command itself ended in error, the underlying message about the transport
458 filter timeout was being overwritten with the pipe command error. Now the
459 underlying error message should be appended to the second error message.
461 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
463 PH/02 Reduce the amount of output that "make" produces by default. Full output
464 can still be requested.
466 PH/03 The warning log line about a condition test deferring for a "warn" verb
467 was being output only once per connection, rather than after each
468 occurrence (because it was using the same function as for successful
469 "warn" verbs). This seems wrong, so I have changed it.
471 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
472 it should not have, which might have caused a crash in the right
473 circumstances, but probably never did.
475 PH/04 Installed a modified version of Tony Finch's patch to make submission
476 mode fix the return path as well as the Sender: header line, and to
477 add a /name= option so that you can make the user's friendly name appear
480 TF/03 Added the control = fakedefer ACL modifier.
482 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
483 Mark Lowes for thorough testing.
485 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
487 TK/03 Merged latest SRS patch from Miles Wilton.
489 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
490 with the definition in sysexits.h (which is #included earlier).
491 Fortunately, Exim does not actually use EX_OK. The code used to try to
492 preserve the sysexits.h value, by assumimg that macro definitions were
493 scanned for macro replacements. I have been disabused of this notion,
494 so now the code just undefines EX_OK before #including unistd.h.
496 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
497 in the smtp transport. When a block could not be written in a single
498 write() function, the timeout was being re-applied to each part-write.
499 This seems wrong - if the receiver was accepting one byte at a time it
500 would take for ever. The timeout is now adjusted when this happens. It
501 doesn't have to be particularly precise.
503 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
504 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
506 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
507 without a domain if the check to <postmaster@domain> fails.
509 SC/01 Eximstats: added -xls and the ability to specify output files
510 (patch written by Frank Heydlauf).
512 SC/02 Eximstats: use FileHandles for outputing results.
514 SC/03 Eximstats: allow any combination of xls, txt, and html output.
516 SC/04 Eximstats: fixed display of large numbers with -nvr option
518 SC/05 Eximstats: fixed merging of reports with empty tables.
520 SC/06 Eximstats: added the -include_original_destination flag
522 SC/07 Eximstats: removed tabs and trailing whitespace.
524 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
526 TK/06 MBOX spool code: Add real "From " MBOX separator line
527 so the .eml file is really in mbox format (even though
528 most programs do not really care). Patch from Alex Miller.
530 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
531 The latter is generated from $received_to and is only set if the
532 message has one envelope recipient. SA can use these headers,
533 obviously out-of-the-box. Patch from Alex Miller.
535 PH/08 The ${def test on a variable was returning false if the variable's
536 value was "0", contrary to what the specification has always said!
537 The result should be true unless the variable is empty.
539 PH/09 The syntax error of a character other than { following "${if
540 def:variable_name" (after optional whitespace) was not being diagnosed.
541 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
542 accidental colon was present, for example, could give incorrect results.
544 PH/10 Tidied the code in a number of places where the st_size field of a stat()
545 result is used (not including appendfile, where other changes are about
548 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
549 This involved changing a lot of size variables from int to off_t. It
550 should work with maildirs and everything.
552 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
553 spamd dying while we are connected to it.
555 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
558 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
559 function for BDB 4.3.
561 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
563 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
564 the sources for PCRE have been re-arranged and more files are now
567 PH/15 The code I had for printing potentially long long variables in PH/11
568 above was not the best (it lost precision). The length of off_t variables
569 is now inspected at build time, and an appropriate printing format (%ld
570 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
571 to be "long long int" or "long int". This is needed for the internal
572 formatting function string_vformat().
574 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
575 the configuration file to be ":syslog", then the script "guesses" where
576 the logs files are, rather than using the compiled in default. In our
577 case the guess is not the same as the compiled default, so the script
578 suddenly stopped working when I started to use syslog. The patch checks
579 to see if log_file_path is "". If so, it attempts to read it from exim
580 with no configuration file to get the compiled in version, before it
581 falls back to the previous guessing code."
583 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
584 implementing BATV in an Exim configuration. See NewStuff for the gory
587 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
588 Makefile that are specific to HP-UX.
590 PH/18 If the "use_postmaster" option was set for a recipient callout together
591 with the "random" option, the postmaster address was used as the MAIL
592 FROM address for the random test, but not for the subsequent recipient
593 test. It is now used for both.
595 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
596 patch removes a few documentation additions to RFC 3028, because the
597 latest draft now contains them. It adds the new en;ascii-case comparator
598 and a new error check for 8bit text in MIME parts. Comparator and
599 require names are now matched exactly. I enabled the subaddress
600 extension, but it is not well tested yet (read: it works for me)."
602 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
603 rework some of the code of TK/09 above to avoid the hardwired use of
604 "%lld" and "long long". Replaced the call to snprintf() with a call to
607 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
608 records point to non-existent hosts", "retry timeout exceeded", and
609 "retry time not reached for any host after a long failure period".
611 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
612 experimental DomainKeys support:
614 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
615 (2) On an error such as an illegally used "control", the wrong name for
616 the control was given.
618 These problems did NOT occur unless DomainKeys support was compiled.
620 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
622 PH/24 Added ${if match_ip condition.
624 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
625 databases so that it will be absolutely obvious if a crash occurs in the
626 DB library. This is a regular occurrence (often caused by mis-matched
629 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
630 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
631 picked up on a user's system that detects such things. There doesn't seem
632 to be a gcc warning option for this - only an attribute that has to be
633 put on the function's prototype. It seems that in Fedora Core 4 they have
634 set this on a number of new functions. No doubt there will be more in due
637 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
638 specify a transport. However, if an address that was verified by such a
639 router was the subject of a callout, Exim crashed because it tried to
640 read the rcpt_include_affixes from the non-existent transport. Now it
641 just assumes that the setting of that option is false. This bug was
642 introduced by 4.51/PH/31.
644 PH/28 Changed -d+all to exclude +memory, because that information is very
645 rarely of interest, but it makes the output a lot bigger. People tend to
646 do -d+all out of habit.
648 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
649 code in os-type was giving problems when libc.so lives in lib64, like on
652 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
653 aren't the modern standard, and it seems that some systems' include files
654 don't always have them. Exim was already checking for some of the newer
655 ones like T_AAAA, and defining it itself. I've added checks for all the
656 record types that Exim uses.
658 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
659 not automatically generating a new one, as it is supposed to. This
660 prevented TLS from working. If the file did exist, but contained invalid
661 data, a new version was generated, as expected. It was only the case of a
662 non-existent file that was broken.
664 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
665 with a change in libdomainkeys > 0.64.
667 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
668 from DNS. If the selector record carries the flag, it now has
669 precedence over the domain-wide flag.
671 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
673 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
674 the use of a port name that isn't defined in /etc/services) occurred, the
675 message was deferred as in a normal delivery, and thus remained on the
676 spool, instead of being failed because of the mua_wrapper setting. This
677 is now fixed, and I tidied up some of the mua_wrapper messages at the
680 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
681 the messages in a hash of arrays rather than using individual hashes.
682 This is a bit cleaner and results in dramatic memory savings, albeit
683 at a slight CPU cost.
685 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
686 as requested by Marc Sherman.
688 SC/10 Eximstats: added histograms for user specified patterns as requested
691 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
693 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
694 fopen() in the content-scanning modules that did not already have it.
700 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
701 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
703 TK/02 Fix ACL "control" statement not being available in MIME ACL.
705 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
707 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
708 to test Sieve filters that use "vacation".
710 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
711 that changes the way the GnuTLS parameters are stored in the cache file.
712 The new format can be generated externally. For backward compatibility,
713 if the data in the cache doesn't make sense, Exim assumes it has read an
714 old-format file, and it generates new data and writes a new file. This
715 means that you can't go back to an older release without removing the
718 PH/03 A redirect router that has both "unseen" and "one_time" set does not
719 work if there are any delivery delays because "one_time" forces the
720 parent to be marked "delivered", so its unseen clone is never tried
721 again. For this reason, Exim now forbids the simultaneous setting of
724 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
725 redirected to themselves ("homonym" addresses). Read the long ChangeLog
726 entry if you want to know the details. The fix, however, neglected to
727 consider the case when local delivery batching is involved. The test for
728 "previously delivered" was not happening when checking to see if an
729 address could be batched with a previous (undelivered) one; under
730 certain circumstances this could lead to multiple deliveries to the same
733 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
734 in its include files, and this causes problems building Exim.
736 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
737 header_syntax) but Exim was just ignoring anything given after a slash.
738 In particular, this caused confusion with an attempt to use "verify =
739 reverse_host_lookup/defer_ok". An error is now given when options are
740 supplied for verify items that do not have them. (Maybe reverse_host_
741 lookup should have a defer_ok option, but that's a different point.)
743 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
744 defined by RFC 821) to 2048, because there were problems with some AUTH
745 commands, and RFC 1869 says the size should be increased for extended
746 SMTP commands that take arguments.
748 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
751 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
752 "unknown" error; now it says that the functionality isn't in the binary.
754 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
755 an address' error message when a string expansion fails (syntax or
756 whatever). Otherwise the password may appear in the log. Following change
757 PH/42 below, there is no longer a chance of it appearing in a bounce
760 PH/11 Installed exipick version 20050225.0 from John Jetmore.
762 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
763 of its addresses was ever tried. (Bugzilla bug #2.)
765 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
766 the result incorrectly in the debug output. (It correctly added a newline
767 to what was transported.)
769 TF/01 Added $received_time.
771 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
772 commented out examples of how to interface to a virus scanner and to
773 SpamAssassin. Also added commented examples of av_scanner and
774 spamd_address settings.
776 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
777 and controls are allowed in which ACLs. There were a couple of minor
778 errors. Some of the entries in the conditions table (which is a table of
779 where they are NOT allowed) were getting very unwieldy; rewrote them as a
780 negation of where the condition IS allowed.
782 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
784 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
785 header file does not have a version number, so I've had to invent a new
786 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
787 API. The code is untested by me (my Linux distribution still has 0.3.2 of
788 radiusclient), but it was contributed by a Radius user.
790 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
791 files or directories, for OpenSSL.
793 PH/19 When an Exim process that is running as root has to create an Exim log
794 file, it does so in a subprocess that runs as exim:exim so as to get the
795 ownership right at creation (otherwise, other Exim processes might see
796 the file with the wrong ownership). There was no test for failure of this
797 fork() call, which would lead to the process getting stuck as it waited
798 for a non-existent subprocess. Forks do occasionally fail when resources
799 run out. I reviewed all the other calls to fork(); they all seem to check
802 PH/20 When checking for unexpected SMTP input at connect time (before writing
803 the banner), Exim was not dealing correctly with a non-positive return
804 from the read() function. If the client had disconnected by this time,
805 the result was a log entry for a synchronization error with an empty
806 string after "input=" when read() returned zero. If read() returned -1
807 (an event I could not check), uninitialized data bytes were printed.
808 There were reports of junk text (parts of files, etc) appearing after
811 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
813 PH/22 Added support for macro redefinition, and (re)definition in between
814 driver and ACL definitions.
816 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
817 forgetting to use the resulting value; it was using the unexpanded value.
819 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
820 hadn't been configured. The fix is from Juergen Kreileder, who
821 understands it better than I do:
823 "Here's what I see happening with three configured cyrus_sasl
824 authenticators configured (plain, login, cram-md5):
826 On startup auth_cyrus_sasl_init() gets called for each of these.
827 This means three calls to sasl_listmech() without a specified mech_list.
828 => SASL tests which mechs of all available mechs actually work
829 => three warnings about OTP not working
830 => the returned list contains: plain, login, cram-md5, digest-md5, ...
832 With the patch, sasl_listmech() also gets called three times. But now
833 SASL's mech_list option is set to the server_mech specified in the the
834 authenticator. Or in other words, the answer from sasl_listmech()
835 gets limited to just the mech you're testing for (which is different
837 => the return list contains just 'plain' or 'login', 'cram-md5' or
838 nothing depending on the value of ob->server_mech.
840 I've just tested the patch: Authentication still works fine,
841 unavailable mechs specified in the exim configuration are still
842 caught, and the auth.log warnings about OTP are gone."
844 PH/25 When debugging is enabled, the contents of the command line are added
845 to the debugging output, even when log_selector=+arguments is not
848 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
849 answer is "GNU", and only if the return is "GNU/something" is the answer
852 PH/27 $acl_verify_message is now set immediately after the failure of a
853 verification in an ACL, and so is available in subsequent modifiers. In
854 particular, the message can be preserved by coding like this:
856 warn !verify = sender
857 set acl_m0 = $acl_verify_message
859 Previously, $acl_verify_message was set only while expanding "message"
860 and "log_message" when a very denied access.
862 PH/28 Modified OS/os.c-Linux with
864 -#ifndef OS_LOAD_AVERAGE
865 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
867 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
868 nomenclature these days.)
870 PH/29 Installed patch from the Sieve maintainer that adds the options
871 sieve_useraddress and sieve_subaddress to the redirect router.
873 PH/30 In these circumstances:
874 . Two addresses routed to the same list of hosts;
875 . First host does not offer TLS;
876 . First host accepts first address;
877 . First host gives temporary error to second address;
878 . Second host offers TLS and a TLS session is established;
879 . Second host accepts second address.
880 Exim incorrectly logged both deliveries with the TLS parameters (cipher
881 and peerdn, if requested) that were in fact used only for the second
884 PH/31 When doing a callout as part of verifying an address, Exim was not paying
885 attention to any local part prefix or suffix that was matched by the
886 router that accepted the address. It now behaves in the same way as it
887 does for delivery: the affixes are removed from the local part unless
888 rcpt_include_affixes is set on the transport.
