1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.266 2005/11/21 10:24:02 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 Two changes to the default runtime configuration:
11 (1) Move the checks for relay_from_hosts and authenticated clients from
12 after to before the (commented out) DNS black list checks.
14 (2) Add control=submission to the relay_from_hosts and authenticated
15 clients checks, on the grounds that messages accepted by these
16 statements are most likely to be submissions.
18 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
20 (1) Generate an error if the third argument for the ${prvs expansion is
23 (2) Treat a missing third argument of ${prvscheck as if it were an empty
26 (3) Reset the variables that are obtained from the first argument of
27 ${prvscheck and used in the second argument before leaving the code,
28 because their memory is reclaimed, so using them afterwards may do
31 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
32 one (it's much easier than Tom thought :-).
34 (5) Because of (4), we can now allow for the use of $prvscheck_result
35 inside the third argument.
37 PH/03 For some reason, the default setting of PATH when running a command from
38 a pipe transport was just "/usr/bin". I have changed it to
41 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
42 anything to be listed in the output from -bV.
44 PH/05 When a filter generated an autoreply, the entire To: header line was
45 quoted in the delivery log line, like this:
47 => >A.N.Other <ano@some.domain> <original@ddress> ...
49 This has been changed so that it extracts the operative address. There
50 may be more than one such address. If so, they are comma-separated, like
53 => >ano@some.domain,ona@other.domain <original@ddress> ...
55 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
56 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
57 not being looked up in the rDNS to get a host name, Exim was showing the
58 IP address twice in Received: lines, even though the IP addresses were
59 identical. For example:
61 Received: from [1.2.3.4] (helo=[1.2.3.4])
63 However, if the real host name was known, it was omitting the HELO data
64 if it matched the actual IP address. This has been tidied up so that it
65 doesn't show the same IP address twice.
67 PH/07 When both +timestamp and +memory debugging was on, the value given by
68 $tod_xxx expansions could be wrong, because the tod_stamp() function was
69 called by the debug printing, thereby overwriting the timestamp buffer.
70 Debugging no longer uses the tod_stamp() function when +timestamp is set.
72 PH/08 When the original message was included in an autoreply transport, it
73 always said "this is a copy of the message, including all the headers",
74 even if body_only or headers_only was set. It now gives an appropriate
77 PH/09 Applied a patch from the Sieve maintainer which:
80 o adds the (disabled) notify extension core
81 o adds some debug output for the result of if/elsif tests
82 o points to the current vacation draft in the documentation
83 and documents the missing references header update
87 o fixes a bug in processing the envelope test (when testing
88 multiple envelope elements, the last element determinted the
91 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
92 Electronic Mail") by including:
94 Auto-submitted: auto-generated
96 in the messages that it generates (bounce messages and others, such as
97 warnings). In the case of bounce messages for non-SMTP mesages, there was
98 also a typo: it was using "Auto_submitted" (underscore instead of
99 hyphen). Since every message generated by Exim is necessarily in response
100 to another message, thes have all been changed to:
102 Auto-Submitted: auto-replied
104 in accordance with these statements in the RFC:
106 The auto-replied keyword:
108 - SHOULD be used on messages sent in direct response to another
109 message by an automatic process,
111 - MUST NOT be used on manually-generated messages,
113 - MAY be used on Delivery Status Notifications (DSNs) and Message
114 Disposition Notifications (MDNs),
116 - MUST NOT be used on messages generated by automatic or periodic
117 processes, except for messages which are automatic responses to
120 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
121 to the default Received: header definition.
123 PH/12 Added log selector acl_warn_skipped (default on).
125 PH/13 After a successful wildlsearch lookup, discard the values of numeric
126 variables because (a) they are in the wrong storage pool and (b) even if
127 they were copied, it wouldn't work properly because of the caching.
129 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
130 checking when decoding. Apparently there are clients that generate
131 overlong encoded strings. Why am I not surprised?
133 PH/15 If the first argument of "${if match_address" was not empty, but did not
134 contain an "@" character, Exim crashed. Now it writes a panic log message
135 and treats the condition as false.
137 PH/16 In autoreply, treat an empty string for "once" the same as unset.
139 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
140 extension "envelope-auth". The code is finished and in agreement with
141 other implementations, but there is no documentation so far and in fact,
142 nobody wrote the draft yet. This extension is currently #undef'ed, thus
143 not changing the active code.
145 Print executed "if" and "elsif" statements when debugging is used. This
146 helps a great deal to understand what a filter does.
148 Document more things not specified clearly in RFC3028. I had all this
149 sorted out, when out of a sudden new issues came to my mind. Oops."
151 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
158 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
159 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
162 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
163 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
165 PH/03 Typo: missing ".o" in src/pcre/Makefile.
167 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
168 header line, restrict the check to what is listed in RFCs 2369 and 2929.
169 Also, for "Auto-Submitted", treat anything other than "no" as
170 non-personal, in accordance with RFC 3834. (Previously it treated
171 anything starting "auto-" as non-personal.)
173 TF/01 The control=submission/name=... option had a problem with syntax
174 errors if the name included a slash character. The /name= option
175 now slurps the rest of the string, so it can include any characters
176 but it must come last in the list of options (after /sender_retain
179 PH/05 Some modifications to the interface to the fake nameserver for the new
182 PH/06 Add ${if def:sender_address {(envelope-from <$sender_address>)\n\t}}
183 into the default Received: header string.