890 PH/32 Add the sender address, as F=<...>, to the log line when logging a
891 timeout during the DATA phase of an incoming message.
893 PH/33 Sieve envelope tests were broken for match types other than :is. I have
894 applied a patch sanctioned by the Sieve maintainer.
896 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
897 the uid or gid is negative. A case of a negative gid caused this to be
898 noticed. The fix allows for either to be negative.
900 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
901 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
904 PH/36 The change PH/12 above was broken. Fixed it.
906 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
907 the grounds that routing the same address twice would always produce the
908 same answer. This might have been true once, but it is certainly no
909 longer true now. Routing a child address may depend on the previous
910 routing that produced that child. Some complicated redirection strategies
911 went wrong when messages had multiple recipients, and made Exim's
912 behaviour dependent on the order in which the addresses were given.
914 I have moved the duplicate checking until after the routing is complete.
915 Exim scans the addresses that are assigned to local and remote
916 transports, and removes any duplicates. This means that more work will be
917 done, as duplicates will always all be routed, but duplicates are
918 presumably rare, so I don't expect this is of any significance.
920 For deliveries to pipes, files, and autoreplies, the duplicate checking
921 still happens during the routing process, since they are not going to be
924 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
925 It corrects a timeout issue with spamd. This is Ian's comment: "The
926 background is that sometimes spamd either never reads data from a
927 connection it has accepted, or it never writes response data. The exiscan
928 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
929 blindly assumes that writes won't block so it may never time out."
931 PH/39 Allow G after quota size as well as K and M.
933 PH/40 The value set for $authenticated_id in an authenticator may not contain
934 binary zeroes or newlines because the value is written to log lines and
935 to spool files. There was no check on this. Now the value is run through
936 the string_printing() function so that such characters are converted to
937 printable escape sequences.
939 PH/41 $message_linecount is a new variable that contains the total number of
940 lines in the message. Compare $body_linecount, which is the count for the
943 PH/42 Exim no longer gives details of delivery errors for specific addresses in
944 bounce and delay warning messages, except in certain special cases, which
947 (a) An SMTP error message from a remote host;
948 (b) A message specified in a :fail: redirection;
949 (c) A message specified in a "fail" command in a system filter;
950 (d) A message specified in a FAIL return from the queryprogram router;
951 (e) A message specified by the cannot_route_message router option.
953 In these cases only, Exim does include the error details in bounce and
954 warning messages. There are also a few cases where bland messages such
955 as "unrouteable address" or "local delivery error" are given.
957 PH/43 $value is now also set for the "else" part of a ${run expansion.
959 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
960 being worked on, but at least Exim now implements the latest version to
963 PH/45 In a pipe transport, although a timeout while waiting for the pipe
964 process to complete was treated as a delivery failure, a timeout while
965 writing the message to the pipe was logged, but erroneously treated as a
966 successful delivery. Such timeouts include transport filter timeouts. For
967 consistency with the overall process timeout, these timeouts are now
968 treated as errors, giving rise to delivery failures by default. However,
969 there is now a new Boolean option for the pipe transport called
970 timeout_defer, which, if set TRUE, converts the failures into defers for
971 both kinds of timeout. A transport filter timeout is now identified in
974 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
975 systems where "make" and "gmake" are different, calling "gmake" at top
976 level broke things. I've arranged for the value of $(MAKE) to be passed
977 from the Makefile to this script so that it can call the same version of
981 A note about Exim versions 4.44 and 4.50
982 ----------------------------------------
984 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
985 changes of various kinds. As a consequence, a big documentation update was
986 needed. This delayed the release for rather longer than seemed good, especially
987 in the light of a couple of (minor) security issues. Therefore, the changes
988 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
989 release. So 4.44 and 4.50 are in effect two different branches that both start
992 I have left the 4.50 change log unchanged; it contains all the changes since
993 4.43. The change log for 4.44 is below; many of its items are identical to
994 those for 4.50. This seems to be the most sensible way to preserve the
995 historical information.
1001 1. Minor wording change to the doc/README.SIEVE file.
1003 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1004 computation of the current number of files was incorrect.
1006 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1007 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1010 4. Give more explanation in the error message when the command for a transport
1011 filter fails to execute.
1013 5. There are several places where Exim runs a non-Exim command in a
1014 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1015 was being done only for the command run by the queryprogram router. It is
1016 now done for all such subprocesses. The other cases are: ${run, transport
1017 filters, and the commands run by the lmtp and pipe transports.
1019 6. Added CONFIGURE_GROUP build-time option.
1021 7. Some older OS have a limit of 256 on the maximum number of file
1022 descriptors. Exim was using setrlimit() to set 1000 as a large value
1023 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1024 systems. I've change it so that if it can't get 1000, it tries for 256.
1026 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1027 was an oversight, and furthermore, ever since the addition of extra
1028 controls (e.g. 4.43/32), the checks on when to allow different forms of
1029 "control" were broken. There should now be diagnostics for all cases when a
1030 control that does not make sense is encountered.
1032 9. Added the /retain_sender option to "control=submission".
1034 10. $recipients is now available in the predata ACL (oversight).
1036 11. Tidy the search cache before the fork to do a delivery from a message
1037 received from the command line. Otherwise the child will trigger a lookup
1038 failure and thereby defer the delivery if it tries to use (for example) a
1039 cached ldap connection that the parent has called unbind on.
1041 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1042 of $address_data from the recipient verification was clobbered by the
1043 sender verification.
1045 13. The value of address_data from a sender verification is now available in
1046 $sender_address_data in subsequent conditions in the ACL statement.
1048 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1050 15. Added a new option "connect=<time>" to callout options, to set a different
1053 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1054 was its contents. (It was OK if the option was not defined at all.)
1056 17. A "Completed" log line is now written for messages that are removed from
1057 the spool by the -Mrm option.
1059 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1060 information about exactly what failed.
1062 19. Added -dd to debug only the daemon process.
1064 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1065 handles timeouts, both on the server side and network timeouts. Renamed the
1066 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1068 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1069 It is now set to "smtps".
1071 22. $host_address is now set to the target address during the checking of
1072 ignore_target_hosts.
1074 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1075 being passed; this would have caused $sender_host_name to have been used if
1076 matching the list had actually called for a host name (not very likely,
1077 since this list is usually IP addresses). A host name is now passed as
1080 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1081 code that specifies a non-restarting handler (typically sigaction() in
1082 modern systems) in an attempt to fix a rare and obscure crash bug.
1084 25. Narrowed the window for a race in the daemon that could cause it to ignore
1085 SIGCHLD signals. This is not a major problem, because they are used only to
1086 wake it up if nothing else does.
1088 26. A malformed maildirsize file could cause Exim to calculate negative values
1089 for the mailbox size or file count. Odd effects could occur as a result.
1090 The maildirsize information is now recalculated if the size or filecount
1093 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1094 support for a long time. Removed HAVE_SYS_VFS_H.
1096 28. Installed the latest version of exipick from John Jetmore.
1098 29. In an address list, if the pattern was not a regular expression, an empty
1099 subject address (from a bounce message) matched only if the pattern was an
1100 empty string. Non-empty patterns were not even tested. This was the wrong
1101 because it is perfectly reasonable to use an empty address as part of a
1102 database query. An empty address is now tested by patterns that are
1103 lookups. However, all the other forms of pattern expect the subject to
1104 contain a local part and a domain, and therefore, for them, an empty
1105 address still always fails if the pattern is not itself empty.
1107 30. Exim went into a mad DNS loop when attempting to do a callout where the
1108 host was specified on an smtp transport, and looking it up yielded more
1109 than one IP address.
1111 31. Re-factored the code for checking spool and log partition space into a
1112 function that finds that data and another that does the check. The former
1113 is then used to implement four new variables: $spool_space, $log_space,
1114 $spool_inodes, and $log_inodes.
1116 32. The RFC2047 encoding function was originally intended for short strings
1117 such as real names; it was not keeping to the 75-character limit for
1118 encoded words that the RFC imposes. It now respects the limit, and
1119 generates multiple encoded words if necessary. To be on the safe side, I
1120 have increased the buffer size for the ${rfc2047: expansion operator from
1123 33. It is now permitted to omit both strings after an "if" condition; if the
1124 condition is true, the result is "true". As before, when the second string
1125 is omitted, a false condition yields an empty string. This makes it less
1126 cumbersome to write custom ACL and router conditions.
1128 34. Failure to deliver a bounce message always caused it to be frozen, even if
1129 there was an errors_to setting on the router. The errors_to setting is now
1132 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1133 canonical form (fully expanded) before being placed in
1134 $sender_host_address.
1136 36. The table in the code that translates DNS record types into text (T_A to
1137 "A" for instance) was missing entries for NS and CNAME. It is just possible
1138 that this could have caused confusion if both these types were looked up
1139 for the same domain, because the text type is used as part of Exim's
1140 per-process caching. But the chance of anyone hitting this buglet seems
1143 37. The dnsdb lookup has been extended in a number of ways.
1145 (1) There is a new type, "zns", which walks up the domain tree until it
1146 finds some nameserver records. It should be used with care.
1148 (2) There is a new type, "mxh", which is like "mx" except that it returns
1149 just the host names, not the priorities.
1151 (3) It is now possible to give a list of domains (or IP addresses) to be
1152 looked up. The behaviour when one of the lookups defers can be
1153 controlled by a keyword.
1155 (4) It is now possible to specify the separator character for use when
1156 multiple records are returned.
1158 38. The dnslists ACL condition has been extended: it is now possible to supply
1159 a list of IP addresses and/or domains to be looked up in a particular DNS
1162 39. Added log_selector=+queue_time_overall.
1164 40. When running the queue in the test harness, wait just a tad after forking a
1165 delivery process, to get repeatability of debugging output.
1167 41. Include certificate and key file names in error message when GnuTLS fails
1168 to set them up, because the GnuTLS error message doesn't include the name
1169 of the failing file when there is a problem reading it.
1171 42. Allow both -bf and -bF in the same test run.
1173 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1175 44. The "Exiscan patch" is now merged into the mainline Exim source.
1177 45. Sometimes the final signoff response after QUIT could fail to get
1178 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1179 < 0 before doing a fflush(). This bug looks as though it goes back to the
1180 introduction of TLS in release 3.20, but "sometimes" must have been rare
1181 because the tests only now provoked it.
1183 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1184 (this can affect the format of dates).
1186 47. exim_tidydb, when checking for the continued existence of a message for
1187 which it has found a message-specific retry record, was not finding
1188 messages that were in split spool directories. Consequently, it was
1189 deleting retry records that should have stayed in existence.
1191 48. Steve fixed some bugs in eximstats.
1193 49. The SPA authentication driver was not abandoning authentication and moving
1194 on to the next authenticator when an expansion was forced to fail,
1195 contradicting the general specification for all authenticators. Instead it
1196 was generating a temporary error. It now behaves as specified.
1198 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1199 (the order specifies the preference for clients). The order is now AES256,
1200 AES128, 3DES, ARCFOUR128.
1202 51. Small patch to Sieve code - explicitly set From: when generating an
1205 52. Exim crashed if a remote delivery caused a very long error message to be
1206 recorded - for instance if somebody sent an entire SpamAssassin report back
1207 as a large number of 550 error lines. This bug was coincidentally fixed by
1208 increasing the size of one of Exim's internal buffers (big_buffer) that
1209 happened as part of the Exiscan merge. However, to be on the safe side, I
1210 have made the code more robust (and fixed the comments that describe what
1213 53. Now that there can be additional text after "Completed" in log lines (if
1214 the queue_time_overall log selector is set), a one-byte patch to exigrep
1215 was needed to allow it to recognize "Completed" as not the last thing in
1218 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1219 patch that reportedly fixes this has been added. I am not expert enough to
1220 create a test for it. This is what the patch creator wrote:
1222 "I found a little strange behaviour of ldap code when working with
1223 Windows 2003 AD Domain, where users was placed in more than one
1224 Organization Units. When I tried to give exim partial DN, the exit code
1225 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1226 But simultaneously result of request was absolutely normal ldap result,
1227 so I produce this patch..."
1229 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1230 REFERENCE, so I have modified the code to exclude the patch when that macro
1233 55. Some experimental protocols are using DNS PTR records for new purposes. The
1234 keys for these records are domain names, not reversed IP addresses. The
1235 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1236 leaves it alone. Component reversal etc. now happens only for IP addresses.
1239 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1241 57. Double the size of the debug message buffer (to 2048) so that more of very
1242 long debug lines gets shown.
1244 58. The exicyclog utility now does better if the number of log files to keep
1245 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1247 59. Two changes related to the smtp_active_hostname option:
1249 (1) $smtp_active_hostname is now available as a variable.
1250 (2) The default for smtp_banner uses $smtp_active_hostname instead
1251 of $primary_hostname.
1253 60. The host_aton() function is supposed to be passed a string that is known
1254 to be a valid IP address. However, in the case of IPv6 addresses, it was
1255 not checking this. This is a hostage to fortune. Exim now panics and dies
1256 if the condition is not met. A case was found where this could be provoked
1257 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1258 components; fortuitously, this particular loophole had already been fixed
1259 by change 4.50/55 above.
1261 If there are any other similar loopholes, the new check in host_aton()
1262 itself should stop them being exploited. The report I received stated that
1263 data on the command line could provoke the exploit when Exim was running as
1264 exim, but did not say which command line option was involved. All I could
1265 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1266 running as the user.
1269 61. There was a buffer overflow vulnerability in the SPA authentication code
1270 (which came originally from the Samba project). I have added a test to the
1271 spa_base64_to_bits() function which I hope fixes it.
1274 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1275 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1277 63. The daemon start-up calls getloadavg() while still root for those OS that
1278 need the first call to be done as root, but it missed one case: when
1279 deliver_queue_load_max is set with deliver_drop_privilege. This is
1280 necessary for the benefit of the queue runner, because there is no re-exec
1281 when deliver_drop_privilege is set.
1283 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1284 This has been fixed.
1286 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1287 was in use, was not putting the data itself into the right store pool;
1288 consequently, it could be overwritten for a subsequent message in the same
1289 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1292 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1294 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1295 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1296 the function were treating the return as a boolean value, which happened to
1297 work because 0=false and not-0=true, but is not correct code.
1299 68. The host_aton() function was not handling scoped IPv6 addresses (those
1300 with, for example, "%eth0" on the end) correctly.
1302 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1303 negated items (that is, ~something) in unsigned ints. Some compilers
1304 apparently mutter when there is no cast.
1306 70. If an address verification called from an ACL failed, and did not produce a
1307 user-specific message (i.e. there was only a "system" message), nothing was
1308 put in $acl_verify_message. In this situation, it now puts the system
1311 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1312 session; change 4.31/43 added the unwanted input to the log line - except
1313 that it did not do this in the start of session case. It now does.
1315 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1316 This is wrong and can cause the other end to generate a synchronization
1317 error if it is another Exim or anything else that does the synchronization
1318 check. A QUIT command is no longer sent after a timeout.
1320 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1321 during host lookups.
1323 74. The defer_ok option of callout verification was not working if it was used
1324 when verifying addresses in header lines, that is, for this case:
1326 verify = header_sender/callout=defer_ok
1328 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1329 those file descriptors could be used for SMTP connections. If anything
1330 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1331 could be sent to the SMTP client, causing chaos. The daemon now opens
1332 stdin, stdout, and stderr to /dev/null when it puts itself into the
1335 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1336 log by default. The user can override this with suitable Perl magic.
1338 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1339 the log message when discard triggers, was not working for the DATA ACL or
1340 for the non-SMTP ACL.
1342 78. Error message wording change in sieve.c.
1344 79. If smtp_accept_max_per_host was set, the number of connections could be
1345 restricted to fewer than expected, because the daemon was trying to set up
1346 a new connection before checking whether the processes handling previous
1347 connections had finished. The check for completed processes is now done
1348 earlier. On busy systems, this bug wouldn't be noticed because something
1349 else would have woken the daemon, and it would have reaped the completed
1352 80. If a message was submitted locally by a user whose login name contained one
1353 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1354 It caused a spool format error. I have fixed the spool reading code. A
1355 related problem was that the "from" clause in the Received: line became
1356 illegal because of the space(s). It is now covered by ${quote_local_part.
1358 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1361 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1363 83. Patch from Sieve maintainer to fix unterminated string problem in
1364 "vacation" handling.
1366 84. Some minor changes to the Linux configuration files to help with other
1367 OS variants using glibc.
1369 85. One more patch for Sieve to update vacation handling to latest spec.
1372 ----------------------------------------------------
1373 See the note above about the 4.44 and 4.50 releases.
1374 ----------------------------------------------------
1380 1. Change 4.43/35 introduced a bug that caused file counts to be
1381 incorrectly computed when quota_filecount was set in an appendfile
1384 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1385 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1388 3. Give more explanation in the error message when the command for a transport
1389 filter fails to execute.
1391 4. There are several places where Exim runs a non-Exim command in a
1392 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1393 was being done only for the command run by the queryprogram router. It is
1394 now done for all such subprocesses. The other cases are: ${run, transport
1395 filters, and the commands run by the lmtp and pipe transports.
1397 5. Some older OS have a limit of 256 on the maximum number of file
1398 descriptors. Exim was using setrlimit() to set 1000 as a large value
1399 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1400 systems. I've change it so that if it can't get 1000, it tries for 256.
1402 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1403 was an oversight, and furthermore, ever since the addition of extra
1404 controls (e.g. 4.43/32), the checks on when to allow different forms of
1405 "control" were broken. There should now be diagnostics for all cases when a
1406 control that does not make sense is encountered.
1408 7. $recipients is now available in the predata ACL (oversight).
1410 8. Tidy the search cache before the fork to do a delivery from a message
1411 received from the command line. Otherwise the child will trigger a lookup
1412 failure and thereby defer the delivery if it tries to use (for example) a
1413 cached ldap connection that the parent has called unbind on.
1415 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1416 of $address_data from the recipient verification was clobbered by the
1417 sender verification.
1419 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1420 was its contents. (It was OK if the option was not defined at all.)
1422 11. A "Completed" log line is now written for messages that are removed from
1423 the spool by the -Mrm option.
1425 12. $host_address is now set to the target address during the checking of
1426 ignore_target_hosts.
1428 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1429 being passed; this would have caused $sender_host_name to have been used if
1430 matching the list had actually called for a host name (not very likely,
1431 since this list is usually IP addresses). A host name is now passed as
1434 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1435 code that specifies a non-restarting handler (typically sigaction() in
1436 modern systems) in an attempt to fix a rare and obscure crash bug.
1438 15. Narrowed the window for a race in the daemon that could cause it to ignore
1439 SIGCHLD signals. This is not a major problem, because they are used only to
1440 wake it up if nothing else does.
1442 16. A malformed maildirsize file could cause Exim to calculate negative values
1443 for the mailbox size or file count. Odd effects could occur as a result.
1444 The maildirsize information is now recalculated if the size or filecount
1447 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1448 support for a long time. Removed HAVE_SYS_VFS_H.
1450 18. Updated exipick to current release from John Jetmore.
1452 19. Allow an empty sender to be matched against a lookup in an address list.
1453 Previously the only cases considered were a regular expression, or an
1456 20. Exim went into a mad DNS lookup loop when doing a callout where the
1457 host was specified on the transport, if the DNS lookup yielded more than
1460 21. The RFC2047 encoding function was originally intended for short strings
1461 such as real names; it was not keeping to the 75-character limit for
1462 encoded words that the RFC imposes. It now respects the limit, and
1463 generates multiple encoded words if necessary. To be on the safe side, I
1464 have increased the buffer size for the ${rfc2047: expansion operator from
1467 22. Failure to deliver a bounce message always caused it to be frozen, even if
1468 there was an errors_to setting on the router. The errors_to setting is now
1471 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1472 canonical form (fully expanded) before being placed in
1473 $sender_host_address.
1475 24. Updated eximstats to version 1.33
1477 25. Include certificate and key file names in error message when GnuTLS fails
1478 to set them up, because the GnuTLS error message doesn't include the name
1479 of the failing file when there is a problem reading it.
1481 26. Expand error message when OpenSSL has problems setting up cert/key files.
1484 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1485 (this can affect the format of dates).
1487 28. exim_tidydb, when checking for the continued existence of a message for
1488 which it has found a message-specific retry record, was not finding
1489 messages that were in split spool directories. Consequently, it was
1490 deleting retry records that should have stayed in existence.
1492 29. eximstats updated to version 1.35
1493 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1494 1.35 - bugfix such that pie charts by volume are generated correctly
1496 30. The SPA authentication driver was not abandoning authentication and moving
1497 on to the next authenticator when an expansion was forced to fail,
1498 contradicting the general specification for all authenticators. Instead it
1499 was generating a temporary error. It now behaves as specified.
1501 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1502 (the order specifies the preference for clients). The order is now AES256,
1503 AES128, 3DES, ARCFOUR128.
1505 31. Small patch to Sieve code - explicitly set From: when generating an
1508 32. Exim crashed if a remote delivery caused a very long error message to be
1509 recorded - for instance if somebody sent an entire SpamAssassin report back
1510 as a large number of 550 error lines. This bug was coincidentally fixed by
1511 increasing the size of one of Exim's internal buffers (big_buffer) that
1512 happened as part of the Exiscan merge. However, to be on the safe side, I
1513 have made the code more robust (and fixed the comments that describe what
1516 33. Some experimental protocols are using DNS PTR records for new purposes. The
1517 keys for these records are domain names, not reversed IP addresses. The
1518 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1519 leaves it alone. Component reversal etc. now happens only for IP addresses.
1522 34. The host_aton() function is supposed to be passed a string that is known
1523 to be a valid IP address. However, in the case of IPv6 addresses, it was
1524 not checking this. This is a hostage to fortune. Exim now panics and dies
1525 if the condition is not met. A case was found where this could be provoked
1526 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1527 components; fortuitously, this particular loophole had already been fixed
1528 by change 4.50/55 or 4.44/33 above.
1530 If there are any other similar loopholes, the new check in host_aton()
1531 itself should stop them being exploited. The report I received stated that
1532 data on the command line could provoke the exploit when Exim was running as
1533 exim, but did not say which command line option was involved. All I could
1534 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1535 running as the user.
1538 35. There was a buffer overflow vulnerability in the SPA authentication code
1539 (which came originally from the Samba project). I have added a test to the
1540 spa_base64_to_bits() function which I hope fixes it.
1543 36. The daemon start-up calls getloadavg() while still root for those OS that
1544 need the first call to be done as root, but it missed one case: when
1545 deliver_queue_load_max is set with deliver_drop_privilege. This is
1546 necessary for the benefit of the queue runner, because there is no re-exec
1547 when deliver_drop_privilege is set.
1549 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1550 was in use, was not putting the data itself into the right store pool;
1551 consequently, it could be overwritten for a subsequent message in the same
1552 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1555 38. Sometimes the final signoff response after QUIT could fail to get
1556 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1557 < 0 before doing a fflush(). This bug looks as though it goes back to the
1558 introduction of TLS in release 3.20, but "sometimes" must have been rare
1559 because the tests only now provoked it.
1565 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1566 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1567 Now it returns an integer. A number of calls were still expecting a T/F
1568 return. Fortuitously, in all cases, the tests worked in OK situations,
1569 which is the norm. However, things would have gone wrong on any write
1570 failures on the smtp file descriptor. This function is used when sending
1571 messages over SMTP and also when doing verify callouts.
1573 2. When Exim is called to do synchronous delivery of a locally submitted
1574 message (the -odf or -odi options), it no longer closes stderr before doing
1577 3. Implemented the mua_wrapper option.
1579 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1581 5. Implemented the functions header_remove(), header_testname(),
1582 header_add_at_position(), and receive_remove_recipient(), and exported them
1585 6. If an ACL "warn" statement specified the addition of headers, Exim already
1586 inserted X-ACL-Warn: at the start if there was no header name. However, it
1587 was not making this test for the second and subsequent header lines if
1588 there were newlines in the string. This meant that an invalid header could
1589 be inserted if Exim was badly configured.
1591 7. Allow an ACL "warn" statement to add header lines at the start or after all
1592 the Received: headers, as well as at the end.
1594 8. Added the rcpt_4xx retry error code.
1596 9. Added postmaster_mailfrom=xxx to callout verification option.
1598 10. Added mailfrom=xxxx to the callout verification option, for verify=
1601 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1602 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1604 12. Inserted some casts to stop certain compilers warning when using pointer
1605 differences as field lengths or precisions in printf-type calls (mostly
1606 affecting debugging statements).
1608 13. Added optional readline() support for -be (dynamically loaded).
1610 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1611 same clock tick as a message's arrival, so that its received time was the
1612 same as the "first fail" time on the retry record, and that message
1613 remained on the queue past the ultimate address timeout, every queue runner
1614 would try a delivery (because it was past the ultimate address timeout) but
1615 after another failure, the ultimate address timeout, which should have then
1616 bounced the address, did not kick in. This was a "< instead of <=" error;
1617 in most cases the first failure would have been in the next clock tick
1618 after the received time, and all would be well.
1620 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1621 being recognized when the domain list was tested by the match_domain
1622 condition in an expansion string.
1624 16. Added the ${str2b64: operator.
1626 17. Exim was always calling setrlimit() to set a large limit for the number of
1627 processes, without checking whether the existing limit was already
1628 adequate. (It did check for the limit on file descriptors.) Furthermore,
1629 errors from getrlimit() and setrlimit() were being ignored. Now they are
1630 logged to the main and panic logs, but Exim does carry on, to try to do its
1631 job under whatever limits there are.
1633 18. Imported PCRE 5.0.
1635 19. Trivial typo in log message " temporarily refused connection" (the leading
1638 20. If the log selector return_path_on_delivery was set and an address was
1639 redirected to /dev/null, the delivery process crashed because it assumed
1640 that a return path would always be set for a "successful" delivery. In this
1641 case, the whole delivery is bypassed as an optimization, and therefore no
1644 21. Internal re-arrangement: the function for sending a challenge and reading
1645 a response while authentication was assuming a zero-terminated challenge
1646 string. It's now changed to take a pointer and a length, to allow for
1647 binary data in such strings.
1649 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1651 23. Exim was not respecting finduser_retries when seeking the login of the
1652 uid under which it was called; it was always trying 10 times. (The default
1653 setting of finduser_retries is zero.) Also, it was sleeping after the final
1654 failure, which is pointless.
1656 24. Implemented tls_on_connect_ports.
1658 25. Implemented acl_smtp_predata.