190 TK/01 Added the "success_on_redirect" address verification option. See
191 NewStuff for rationale and an example.
193 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
195 PH/02 Patch to exigrep to allow it to work on syslog lines.
197 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
198 fread() to skip over the body file's header line, because in Cygwin the
199 header line is locked and is inaccessible.
201 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
202 co-exist for some time) to make it clear that it is the Exim ID that is
203 referenced, not the Message-ID: header line.
205 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
206 string_format() function, because snprintf() does not exist on all
209 PH/06 The use of forbid_filter_existstest now also locks out the use of the
210 ${stat: expansion item.
212 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
213 protocol synchronization error", to keep the pedants happy.
215 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
216 well as for IRIX systems, when gcc is being used. See the host.c source
219 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
221 PH/10 Named domain lists were not working if used in a queue_smtp_domains
224 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
225 transport and to the smtp transport in LMTP mode.
227 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
229 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
230 run a filter in a subprocess. This could lead to confusion in subsequent
231 lookups in the parent process. There should also be a search_tidyup() at
232 the end of the subprocess.
234 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
235 only if the host matched helo_try_verify_hosts, which caused the
236 verification to occur when the EHLO/HELO command was issued. The ACL just
237 tested the remembered result. Now, if a previous verification attempt has
238 not happened, "verify = helo" does it there and then.
240 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
242 TK/03 Fix log output including CR from clamd.
244 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
245 exist provoked a memory error which could cause a segfault.
247 PH/15 Installed PCRE 6.2
249 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
251 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
252 of the problem. Specifically, suggested +O2 rather than +O1 for the
255 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
257 PH/20 If a delivery was routed to a non-standard port by means of an SRV
258 record, the port was not correctly logged when the outgoing_port log
259 selector was set (it logged the transort's default port).
261 PH/21 Added support for host-specific ports to manualroute, queryprogram,
262 fallback_hosts, and "hosts" in the smtp transport.
264 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
265 host errors such as "Connection refused".
267 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
268 authentication with radiusclient 0.4.9:
270 - Error returned from rc_read_config was caught wrongly
271 - Username/password not passed on to radius server due to wrong length.
273 The presumption is that some radiusclient API changes for 4.51/PH/17
274 were not taken care of correctly. The code is still untested by me (my
275 Linux distribution still has 0.3.2 of radiusclient), but it was
276 contributed by a Radius user.
278 PH/24 When doing a callout, the value of $domain wasn't set correctly when
279 expanding the "port" option of the smtp transport.
281 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
282 while reading a MIME header. Thanks to Tom Hughes for a patch.
284 PH/24 Include config.h inside local_scan.h so that configuration settings are
287 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
288 that in an ACL for RCPT (for example), you can examine exactly what was
291 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
292 commands, but it was not correctly comparing the address with the actual
293 client host address. Thus, it would show the EHLO address in Received:
294 header lines when this was not necessary.
296 PH/27 Added the % operator to ${eval:}.
298 PH/28 Exim tries to create and chdir to its spool directory when it starts;
299 it should be ignoring failures (because with -C, for example, it has lost
300 privilege). It wasn't ignoring creation failures other than "already
303 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
306 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
307 because an input line was too long, either on its own, or by virtue of
308 too many continuations, the temporary file was not being removed, and the
309 return code was incorrect.
311 PH/31 Missing "BOOL" in function definition in filtertest.c.
313 PH/32 Applied Sieve patches from the maintainer.
315 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
317 PH/33 Added "verify = not_blind".
319 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
320 Local/Makefile (with some defaults set). These are used in built scripts
321 such as exicyclog, but they have never been used in the exim_install
322 script (though there are many overriding facilities there). I have
323 arranged that the exim_install script now takes note of these two
326 PH/35 Installed configuration files for Dragonfly.
328 PH/36 When a locally submitted message by a trusted user did not contain a
329 From: header, and the sender address was obtained from -f or from an SMTP
330 MAIL command, and the trusted user did not use -F to supply a sender
331 name, $originator_name was incorrectly used when constructing a From:
332 header. Furthermore, $originator_name was used for submission mode
333 messages from external hosts without From: headers in a similar way,
334 which is clearly wrong.
336 PH/37 Added control=suppress_local_fixups.
338 PH/38 When log_selector = +received_sender was set, and the addition of the
339 sender made the log line's construction buffer exactly full, or one byte
340 less than full, an overflow happened when the terminating "\n" was
343 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
344 when the result of a list match is failure because a DNS lookup failed.
346 PH/40 RM_COMMAND is now used in the building process.
348 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
349 the "build-* directories that it finds.
351 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
352 address in a domain literal was a prefix of an interface address.
354 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
355 when verifying a sender address, unless rewrite_headers is false.
357 PH/44 Wrote a long comment about why errors_to addresses are verified as
358 recipients, not senders.
360 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
361 the ratelimit ACL was added.
363 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
365 PH/46 Added extra information about PostgreSQL errors to the error string.
367 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
368 avoiding the need to install special zones in a real server. This is
369 backwards compatible; if it can't find the fake resolver, it drops back.
370 Thus, both old and new test suites can be run.
372 TF/02 Added util/ratelimit.pl
374 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
375 clock is set back in time.
377 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
378 Candler <B.Candler@pobox.com>.
380 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
381 OK for addresses that are the result of redirections.
383 PH/48 A number of further additions for the benefit of the new test suite,
384 including a fake gethostbyname() that interfaces to the fake DNS resolver