1660 26. If the domain in control=submission is set empty, Exim assumes that the
1661 authenticated id is a complete email address when it generates From: or
1662 Sender: header lines.
1664 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1665 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1666 chown and chgrp in /bin and hostname in /usr/bin.
1668 28. Exim was keeping the "process log" file open after each use, just as it
1669 does for the main log. This opens the possibility of it remaining open for
1670 long periods when the USR1 signal hits a daemon. Occasional processlog
1671 errors were reported, that could have been caused by this. Anyway, it seems
1672 much more sensible not to leave this file open at all, so that is what now
1675 29. The long-running daemon process does not normally write to the log once it
1676 has entered its main loop, and it closes the log before doing so. This is
1677 so that log files can straightforwardly be renamed and moved. However,
1678 there are a couple of unusual error situations where the daemon does write
1679 log entries, and I had neglected to close the log afterwards.
1681 30. The text of an SMTP error response that was received during a remote
1682 delivery was being truncated at 512 bytes. This is too short for some of
1683 the long messages that one sometimes sees. I've increased the limit to
1686 31. It is now possible to make retry rules that apply only when a message has a
1687 specific sender, in particular, an empty sender.
1689 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1690 it possible to be selective about when SMTP synchronization is enforced.
1692 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1694 32. Implemented hosts_connection_nolog.
1696 33. Added an ACL for QUIT.
1698 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1701 35. Added mailbox_size and mailbox_filecount to appendfile.
1703 36. Added control = no_multiline_responses to ACLs.
1705 37. There was a bug in the logic of the code that waits for the clock to tick
1706 in the case where the clock went backwards by a substantial amount such
1707 that the microsecond fraction of "now" was more than the microsecond
1708 fraction of "then" (but the whole seconds number was less).
1710 38. Added support for the libradius Radius client library this is found on
1711 FreeBSD (previously only the radiusclient library was supported).
1717 1. When certain lookups returned multiple values in the form name=value, the
1718 quoting of the values was not always being done properly. Specifically:
1719 (a) If the value started with a double quote, but contained no whitespace,
1721 (b) If the value contained whitespace other than a space character (i.e.
1722 tabs or newlines or carriage returns) it was not quoted.
1723 This fix has been applied to the mysql and pgsql lookups by writing a
1724 separate quoting function and calling it from the lookup code. The fix
1725 should probably also be applied to nisplus, ibase and oracle lookups, but
1726 since I cannot test any of those, I have not disturbed their existing code.
1728 2. A hit in the callout cache for a specific address caused a log line with no
1729 reason for rejecting RCPT. Now it says "Previous (cached) callout
1730 verification failure".
1732 3. There was an off-by-one bug in the queryprogram router. An over-long
1733 return line was truncated at 256 instead of 255 characters, thereby
1734 overflowing its buffer with the terminating zero. As well as fixing this, I
1735 have increased the buffer size to 1024 (and made a note to document this).
1737 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1738 when Exim is waiting for an SMTP response from a remote server, Exim
1739 restarts its select() call on the socket, thereby resetting its timeout.
1740 This is not a problem when such interrupts are rare. Somebody set up a cron
1741 job to run exiwhat every 2 minutes, which is less than the normal select()
1742 timeout (5 or 10 minutes). This meant that the select() timeout never
1743 kicked in because it was always reset. I have fixed this by comparing the
1744 time when an interrupt arrives with the time at the start of the first call
1745 to select(). If more time than the timeout has elapsed, the interrupt is
1746 treated as a timeout.
1748 5. Some internal re-factoring in preparation for the addition of Sieve
1749 extensions (by MH). In particular, the "personal" test is moved to a
1750 separate function, and given an option for scanning Cc: and Bcc: (which is
1751 not set for Exim filters).
1753 6. When Exim created an email address using the login of the caller as the
1754 local part (e.g. when creating a From: or Sender: header line), it was not
1755 quoting the local part when it contained special characters such as @.
1757 7. Installed new OpenBSD configuration files.
1759 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1760 try to make them clearer.
1762 9. Callout options, other than the timeout value, were being ignored when
1763 verifying sender addresses in header lines. For example, when using
1765 verify = header_sender/callout=no_cache
1767 the cache was (incorrectly) being used.
1769 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1770 only the Cygwin environment.
1772 11. When return_path_on_delivery was set as a log selector, if different remote
1773 addresses in the same message used different return paths and parallel
1774 remote delivery occurred, the wrong values would sometimes be logged.
1775 (Whenever a remote delivery process finished, the return path value from
1776 the most recently started remote delivery process was logged.)
1778 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1779 lines when AUTH and/or TLS are in use. This is the "received protocol"
1780 field. Exim used to use "asmtp" for authenticated SMTP, without any
1781 indication (in the protocol name) for TLS use. Now it follows the RFC and
1782 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1783 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1784 names appear in log lines as well as in Received: header lines.
1786 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1787 extensions, and comparison tests, and to fix some bugs.
1789 14. Changes to the "personal" filter test:
1791 (1) The test was buggy in that it was just doing the equivalent of
1792 "contains" tests on header lines. For example, if a user's address was
1793 anne@some.where, the "personal" test would incorrectly be true for
1795 To: susanne@some.where
1797 This test is now done by extracting each address from the header in turn,
1798 and checking the entire address. Other tests that are part of "personal"
1799 are now done using regular expressions (for example, to check local parts
1800 of addresses in From: header lines).
1802 (2) The list of non-personal local parts in From: addresses has been
1803 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1804 taken from the Sieve specification recommendations.
1806 (3) If the message contains any header line starting with "List-" it is
1807 treated as non-personal.
1809 (4) The test for "circular" in the Subject: header line has been removed
1810 because it now seems ill-conceived.
1812 15. Minor typos in src/EDITME comments corrected.
1814 16. Installed latest exipick from John Jetmore.
1816 17. If headers_add on a router specified a text string that was too long for
1817 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1818 of string_sprintf() is now avoided.
1820 18. $message_body_size was not set (it was always zero) when running the DATA
1821 ACL and the local_scan() function.
1823 19. For the "mail" command in an Exim filter, no default was being set for
1824 the once_repeat time, causing a random time value to be used if "once" was
1825 specified. (If the value happened to be <= 0, no repeat happened.) The
1826 default is now 0s, meaning "never repeat". The "vacation" command was OK
1827 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1828 (I found it when inspecting the code).
1830 20. There is now an overall timeout for performing a callout verification. It
1831 defaults to 4 times the callout timeout, which applies to individual SMTP
1832 commands during the callout. The overall timeout applies when there is more
1833 than one host that can be tried. The timeout is checked before trying the
1834 next host. This prevents very long delays if there are a large number of
1835 hosts and all are timing out (e.g. when the network connections are timing
1836 out). The value of the overall timeout can be changed by specifying an
1837 additional sub-option for "callout", called "maxwait". For example:
1839 verify = sender/callout=5s,maxwait=20s
1841 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1842 to the end before writing, but this should make it even safer).
1844 22. Exim was forgetting that it had advertised PIPELINING for the second and
1845 subsequent messages on an SMTP connection. It was also not resetting its
1846 memory on STARTTLS and an internal HELO.
1848 23. When Exim logs an SMTP synchronization error within a session, it now
1849 records whether PIPELINING has been advertised or not.
1851 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1852 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1853 rather than long int.
1855 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1857 26. Added the never_mail option to autoreply.
1863 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1864 crash if the getsockname() call failed; this can happen if a connection is
1865 closed very soon after it is established. The problem was simply in the
1866 order in which certain operations were done, causing Exim to try to write
1867 to the SMTP stream before it had set up the file descriptor. The bug has
1868 been fixed by making things happen in the correct order.
1874 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1875 before the connection was closed, thus losing the rejection response.
1877 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1878 some early Solaris releases, but causes trouble in current releases where
1879 socklen_t is defined.
1881 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1884 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1887 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1888 that releasing the top of it at the end releases what was used for sub-
1889 expansions (unless the block got too big). However, discard this block if
1890 the first thing is a variable or header, so that we can use its block when
1891 it is dynamic (useful for very large $message_headers, for example).
1893 6. Lookups now cache *every* query, not just the most recent. A new, separate
1894 store pool is used for this. It can be recovered when all lookup caches are
1895 flushed. Lookups now release memory at the end of their result strings.
1896 This has involved some general refactoring of the lookup sources.
1898 7. Some code has been added to the store_xxx() functions to reduce the amount
1899 of flapping under certain conditions.
1901 8. log_incoming_interface used to affect only the <= reception log lines. Now
1902 it causes the local interface and port to be added to several more SMTP log
1903 lines, for example "SMTP connection from", and rejection lines.
1905 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1907 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1909 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1910 could be overwritten at the end of the current message (or the start of a
1911 new message if it was set in a HELO ACL). The value is now preserved for
1912 the duration of the SMTP connection.
1914 12. If a transport had a headers_rewrite setting, and a matching header line
1915 contained an unqualified address, that address was qualified, even if it
1916 did not match any rewriting rules. The underlying bug was that the values
1917 of the flags that permit the existence of unqualified sender and recipient
1918 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1919 non-local messages, and by -bnq for local messages) were not being
1920 preserved with the message after it was received.
1922 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1923 "next input=" as part of the text comprising the host identity instead of
1924 the correct text. The code was using the same buffer for two different
1925 strings. However, depending on which order the printing function evaluated
1926 its arguments, the bug did not always show up. Under Linux, for example, my
1927 test suite worked just fine.
1929 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1930 doesn't work with some older versions of Perl. It has been changed to "my",
1931 which in any case is probably the better facility to use.
1933 15. A really picky compiler found some instances of statements for creating
1934 error messages that either had too many or two few arguments for the format
1937 16. The size of the buffer for calls to the DNS resolver has been increased
1938 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1939 for addresses that have a lot of PTR records. This alleviates a problem; it
1940 does not fully solve it.
1942 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1943 buffer now truncates the list and logs the incident, which is the same
1944 action as happens when Exim is looking up a host name and its aliases.
1945 Previously in this situation something unpredictable would happen;
1946 sometimes it was "internal error: store_reset failed".
1948 18. If a server dropped the connection unexpectedly when an Exim client was
1949 using GnuTLS and trying to read a response, the client delivery process
1950 crashed while trying to generate an error log message.
1952 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1953 string, for example:
1955 warn message = H1: something\nH2: something
1957 the text was added as a single header line from Exim's point of view
1958 though it ended up OK in the delivered message. However, searching for the
1959 second and subsequent header lines using $h_h2: did not work. This has been
1960 fixed. Similarly, if a system filter added multiple headers in this way,
1961 the routers could not see them.
1963 20. Expanded the error message when iplsearch is called with an invalid key to
1964 suggest using net-iplsearch in a host list.
1966 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1967 ACLs are no longer actually imposed (and a message to that effect is
1970 22. If a "gecos" field in a passwd entry contained escaped characters, in
1971 particular, if it contained a \" sequence, Exim got it wrong when building
1972 a From: or a Sender: header from that name. A second bug also caused
1973 incorrect handling when an unquoted " was present following a character
1974 that needed quoting.
1976 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1977 was not being matched caselessly.
1979 24. Arranged for all hyphens in the exim.8 source to be escaped with
1982 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1983 properly. Recipient callout cache records were still being keyed to include
1984 the sender, even when use_sender was set false. This led to far more
1985 callouts that were necessary. The sender is no longer included in the key
1986 when use_sender is false.
1988 26. Added "control = submission" modifier to ACLs.
1990 27. Added the ${base62d: operator to decode base 62 numbers.
1992 28. dnsdb lookups can now access SRV records.
1994 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1995 the configuration file.
1997 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1998 (-v) mode. This makes the output for a verbose queue run more intelligible.
2000 31. Added a use_postmaster feature to recipient callouts.
2002 32. Added the $body_zerocount variable, containing the number of binary zero
2003 bytes in the message body.
2005 33. The time of last modification of the "new" subdirectory is now used as the
2006 "mailbox time last read" when there is a quota error for a maildir
2009 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2011 35. Added +ignore_unknown as a special item in host lists.
2013 36. Code for decoding IPv6 addresses in host lists is now included, even if
2014 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2015 address was recognized as an IP address, but was then not correctly decoded
2016 into binary, causing unexpected and incorrect effects when compared with
2023 1. Very minor rewording of debugging text in manualroute to say "list of
2024 hosts" instead of "hostlist".
2026 2. If verify=header_syntax was set, and a header line with an unqualified
2027 address (no domain) and a large number of spaces between the end of the
2028 name and the colon was received, the reception process suffered a buffer
2029 overflow, and (when I tested it) crashed. This was caused by some obsolete
2030 code that should have been removed. The fix is to remove it!
2032 3. When running in the test harness, delay a bit after writing a bounce
2033 message to get a bit more predictability in the log output.
2035 4. Added a call to search_tidyup() just before forking a reception process. In
2036 theory, someone could use a lookup in the expansion of smtp_accept_max_
2037 per_host which, without the tidyup, could leave open a database connection.
2039 5. Added the variables $recipient_data and $sender_data which get set from a
2040 lookup success in an ACL "recipients" or "senders" condition, or a router
2041 "senders" option, similar to $domain_data and $local_part_data.
2043 6. Moved the writing of debug_print from before to after the "senders" test
2046 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2047 problems for message scanning, either using a data ACL, or using
2048 local_scan() because the Received: header was not generated till after they
2049 were called (in order to set the time as the time of reception completion).
2050 I have revised the way this works. The header is now generated after the
2051 body is received, but before the ACL or local_scan() are called. After they
2052 are run, the timestamp in the header is updated.
2058 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2059 before starting a queue runner without re-exec. This happened only when
2060 deliver_drop_privilege was set or when the Exim user was set to root. The
2061 effect of the bug was that timeouts during subsequent deliveries caused
2062 crashes instead of being properly handled. The handler is now left at its
2063 default (and expected) setting.
2065 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2066 message, again when deliver_drop_privilege is set or Exim is run as root.
2067 The bug described in (1) was not present in this case, but the tidying up
2068 of the other signals was missing. I have made the two cases consistent.
2070 3. The ignore_target_hosts setting on a manualroute router was being ignored
2071 for hosts that were looked up using the /MX notation.
2073 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2076 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2077 operated on the sender address. After changing the $sender_address to <>
2078 for the sender address verify, Exim was re-instated it as the original
2079 (before rewriting) address, but remembering that it had rewritten it, so it
2080 wasn't rewriting it again. This bug also had the effect of breaking the
2081 sender address verification caching when the sender address was rewritten.
2083 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2084 This has been changed so that if the ip literal address matches
2085 ignore_target_hosts, the router declines.
2087 7. Added expansion conditions match_domain, match_address, and match_local_
2088 part (NOT match_host).
2090 8. The placeholder for the Received: header didn't have a length field set.
2092 9. Added code to Exim itself and to exim_lock to test for a specific race
2093 condition that could lead to file corruption when using MBX delivery. The
2094 issue is with the lockfile that is created in /tmp. If this file is removed
2095 after a process has opened it but before that process has acquired a lock,
2096 there is the potential for a second process to recreate the file and also
2097 acquire a lock. This could lead to two Exim processes writing to the file
2098 at the same time. The added code performs the same test as UW imapd; it
2099 checks after acquiring the lock that its file descriptor still refers to
2100 the same named file.
2102 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2103 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2106 11. Added the smtp_active_hostname option. If used, this will typically be made
2107 to depend on the incoming interface address. Because $interface_address is
2108 not set up until the daemon has forked a reception process, error responses
2109 that can happen earlier (such as "too many connections") no longer contain
2112 12. If an expansion in a condition on a "warn" statement fails because a lookup
2113 defers, the "warn" statement is abandoned, and the next ACL statement is
2114 processed. Previously this caused the whole ACL to be aborted.
2116 13. Added the iplsearch lookup type.
2118 14. Added ident_timeout as a log selector.
2120 15. Added tls_certificate_verified as a log selector.
2122 16. Added a global option tls_require_ciphers (compare the smtp transport
2123 option of the same name). This controls incoming TLS connections.
2125 17. I finally figured out how to make tls_require_ciphers do a similar thing
2126 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2127 before starting the TLS session.
2129 18. Tabs are now shown as \t in -bP output.
2131 19. If the log selector return_path_on_delivery was set, Exim crashed when
2132 bouncing a message because it had too many Received: header lines.
2134 20. If two routers both had headers_remove settings, and the first one included
2135 a superfluous trailing colon, the final name in the first list and the
2136 first name in the second list were incorrectly joined into one item (with a
2137 colon in the middle).
2143 1. Added -C and -D options to the exinext utility, mainly to make it easier
2144 to include in the automated testing, but these could be helpful when
2145 multiple configurations are in use.
2147 2. The exinext utility was not formatting the output nicely when there was
2148 an alternate port involved in the retry record key, nor when there was a
2149 message id as well (for retries that were specific to a specific message
2150 and a specific host). It was also confused by IPv6 addresses, because of
2151 the additional colons they contain. I have fixed the IPv4 problem, and
2152 patched it up to do a reasonable job for IPv6.
2154 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2155 delivery, the log line now contains "pipelined" if PIPELINING was used.
2157 4. An SMTP transport process used to panic and die if the bind() call to set
2158 an explicit outgoing interface failed. This has been changed; it is now
2159 treated in the same way as a connect() failure.
2161 5. A reference to $sender_host_name in the part of a conditional expansion
2162 that was being skipped was still causing a DNS lookup. This no longer
2165 6. The def: expansion condition was not recognizing references to header lines
2166 that used bh_ and bheader_.
2168 7. Added the _cache feature to named lists.
2170 8. The code for checking quota_filecount in the appendfile transport was
2171 allowing one more file than it should have been.
2173 9. For compatibility with Sendmail, the command line option
2181 and sets the incoming protocol and host name (for trusted callers). The
2182 host name and its colon can be omitted when only the protocol is to be set.
2183 Note the Exim already has two private options, -pd and -ps, that refer to
2184 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2185 "s", but I don't think that's a major issue.
2187 10. A number of refactoring changes to the code, none of which should affect
2190 (a) The number of logging options was getting close to filling up the
2191 32-bit word that was used as a bit map. I have split them into two classes:
2192 those that are passed in the argument to log_write(), and those that are
2193 only ever tested independently outside of that function. These are now in
2194 separate 32-bit words, so there is plenty of room for expansion again.
2195 There is no change in the user interface or the logging behaviour.
2197 (b) When building, for example, log lines, the code previously used a
2198 macro that called string_cat() twice, in order to add two strings. This is
2199 not really sufficiently general. Furthermore, there was one instance where
2200 it was actually wrong because one of the argument was used twice, and in
2201 one call a function was used. (As it happened, calling the function twice
2202 did not affect the overall behaviour.) The macro has been replaced by a
2203 function that can join an arbitrary number of extra strings onto a growing
2206 (c) The code for expansion conditions now uses a table and a binary chop
2207 instead of a serial search (which was left over from when there were very
2208 few conditions). Also, it now recognizes conditions like "pam" even when
2209 the relevant support is not compiled in: a suitably worded error message is
2210 given if an attempt is made to use such a condition.
2212 11. Added ${time_interval:xxxxx}.
2214 12. A bug was causing one of the ddress fields not to be passed back correctly
2215 from remote delivery subprocesses. The field in question was not being
2216 subsequently used, so this caused to problems in practice.
2218 13. Added new log selectors queue_time and deliver_time.
2220 14. Might have fixed a bug in maildirsizefile handling that threw up
2221 "unexpected character" debug warnings, and recalculated the data
2222 unnecessarily. In any case, I expanded the warning message to give more
2225 15. Added the message "Restricted characters in address" to the statements in
2226 the default ACL that block characters like @ and % in local parts.
2228 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2229 Three changes have been made:
2231 (a) There was a serious bug; a negative response to MAIL caused the whole
2232 recipient domain to be cached as invalid, thereby blocking all messages
2233 to all local parts at the same domain, from all senders. This bug has
2234 been fixed. The domain is no longer cached after a negative response to
2235 MAIL if the sender used is not empty.
2237 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2240 (c) A new callout option, "use_sender" has been added for people who want
2241 the modified behaviour.
2247 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2250 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2251 indeed breaks things for older releases.
2253 3. Added additional logging to the case where there is a problem reading data
2254 from a filter that is running in a subprocess using a pipe, in order to
2255 try to track down a specific problem.
2257 4. Testing facility fudge: when running in the test harness and attempting
2258 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2259 getting "No route to host". Convert this to a timeout.
2261 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2264 6. Some OS don't have socklen_t but use size_t instead. This affects the
2265 fifth argument of getsockopt() amongst other things. This is now
2266 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2267 can be set for individual OS. I have set it for SunOS5, OSF1, and
2268 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2269 some earlier ones do not.
2271 7. Change 4.30/15 was not doing the test caselessly.
2273 8. The standard form for an IPv6 address literal was being rejected by address
2274 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2275 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2276 this, as well as the form without the "IPv6" on the front (but only when
2277 address literals are enabled, of course).
2279 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2281 10. Exim crashed if a message with an empty sender address specified by -f
2282 encountered a router with an errors_to setting. This could be provoked only
2283 by a command such as
2287 where an empty string was supplied; "<>" did not hit this bug.
2289 11. Installed PCRE release 4.5.
2291 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2292 remained set. It is now erased.
2294 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2295 times from message ids (which are base 36 rather than the normal 62).
2297 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2298 were being counted as actual protocol errors, and logged if the log
2299 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2300 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2301 and DATA following a set of rejected RCPTs do not count as protocol errors.
2302 In other words, Exim assumes they were pipelined, though this may not
2303 actually be the case. Of course, in all cases the client gets an
2304 appropriate error code.
2306 15. If a lookup fails in an ACL condition, a message about the failure may
2307 be available; it is used if testing the ACL cannot continue, because most
2308 such messages specify what the cause of the deferral is. However, some
2309 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2310 that caused an old message to be retained and used if a later statement
2311 caused a defer, replacing the real cause of the deferral.
2313 16. If an IP address had so many PTR records that the DNS lookup buffer
2314 was not large enough to hold them, Exim could crash while trying to process
2315 the truncated data. It now detects and logs this case.
2317 17. Further to 4.21/58, another change has been made: if (and only if) the
2318 first line of a message (the first header line) ends with CRLF, a bare LF
2319 in a subsequent header line has a space inserted after it, so as not to
2320 terminate the header.
2322 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2323 unnecessarily, used atoi() instead of strtol(), and didn't check the
2324 termination when getting file sizes from file names by regex.
2326 19. Completely re-implemented the support for maildirsize files, in the light
2327 of a number of problems with the previous contributed implementation
2328 (4.30/29). In particular:
2330 . If the quota is zero, the maildirsize file is maintained, but no quota is
2333 . If the maildir directory does not exist, it is created before any attempt
2334 to write a maildirsize file.
2336 . The quota value in the file is just a cache; if the quota is changed in
2337 the transport, the new value overrides.
2339 . A regular expression is available for excluding directories from the
2342 20. The autoreply transport checks the characters in options that define the
2343 message's headers; it allows continued headers, but it was checking with
2344 isspace() after an embedded newline instead of explicitly looking for a
2347 21. If all the "regular" hosts to which an address was routed had passed their
2348 expiry times, and had not reached their retry times, the address was
2349 bounced, even if fallback hosts were defined. Now Exim should go on to try
2352 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2353 equivalent code in the SMTP transport. Some hosts send humungous responses
2354 to HELO/EHLO, more than 1024 it seems.
2356 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2357 gives compiler warnings in some environments. I've now done it "properly",
2360 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2361 (because of problems with the built-in one) was declared to return uschar *
2362 instead of char *, causing compiler failure.
2364 25. Fixed a file descriptor leak when processing alias/forward files.
2366 26. Fixed a minor format string issue in dbfn.c.
2368 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2370 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2371 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2374 29. When a sender address is verified, it is cached, to save repeating the test
2375 when there is more than one recipient in a message. However, when the
2376 verification involves a callout, it is possible for different callout
2377 options to be set for different recipients. It is too complicated to keep
2378 track of this in the cache, so now Exim always runs a verification when a
2379 callout is required, relying on the callout cache for the optimization.
2380 The overhead is duplication of the address routing, but this should not be
2383 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2384 to be verified with callout=postmaster, and the main callout worked but the
2385 postmaster check failed, the verification correctly failed. However, if a
2386 subsequent RCPT command asked for sender verification *without* the
2387 postmaster check, incorrect caching caused this verification also to fail,
2390 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2391 it was not caching the DNS options (qualify_single, search_parents) that
2392 were used when the lookup failed. A subsequent lookup with different
2393 options therefore always gave the same answer, though there were cases
2394 where it should not have. (Example: a "domains = !$mx_any" option on a
2395 dnslookup router: the "domains" option is always processed without any
2396 widening, but the router might have qualify_single set.) Now Exim uses the
2397 cached value only when the same options are set.
2399 32. Added John Jetmore's "exipick" utility to the distribution.
2401 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2402 than a timeout (e.g. a certificate is required, and is not provided), an
2403 Exim server now closes the connection immediately. Previously it waited for
2404 the client to close - but if the client is SSL, it seems that they each
2405 wait for each other, leading to a delay before one of them times out.
2407 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2408 maintained 0.8.x compatibility because I don't think many are using it, and
2409 it is clearly obsolete.
2411 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2414 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2415 client certificate was expired. A simple patch fixes this, though I don't
2416 understand the full logic of why the verify callback is called multiple
2419 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2420 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2421 which causes problems with some clients (such as the Certicom SSL Plus
2422 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2423 disables the coutermeasure allowing Eudora to connect."
2425 38. Exim was not checking that a write() to a log file succeeded. This could
2426 lead to Bad Things if a log got too big, in particular if it hit a file
2427 size limit. Exim now panics and dies if it cannot write to a log file, just
2428 as it does if it cannot open a log file.
2430 39. Modified OS/Makefile-Linux so that it now contains
2432 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2434 The two -D definitions ensure that Exim is compiled with large file
2435 support, which makes it possible to handle log files that are bigger than
2438 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2439 instance) a domain was checked against a named list that involved a lookup,
2440 causing $domain_data to be set, then another domain was checked against the
2441 same list, then the first domain was re-checked, the value of $domain_data
2442 after the final check could be wrong. In particular, if the second check
2443 failed, it could be set empty. This bug probably also applied to
2446 41. The strip_trailing_dot option was not being applied to the address given
2447 with the -f command-line option.
2449 42. The code for reading a message's header from the spool was incrementing
2450 $received_count, but never initializing it. This meant that the value was
2451 incorrect (doubled) while delivering a message in the same process in which
2452 it was received. In the most common configuration of Exim, this never
2453 happens - a fresh exec is done - but it can happen when
2454 deliver_drop_privilege is set.
2456 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2457 it now includes up to 150 characters of the unexpected data in the log
2460 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2461 and building data strings. The size of both of these buffers was 10 000
2462 bytes - far larger than anybody would *ever* want, thought I. Needless to
2463 say, somebody hit the limit. I have increased the maximum line length to
2464 20 000 and the maximum data length of concatenated lines to 100 000. I have
2465 also fixed two bugs, because there was no checking on these buffers. Tsk,
2466 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2467 buffer is too small.
2469 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2470 lsearch lookups. Now it does.
2472 46. When parsing a route_list item in a manualroute router, a fixed-length
2473 buffer was used for the list of hosts. I made this 1024 bytes long,
2474 thinking that nobody would ever have a list of hosts that long. Wrong.
2475 Somebody had a whole pile of complicated expansion conditions, and the
2476 string was silently truncated, leading to an expansion error. It turns out
2477 that it is easier to change to an unlimited length (owing to other changes
2478 that have happened since this code was originally written) than to build
2479 structure for giving a limitation error. The length of the item that
2480 expands into the list of hosts is now unlimited.
2482 47. The lsearch lookup could not handle data where the length of text line was
2483 more than 4095 characters. Such lines were truncated, leading to shortened
2484 data being returned. It should now handle lines of any length.
2486 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2487 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2490 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2491 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2492 no longer happens in comment lines. A list of replacements is now placed
2493 at the head of all of the source files, except those whose only change is
2494 to replace PERL_COMMAND in the very first #! line.
2496 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2497 messages on the queue, with a bottom-up merge sort, using code contributed
2498 by Michael Haardt. This should make operations like -bp somewhat faster on
2499 large queues. It won't affect queue runners, except when queue_run_in_order
2502 51. Installed eximstats 1.31 in the distribution.
2504 52. Added support for SRV lookups to the dnslookup router.
2506 53. If an ACL referred to $message_body or $message_body_end, the value was not
2507 reset for any messages that followed in the same SMTP session.
2509 54. The store-handling optimization for building very long strings was not
2510 differentiating between the different store pools. I don't think this
2511 actually made any difference in practice, but I've tidied it.
2513 55. While running the routers to verify a sender address, $sender_address
2514 was still set to the sender address. This is wrong, because when routing to
2515 send a bounce to the sender, it would be empty. Therefore, I have changed
2516 it so that, while verifying a sender address, $sender_address is set to <>.
2517 (There is no change to what happens when verifying a recipient address.)
2519 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2520 target A or AAAA records (if not already returned) without resetting the
2521 qualify_single or search_parents options of the DNS resolver. These are
2522 inappropriate in this case because the targets of MX and SRV records must
2523 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2524 target that, when qualified, matched something in the local domain. These
2525 two options are now turned off when doing these lookups.
2527 57. It seems that at least some releases of Reiserfs (which does not have the
2528 concept of a fixed number of inodes) returns zero and not -1 for the
2529 number of available inodes. This interacted badly with check_spool_inodes,
2530 which assumed that -1 was the "no such thing" setting. What I have done is
2531 to check that the total number of inodes is greater than zero before doing
2532 the test of how many are available.
2534 58. When a "warn" ACL statement has a log_message modifier, the message is
2535 remembered, and not repeated. This is to avoid a lot of repetition when a
2536 message has many recipients that cause the same warning to be written.
2537 Howewer, Exim was preserving the list of already written lines for an
2538 entire SMTP session, which doesn't seem right. The memory is now reset if a
2539 new message is started.
2541 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2542 debugging output unless log_rewrite was also set.
2544 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2545 of (int)(handle) when we know that handle contains (void *)(-1).
2547 61. The Exim daemon panic-logs an error return when it closes the incoming
2548 connection. However "connection reset by peer" seems to be common, and
2549 isn't really an error worthy of noting specially, so that particular error
2552 62. When Exim is trying to find all the local interfaces, it used to panic and
2553 die if the ioctl to get the interface flags failed. However, it seems that
2554 on at least one OS (Solaris 9) it is possible to have an interface that is
2555 included in the list of interfaces, but for which you get a failure error
2556 for this call. This happens when the interface is not "plumbed" into a
2557 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2558 failure of the "get flags" call assumes that the interface is down.
2560 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2561 makes life easier for people who are doing arithmetic on fields extracted
2562 from dates, where you often get leading zeros that should not be
2563 interpreted as octal.
2565 64. Added qualify_domain to the redirect router, to override the global
2568 65. If a pathologically long header line contained very many addresses (the
2569 report of this problem mentioned 10 000) and each of them was rewritten,
2570 Exim could use up a very large amount of memory. (It kept on making new
2571 copies of the header line as it rewrote, and never released the old ones.)
2572 At the expense of a bit more processing, the header rewriting function has
2573 been changed so that it no longer eats memory in this way.
2575 66. The generation of the Received: header has been moved from the time that a
2576 message starts to be received, to the time that it finishes. The timestamp
2577 in the Received: header should now be very close to that of the <= log
2578 line. There are two side-effects of this change:
2580 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2581 logged header lines no longer include the local Received: line, because
2582 it has not yet been created. The same applies to a copy of the message
2583 that is returned to a non-SMTP sender when a message is rejected.
2585 (b) When a filter file is tested using -bf, no additional Received: header
2586 is added to the test message. After some thought, I decided that this
2589 This change does not affect the value of $received_for. It is still set
2590 after address rewriting, but before local_scan() is called.
2592 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2594 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2595 gave an unhelpful panic error message, and a defer error. I have managed to
2596 change this behaviour so that it now rejects any supplied certificate,
2597 which seems right, as the list of acceptable certificates is empty.
2599 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2600 gave an unhelpful defer error. I have not managed to make this reject any
2601 supplied certificates, but the error message it gives is "no certificate
2602 supplied", which is not helpful.
2604 70. exigrep's output now also includes lines that are not associated with any
2605 message, but which match the given pattern. Implemented by a patch from
2606 Martin Sluka, which also tidied up the Perl a bit.
2608 71. Recipient callout verification, like sender verification, was using <> in
2609 the MAIL FROM command. This isn't really the right thing, since the actual
2610 sender may affect whether the remote host accepts the recipient or not. I
2611 have changed it to use the actual sender in the callout; this means that
2612 the cache record is now keyed on a recipient/sender pair, not just the
2613 recipient address. There doesn't seem to be a real danger of callout loops,
2614 since a callout by the remote host to check the sender would use <>.
2615 [SEE ABOVE: changed after hitting problems.]
2617 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2618 temporary errors. However, in the case of such a code being given after
2619 the end of a data transmission (i.e. after ".") Exim was failing to write
2620 a retry record for the message. (Yes, there was some broken host that was
2621 actually sending 8xx at this point.)
2623 73. An unknown lookup type in a host list could cause Exim to panic-die when
2624 the list was checked. (An example that provoked this was putting <; in the
2625 middle of a list instead of at the start.) If this happened during a DATA
2626 ACL check, a -D file could be left lying around. This kind of configuration
2627 error no longer causes Exim to die; instead it causes a defer errror. The
2628 incident is still logged to the main and panic logs.
2630 74. Buglet left over from Exim 3 conversion. The message "too many messages
2631 in one connection" was written to the rejectlog but not the mainlog, except
2632 when address rewriting (yes!) was being logged.
2634 75. Added write_rejectlog option.
2636 76. When a system filter was run not as root (that is, when system_filter_user
2637 was set), the values of the $n variables were not being returned to the
2638 main process; thus, they were not subsequently available in the $sn
2641 77. Added +return_path_on_delivery log selector.
2643 78. A connection timeout was being treated differently from recipients deferred
2644 when testing hosts_max_try with a message that was older than the host's
2645 retry timeout. (The host should not be counted, thus allowing all hosts to
2646 be tried at least once before bouncing.) This may have been the cause of an
2647 occasionally reported bug whereby a message would remain on the queue
2648 longer than the retry timeout, but would be bounced if a delivery was
2649 forced. I say "may" because I never totally pinned down the problem;
2650 setting up timeout/retry tests is difficult. See also the next item.
2652 79. The ultimate address timeout was not being applied to errors that involved
2653 a combination of host plus message (for example, a timeout on a MAIL
2654 command). When an address resolved to a number of possible hosts, and they
2655 were not all tried for each delivery (e.g. because of hosts_max_try), a
2656 message could remain on the queue longer than the retry timeout.
2658 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2661 81. Fixed an obscure SMTP outgoing bug which required at least the following
2662 conditions: (a) there was another message waiting for the same server;
2663 (b) the server returned 5xx to all RCPT commands in the first message so
2664 that the message was not completed; (c) the server dropped the connection
2665 or gave a negative response to the RSET that Exim sends to abort the
2666 transaction. The observed case was a dropped connection after DATA that had
2667 been sent in pipelining mode. That is, the server had advertised PIPELINING
2668 but was not implementing it correctly. The effect of the bug was incorrect
2669 behaviour, such as trying another host, and this could lead to a crash.
2675 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2676 and daemon.c were passed as pointers to ints; they should have been
2677 pointers to socklen_t variables (which are typically unsigned ints).
2679 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2682 3. Fixed a really odd bug that affected only the testing scheme; patching a
2683 certain fixed string in the binary changed the value of another string that
2684 happened to be identical to the end of the original first string.
2686 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2687 name", it returns that address as the IP address. On some operating
2688 systems (e.g. Solaris), it also passes back the IP address string as the
2689 "host name". However, on others (e.g. Linux), it passes back an empty
2690 string. Exim wasn't checking for this, and was changing the host name to an
2691 empty string, assuming it had been canonicized.
2693 5. Although rare, it is permitted to have more than one PTR record for a given
2694 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2695 all the names associated with an address, because they do in Solaris.
2696 However, it seems that they do not in Linux for data that comes from the
2697 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2698 I found this out when I moved to a new Linux workstation and tried to run
2699 the Exim test suite.
2701 To get round this problem I have changed the code so that it now does its
2702 own call to the DNS to look up PTR records when searching for a host name.
2703 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2704 addresses that are only in /etc/hosts are still found.
2706 This behaviour is, however, controlled by an option called host_lookup_
2707 order, which defaults to "bydns:byaddr". If people want to use the other
2708 order, or indeed, just use one or the other means of lookup, they can
2709 specify it in this variable.
2711 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2712 some operating systems, this comes back from gethostbyaddr() as an empty
2713 string, and this is what Exim used to test for. However, it seems that in
2714 other systems, "." is yielded. Exim now tests for this case too.
2716 7. The values of check_spool_space and check_log_space are now held internally
2717 as a number of kilobytes instead of an absolute number of bytes. If a
2718 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2719 kilobyte. This means that much larger values can be stored.
2721 8. Exim monitor: an attempt to get the action menu when not actually pointing
2722 at a message produces an empty menu entitled "No message selected". This
2723 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2724 no entries in it ("Shell widget menu has zero width and/or height"). So I
2725 have added a single, blank menu entry in this case.
2727 9. Added ${quote_local_part.
2729 10. MIME decoding is now applied to the contents of Subject: header lines when
2732 11. Now that a reference to $sender_host_address automatically causes a reverse
2733 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2734 host lookup before query-style lookups in lists that might use this
2735 variable. This has therefore been abolished, and the "net-" prefix is no
2736 longer necessary for query-style lookups.
2738 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2739 have been a typo for LFLAGS, so it has been changed.
2741 13. The install script calls Exim with "-C /dev/null" in order to find the
2742 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2743 to be output. Howeve, since Exim outputs its version number before the
2744 error, it didn't break the script. It just looked ugly. I fixed this by
2745 always allowing "-C /dev/null" if the caller is root.
2747 14. Ignore overlarge ACL variable number when reading spool file - insurance
2748 against a later release with more variables having written the file.
2750 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2751 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2752 this, as well as the form without the "IPv6" on the front.
2754 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2755 OS/Makefile-Darwin file.
2757 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2758 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2760 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2761 variable. However, some OS use ints for inodes. I've added cast to long int
2762 to get rid of the compiler warning.
2764 19. I had forgotten to lock out "/../" in configuration file names when
2765 ALT_CONFIG_PREFIX was set.
2767 20. Routers used for verification do not need to specify transports. However,
2768 if such a router generated a host list, and callout was configured, Exim
2769 crashed, because it could not find a port number from the (non-existent)
2770 transport. It now assumes port 25 in this circumstance.
2772 21. Added the -t option to exigrep.
2774 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2775 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2776 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2777 them is set, LOOKUP_LSEARCH is forced.
2779 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2781 24. Added sender and host information to the "rejected by local_scan()" log
2782 line; previously there was no indication of these.
2784 25. Added .include_if_exists.
2786 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2787 Linux. It turns out that not all file systems support this. Apparently some
2788 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2789 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2790 not supported on the file descriptor, is now ignored when Exim is trying to
2791 sync a directory. This applies only to Linux.
2793 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2795 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2796 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2799 29. Added support for maildirsize files from supplied patch (modified a bit).
2801 30. The use of :fail: followed by an empty string could lead Exim to respond to
2802 sender verification failures with (e.g.):
2804 550 Verification failed for <xxx>
2805 550 Sender verify failed
2807 where the first response line was missing the '-' that indicates it is not
2808 the final line of the response.
2810 31. The loop for finding the name of the user that called Exim had a hardwired
2811 limit of 10; it now uses the value of finduser_retries, which is used for
2812 all other user lookups.
2814 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2817 33. Exim was neglecting to zero errno before one call of strtol() when
2818 expanding a string and expecting an integer value. On some systems this
2819 resulted in spurious "integer overflow" errors. Also, it was casting the
2820 result into an int without checking.
2822 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2823 did not work. The code looks as if it has *never* worked, though it appears
2824 to have been documented since at least releast 1.62. I have made it work.
2826 35. The "timeout_DNS" error in retry rules, also documented since at least
2827 1.62, also never worked. As it isn't clear exactly what this means, and
2828 clearly it isn't a major issue, I have abolished the feature by treating it
2829 as "timeout", and writing a warning to the main and panic logs.
2831 36. The display of retry rules for -brt wasn't always showing the error code
2834 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2835 timeout_connect_A, timeout_connect_MX.
2837 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2838 to the empty sender.
2840 39. The daemon was not analysing the content of -oX till after it had closed
2841 stderr and disconnected from the controlling terminal. This meant that any
2842 syntax errors were only noted on the panic log, and the return code from
2843 the command was 0. By re-arranging the code a little, I've made the
2844 decoding happen first, so such errors now appear on stderr, and the return
2845 code is 1. However, the actual setting up of the sockets still happens in
2846 the disconnected process, so errors there are still only recorded on the
2849 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2850 connections (as happens on some IP stacks) was logged at start up time as
2851 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2852 it from "IPv6 and IPv4", which means that two separate sockets are being
2855 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2856 says whether AF_INET or AF_INET6 was passed as an argument.
2858 42. Exiwhat output was messed up when time zones were included in log
2861 43. Exiwhat now gives more information about the daemon's listening ports,
2862 and whether -tls-on-connect was used.
2864 44. The "port" option of the smtp transport is now expanded.
2866 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2867 silently ignored. Now an error message is written to the main and panic
2870 46. There's a new ACL modifier called "logwrite" which writes to a log file
2871 as soon as it is encountered.
2873 47. Added $local_user_uid and $local_user_gid at routing time.
2875 48. Exim crashed when trying to verify a sender address that was being
2878 49. Exim was recognizing only a space character after ".include". It now also
2879 recognizes a tab character.
2881 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2882 extracting the relevant information from the specification. The man page no
2883 longer contains scrambled data for the -d option, and I've added a section
2884 at the front about calling Exim under different names.
2886 51. Added "extra_headers" argument to the "mail" command in filter files.
2888 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2891 53. Installed eximstats 1.29.
2893 54. Added transport_filter_timeout as a generic transport option.
2895 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2896 Cc: header lines. This was required by RFC 822, but it not required by RFC
2899 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2900 incoming messages that did not have them. Now it does so only if the
2901 message originates locally, that is, if there is no associated remote host
2902 address. When Resent- header lines are present, this applies to the Resent-
2903 lines rather than the non-Resent- lines.
2905 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2906 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2908 58. Messages for configuration errors now include the name of the main
2909 configuration file - useful now that there may be more than one file in a
2910 list (.included file names were always shown).
2912 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2913 for those rare installations that do not start the daemon as root or run it
2914 setuid root. I've cut out the call to initgroups() if the daemon is not
2917 60. The Exim user and group can now be bound into the binary as text strings
2918 that are looked up at the start of Exim's processing.
2920 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2922 62. Added $mailstore_basename variable.
2924 63. Installed patch to sieve.c from Michael Haardt.
2926 64. When Exim failed to open the panic log after failing to open the main log,
2927 the original message it was trying to log was written to stderr and debug
2928 output, but if they were not available (the usual case in production), it
2929 was lost. Now it is written to syslog before the two lines that record the
2930 failures to open the logs.
2932 65. Users' Exim filters run in subprocesses under the user's uid. It is
2933 possible for a "deliver" command or an alias in a "personal" command to
2934 provoke an address rewrite. If logging of address rewriting is configured,
2935 this fails because the process is not running as root or exim. There may be
2936 a better way of dealing with this, but for the moment (because 4.30 needs
2937 to be released), I have disabled address rewrite logging when running a
2938 filter in a non-root, non-exim process.
2944 1. The buildconfig auxiliary program wasn't quoting the value set for
2945 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2946 not defined. This bug was masked in 4.22 by the effect that was fixed in
2949 2. Some messages that were rejected after a message id was allocated were
2950 shown as "incomplete" by exigrep. It no longer does this for messages that
2951 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2953 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2954 have allow_domain_literals set, the ID did not get logged in the <= line.
2955 Domain literals are now always recognized in Message-ID: header lines.
2957 4. The first argument for a ${extract expansion item is the key name or field
2958 number. Leading and trailing spaces in this item were not being ignored,
2959 causing some misleading effects.
2961 5. When deliver_drop_privilege was set, single queue runner processes started
2962 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2963 same command in the process it spins off) were not dropping privilege.
2965 6. When the daemon running as "exim" started a queue runner, it always
2966 re-executed Exim in the spun-off process. This is a waste of effort when
2967 deliver_drop_privilege is set. The new process now just calls the
2968 queue-runner function directly.
2974 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2977 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2978 ignored. Though the use of -oP was forcing the writing of a pid file, it
2979 was always written to the default place.
2981 3. If the message "no IP address found for host xxxx" is generated during
2982 incoming verification, it is now followed by identification of the incoming
2983 connection (so you can more easily find what provoked it).
2985 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2987 5. Added some features to "harden" Exim a bit more against certain attacks:
2989 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2990 be put in Local/Makefile. This is like the never_users runtime option,
2991 but it cannot be overridden. The default setting is "root".
2993 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2994 prefix string with which any file named in a -C command line option
2997 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2998 is retained for -C and -D only if the caller of Exim is root. Without
2999 it, the exim user may also use -C and -D and retain privilege.
3001 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3002 command line option is disabled.
3004 6. Macro names set by the -D option must start with an upper case letter, just
3005 like macro names defined in the configuration file.
3007 7. Added "dereference=" facility to LDAP.
3009 8. Two instances of the typo "uknown" in the source files are fixed.
3011 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3012 the Configure-Makefile script screwed up while processing it.
3014 10. Incorporated PCRE 4.4.
3016 11. The SMTP synchronization check was not operating right at the start of an
3017 SMTP session. For example, it could not catch a HELO sent before the client
3018 waited for the greeting. There is now a check for outstanding input at the
3019 point when the greeting is written. Because of the duplex, asynchronous
3020 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3021 way, but not yet received, when the check is performed.
3023 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3024 on TCP/IP sockets, because this apparently causes some broken clients to
3027 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3028 unchanged) from the Cygwin maintainer.
3030 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3031 was supported instead of testing for mbx. Effectively a typo.
3033 15. The spa authenticator server code was not checking that the input it
3034 received was valid base64.
3036 16. The debug output line for the "set" modifier in ACLs was not showing the
3037 name of the variable that was being set.
3039 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3041 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3042 lookup on its own. Something else was needed to trigger the lookup. For
3043 example, a match in host_lookup or the need for a host name in a host list.
3044 Now, if $sender_host_name is referenced and the host name has not yet been
3045 looked up, a lookup is performed. If the lookup fails, the variable remains
3046 empty, and $host_lookup_failed is set to "1".
3048 19. Added "eqi" as a case-independent comparison operator.
3050 20. The saslauthd authentication condition could segfault if neither service
3051 nor realm was specified.
3053 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3054 error message that was logged was misleading, and incoming SMTP
3055 connections were dropped. The message is now more accurate, and temporary
3056 errors are given to SMTP connections.
3058 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3059 (except RSET) with a 421 error, until QUIT is received. However, it was
3060 failing to send a response to QUIT.
3062 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3063 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3064 ACL is now run after the helo_try_verify_hosts code.
3066 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3067 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3068 case-independent, so other case variants are also recognized). Apparently
3069 some systems use these upper case variants.
3071 25. If more than two messages were waiting for the same host, and a transport
3072 filter was specified for the transport, Exim sent two messages over the
3073 same TCP/IP connection, and then failed with "socket operation on non-
3074 socket" when it tried to send the third.
3076 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3078 27. The extern definition of crypt16() in expand.c was not being excluded when
3079 the OS had its own crypt16() function.
3081 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3082 as a preferred synonym for return_size_limit, both as an option and as an
3085 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3087 30. Changed the default configuration ACL to relax the local part checking rule
3088 for addresses that are not in any local domains. For these addresses,
3089 slashes and pipe symbols are allowed within local parts, but the sequence
3090 /../ is explicitly forbidden.
3092 31. SPA server authentication was not clearing the challenge buffer before
3095 32. log_message in a "warn" ACL statement was writing to the reject log as
3096 well as to the main log, which contradicts the documentation and doesn't
3097 seem right (because no rejection is happening). So I have stopped it.
3099 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3100 However, I am unable to do any testing of this.
3102 34. Fixed an infelicity in the appendfile transport. When checking directories
3103 for a mailbox, to see if any needed to be created, it was accidentally
3104 using path names with one or more superfluous leading slashes; tracing
3105 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3107 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3108 contents are added to the log line that is written for every discarded
3109 recipient. (Previously a log_message setting was ignored.)
3111 36. The ${quote: operator now quotes the string if it is empty.
3113 37. The install script runs exim in order to find its version number. If for
3114 some reason other than non-existence or emptiness, which it checks, it
3115 could not run './exim', it was installing it with an empty version number,
3116 i.e. as "exim-". This error state is now caught, and the installation is
3119 38. An argument was missing from the function that creates an error message
3120 when Exim fails to connect to the socket for saslauthd authentication.
3121 This could cause Exim to crash, or give a corrupted message.
3123 39. Added isip, isip4, and isip6 to ${if conditions.
3125 40. The ACL variables $acl_xx are now saved with the message, and can be
3126 accessed later in routers, transports, and filters.
3128 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3129 strings in the file are not string-expanded.
3131 42. If a MAIL command specified a SIZE value that was too large to fit into an
3132 int variable, the check against message_size_limit failed. Such values are
3133 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3134 day this will have to be increased, but I don't think I want to be around
3135 when emails are that large.
3142 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3143 iconv() is not standard in FreeBSD.
3145 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3146 IPv6 enabled. The observed symptom was a segmentation fault on return from
3147 the function os_common_find_running_interfaces() in src/os.c.
3149 3. In the check_special_case() function in daemon.c I had used "errno" as an
3150 argument name, which causes warnings on some systems. This was basically a
3151 typo, since it was named "eno" in the comments!
3153 4. The code that waits for the clock to tick (at a resolution of some fraction
3154 of a second) so as to ensure message-id uniqueness was always waiting for
3155 at least one whole tick, when it could have waited for less. [This is
3156 almost certainly not relevant at current processor speeds, where it is
3157 unlikely to ever wait at all. But we try to future-proof.]
3159 5. The function that sleeps for a time interval that includes fractions of a
3160 second contained a race. It did not block SIGALRM between setting the
3161 timer, and suspending (a couple of lines later). If the interval was short
3162 and the sigsuspend() was delayed until after it had expired, the suspension
3163 never ended. On busy systems this could lead to processes getting stuck for
3166 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3167 process, before it forks any delivery processes. The open lookup caching
3168 mechanism meant that the open file or database connection was passed into
3169 the delivery process. The problem was that delivery processes always tidy
3170 up cached lookup data. This could cause a problem for the next delivery
3171 process started by the queue runner, because the external queue runner
3172 process does not know about the closure. So the next delivery process
3173 still has data in the lookup cache. In the case of a file lookup, there was
3174 no problem because closing a file descriptor in a subprocess doesn't affect
3175 the parent. However, if the lookup was caching a connection to a database,
3176 the connection was closed, and the second delivery process was likely to
3177 see errors such as "PGSQL: query failed: server closed the connection
3178 unexpectedly". The problem has been fixed by closing all cached lookups
3179 in a queue runner before running a delivery process.
3181 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3182 seem to have the "const" qualifier which it has on other OS. I've
3185 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3186 *of the same type* (client or server) with the same public name that an
3187 error should be diagnosed.
3189 9. When Exim looked up a host name for an IP address, but failed to find the
3190 original IP address when looking up the host name (a safety check), it
3191 output the message "<ip address> does not match any IP for NULL", which was
3192 confusing, to say the least. The bug was that the host name should have
3193 appeared instead of "NULL".
3195 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3196 user that is built into the binary, and the -C or -D options is used, root
3197 privilege is dropped before the configuration file is read. In addition,
3198 logging is switched to stderr instead of the normal log files. If the
3199 configuration then re-defines the Exim user, the unprivileged environment
3200 is probably not what is expected, so Exim logs a panic warning message (but
3203 However, if deliver_drop_privilege is set, the unprivileged state may well
3204 be exactly what is intended, so the warning has been cut out in that case,
3205 and Exim is allowed to try to write to its normal log files.
3211 1. smtp_return_error_details was not giving details for temporary sender
3212 or receiver verification errors.
3214 2. Diagnose a configuration error if two authenticators have the same public
3217 3. Exim used not to create the message log file for a message until the first
3218 delivery attempt. This could be confusing when incoming messages were held
3219 for policy or load reasons. The message log file is now created at the time
3220 the message is received, and an initial "Received" line is written to it.
3222 4. The automatically generated man page for command line options had a minor
3223 bug that caused no ill effects; however, a more serious problem was that
3224 the procedure for building the man page automatically didn't always
3225 operate. Consequently, release 4.20 contains an out-of-date version. This
3226 shouldn't happen again.
3228 5. When building Exim with embedded Perl support, the script that builds the
3229 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3230 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3232 6. The freeze_tell option was not being used for messages that were frozen on
3233 arrival, either by an ACL or by local_scan().
3235 7. Added the smtp_incomplete_transaction log selector.
3237 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3238 was accepting AUTH without a new EHLO.
3240 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3241 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3244 10. Logging of TCP/IP connections (when configured) now happens in the main
3245 daemon process instead of the child process, so that the TCP/IP connection
3246 count is more accurate (but it can never be perfect).
3248 11. The use of "drop" in a nested ACL was not being handled correctly in the
3249 outer ACL. Now, if condition failure induced by the nested "drop" causes
3250 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3251 or "require"), the connection is dropped.
3253 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3254 that yield "discard" can now be used with an "accept" or a "discard" verb,
3255 but an error is generated for any others (because I can't see a useful way
3256 to define what should happen).
3258 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3259 are now processed in the same way as lines in the Exim configuration file.
3260 In particular, continuation lines are supported.
3262 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3264 15. Added -ti meaning -t -i.
3266 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3267 domains, and warn by logging if others are found.
3269 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3270 returned from GIFCONF when Exim is trying to find the list of interfaces on
3271 a host. The code in os.c has been modified to copy each ifreq to an aligned
3272 structure in all cases.
3274 Also, in some cases, the returned ifreq's were being copied to a 'struct
3275 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3276 means the last couple of bytes of an IPv6 address could be chopped if the
3277 ifreq contained only a normal sockaddr (14 bytes storage).
3279 18. Named domain lists were not supported in the hosts_treat_as_local option.
3280 An entry such as +xxxx was not recognized, and was treated as a literal
3283 19. Ensure that header lines added by a DATA ACL are included in the reject log
3284 if the ACL subsequently rejects the message.
3286 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3287 MD5 (which is deprecated).
3289 21. When testing a filter file using -bf, Exim was writing a message when it
3290 took the sender from a "From " line in the message, but it was not doing so
3291 when it took $return_path from a Return-Path: header line. It now does.
3293 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3294 with a valid header line field name (a series of printing characters
3295 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3297 23. Changed "disc" in the source to "disk" to conform to the documentation and
3298 the book and for uniformity.
3300 24. Ignore Sendmail's -Ooption=value command line item.
3302 25. When execve() failed while trying to run a command in a pipe transport,
3303 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3304 could be confused with a return value of 69 from the command itself. This
3305 has been changed to 127, the value the shell returns if it is asked to run
3306 a non-existent command. The wording for the related log line suggests a
3307 non-existent command as the problem.
3309 26. If received_header_text expands to an empty string, do not add a Received:
3310 header line to the message. (Well, it adds a token one on the spool, but
3311 marks it "old" so that it doesn't get used or transmitted.)
3313 27. Installed eximstats 1.28 (addition of -nt option).
3315 28. There was no check for failure on the call to getsockname() in the daemon
3316 code. This can fail if there is a shortage of resources on the system, with
3317 ENOMEM, for example. A temporary error is now given on failure.
3319 29. Contrary to the C standard, it seems that in some environments, the
3320 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3321 program. Exim now does this explicitly; it affects the formatting of
3322 timestamps using strftime().
3324 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3325 complaints. I've now initialized all the variables, to avoid this.
3327 32. Header lines added by a system filter were not being "seen" during
3328 transport-time rewrites.
3330 33. The info_callback() function passed to OpenSSL is set up with type void
3331 (*)(SSL *, int, int), as described somewhere. However, when calling the
3332 function (actually a macro) that sets it up, the type void(*)() is
3333 expected. I've put in a cast to prevent warnings from picky compilers.
3335 34. If a DNS black list lookup found a CNAME record, but there were no A
3336 records associated with the domain it pointed at, Exim crashed.
3338 35. If a DNS black list lookup returned more than one A record, Exim ignored
3339 all but the first. It now scans all returned addresses if a particular IP
3340 value is being sought. In this situation, the contents of the
3341 $dnslist_value variable are a list of all the addresses, separated by a
3344 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3345 to accept a host name and all its aliases if the forward lookup for any of
3346 them yielded the IP address of the incoming connection. Now it accepts only
3347 those names whose forward lookup yields the correct IP address. Any other
3348 names are discarded. This closes a loophole whereby a rogue DNS
3349 administrator could create reverse DNS records to break through a
3350 wildcarded host restriction in an ACL.
3352 37. If a user filter or a system filter that ran in a subprocess used any of
3353 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3354 the wrong values were passed to the pipe command ($thisaddress had the
3355 value of $0, $0 had the value of $1, etc). This bug was introduced by
3356 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3358 38. Improved the line breaking for long SMTP error messages from ACLs.
3359 Previously, if there was no break point between 40 and 75 characters, Exim
3360 left the rest of the message alone. Two changes have been made: (a) I've
3361 reduced the minimum length to 35 characters; (b) if it can't find a break
3362 point between 35 and 75 characters, it looks ahead and uses the first one
3363 that it finds. This may give the occasional overlong line, but at least the
3364 remaining text gets split now.
3366 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3367 file descriptors might be low, and that setting 1000 would always raise it.
3368 It turns out that in some environments, the limit is already over 1000 and
3369 that lowering it causes trouble. So now Exim takes care not to decrease it.
3371 40. When delivering a message, the value of $return_path is set to $sender_
3372 address at the start of routing (routers may change the value). By an
3373 oversight, this default was not being set up when an address was tested by
3374 -bt or -bv, which affected the outcome if any router or filter referred to
3377 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3378 log only when "message" or "log_message" are set. However, if one of the
3379 conditions was an address verification, or a call to a nested ACL, the
3380 messages generated by the underlying test were being passed through. This
3381 no longer happens. The underlying message is available in $acl_verify_
3382 message for both "message" and "log_message" expansions, so it can be
3383 passed through if needed.
3385 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3386 new expansion $bh_ to give the encoded byte string without charset
3387 translation. Translation happens only if iconv() is available; HAVE_ICONV
3388 indicates this at build time. HEADERS_CHARSET gives the charset to
3389 translate to; headers_charset can change it in the configuration, and
3390 "headers charset" can change it in an individual filter file.
3392 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3393 that creates RFC 2047 encoded "words" labels them as that charset instead
3394 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3395 expansion operator; (ii) when Exim creates a From: line for a local
3396 message; (iii) when a header line is rewritten to include a "phrase" part.
3398 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3399 buggy, causing it to skip the first lines of messages whose message ID
3400 ended in 'D'. This would not have bitten before Exim release 4.14, because
3401 message IDs were unlikely to end in 'D' before then. The effect was to have
3402 incorrect size information for certain domains.
3404 45. #include "config.h" was missing at the start of the crypt16.c module. This
3405 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3408 46. If there was a timeout during a "random" callout check, Exim treated it as
3409 a failure of the random address, and carried on sending RSET and the real
3410 address. If the delay was just some slowness somewhere, the response to the
3411 original RCPT would be taken as a response to RSET and so on, causing
3412 mayhem of various kinds.
3414 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3415 when I implemented it. It didn't allow for the fact that some option values
3416 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3417 the right test for positive values.
3419 48. Domain names in DNS records are case-independent. Exim always looks them up
3420 in lower case. Some resolvers return domain names in exactly the case they
3421 appear in the zone file, that is, they may contain uppercase letters. Not
3422 all resolvers do this - some return always lower case. Exim was treating a
3423 change of case by a resolver as a change of domain, similar to a widening
3424 of a domain abbreviation. This triggered its re-routing code and so it was
3425 trying to route what was effectively the same domain again. This normally
3426 caused routing to fail (because the router wouldn't handle the domain
3427 twice). Now Exim checks for this case specially, and just changes the
3428 casing of the domain that it ultimately uses when it transmits the message
3431 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3434 50. If a filter generated a file delivery with a non-absolute name (possible if
3435 no home directory exists for the router), the forbid_file option was not
3438 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3439 the existing equality matching.
3441 52. Exim was using ints instead of ino_t variables in some places where it was
3442 dealing with inode numbers.
3444 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3445 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3446 called TMPDIR, and if it finds it is different, it changes its value.
3448 54. The smtp_printf() function is now made available to local_scan() so
3449 additional output lines can be written before returning. There is also an
3450 smtp_fflush() function to enable the detection of a dropped connection.
3451 The variables smtp_input and smtp_batched_input are exported to
3454 55. Changed the default runtime configuration: the message "Unknown user"
3455 has been removed from the ACL, and instead placed on the localuser router,
3456 using the cannot_route_message feature. This means that any verification
3457 failures that generate their own messages won't get overridden. Similarly,
3458 the "Unrouteable address" message that was in the ACL for unverifiable
3459 relay addresses has also been removed.
3461 56. Added hosts_avoid_esmtp to the smtp transport.
3463 57. The exicyclog script was not checking for the esoteric option
3464 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3465 will work only if exicyclog is run under the appropriate euid.
3467 58. Following a discussion on the list, the rules by which Exim recognises line
3468 endings on incoming messages have been changed. The -dropcr and drop_cr
3469 options are now no-ops, retained only for backwards compatibility. The
3470 following line terminators are recognized: LF CRLF CR. However, special
3471 processing applies to CR:
3473 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3474 nor a local message in the state where . is a terminator.
3476 (ii) If a bare CR is encountered in a header line, an extra space is added
3477 after the line terminator so as not to end the header. The reasoning
3478 behind this is that bare CRs in header lines are most likely either
3479 to be mistakes, or people trying to play silly games.
3481 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3482 was being incorrectly given as 18 bytes larger than it should have been.
3483 This is a VOB (very old bug).
3485 60. This may never have affected anything current, but just in case it has:
3486 When the local host is found other than at the start of a list of hosts,
3487 the local host, those with the same MX, and any that follow, are discarded.
3488 When the list in question was part of a longer list of hosts, the following
3489 hosts (not currently being processed) were also being discarded. This no
3490 longer happens. I'm not sure if this situation could ever has previously
3493 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3496 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3497 header. This is something that is recommended in a new Internet Draft, and
3498 is something that is documented as being done by Sendmail. There are two
3499 possible values. For messages generated by the autoreply transport, Exim
3502 Auto-Submitted: auto-replied
3504 whereas for all other generated messages (e.g. bounces) it adds
3506 Auto-Submitted: auto-generated
3508 63. The "personal" condition in filters now includes a test for the
3509 Auto-Submitted: header. If it contains the string "auto-" the message it
3510 not considered personal.
3512 64. Added rcpt_include_affixes as a generic transport option.
3514 65. Added queue_only_override (default true).
3516 66. Added the syslog_duplication option.
3518 67. If what should have been the first header line of a message consisted of
3519 a space followed by a colon, Exim was mis-interpreting it as a header line.
3520 It isn't of course - it is syntactically invalid and should therefore be
3521 treated as the start of the message body. The misbehaviour could have
3522 caused a number of strange effects, including loss of data in subsequent
3523 header lines, and spool format errors.
3525 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3526 client host had authenticated. This control can now be exercised by an ACL
3527 for more flexibility.
3529 69. By default, callouts do not happen when testing with -bh. There is now a
3530 variant, -bhc, which does actually run the callout code, including
3531 consulting and updating the callout cache.
3533 70. Added support for saslauthd authentication, courtesy of Alexander
3536 71. If statvfs() failed on the spool or log directories while checking their
3537 size for availability, Exim confusingly gave the error "space shortage".
3538 Furthermore, in debugging mode it crashed with a floating point exception.
3539 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3540 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3541 serious problem, Exim now writes to the main and panic logs when this
3542 happens, with details of the failure. It then refuses to accept the
3543 incoming message, giving the message "spool directory problem" or "log
3544 directory problem" with a 421 code for SMTP messages.
3546 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3547 0, 1, and 2 exist, because some OS complain for execs without them (see
3548 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3549 descriptors. However, the code omitted to check that the open succeeded,
3550 causing mysterious errors if for some reason the permissions on /dev/null
3551 got screwed. Now Exim writes a message to the main and panic logs, and
3552 bombs out if it can't open /dev/null.
3554 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3555 interact so that it is all more flexible. It is supposed to remain
3556 backwards compatible. Also added extra_local_interfaces.
3558 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3559 to bomb out with an assertion failure - to the client this appears as a
3560 connection drop. This problem occurs in the part of the code that was taken
3561 from the Samba project. Fortunately, the assertion is in a very simple
3562 function, so I have fixed this by reproducing the function inline in the
3563 one place where it is called, and arranging for authentication to fail
3564 instead of killing the process with assert().
3566 75. The SPA client code was not working when the server requested OEM rather
3567 than Unicode encoding.
3569 76. Added code to make require_files with a specific uid setting more usable in
3570 the case where statting the file as root fails - usually a non-root-mounted
3571 NFS file system. When this happens and the failure is EACCES, Exim now
3572 forks a subprocess and does the per-uid checking as the relevant uid.
3574 77. Added process_log_path.
3576 78. If log_file_path was not explicitly set, a setting of check_log_space or
3577 check_log_inodes was ignored.
3579 79. If a space check for the spool or log partitions fails, the incident is now
3580 logged. Of course, in the latter case the data may get lost...
3582 80. Added the %p formatting code to string_format() so that it can be used to
3583 print addresses in debug_print(). Adjusted all the address printing in the
3584 debugging in store.c to use %p rather than %d.
3586 81. There was a concern that a line of code in smtp_in.c could overflow a
3587 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3588 initially expressed, the concern was not well-founded, because trailing
3589 spaces are removed early. However, if the trailing spaces were followed by
3590 a NULL, they did not get removed, so the overflow was possible. Two fixes
3593 (a) I re-wrote the offending code in a cleaner fashion.
3594 (b) If an incoming SMTP command contains a NULL character, it is rejected
3597 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3598 runs initgroups(), so that if the Exim user is in any additional groups,
3599 they will be used during message reception.
3605 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob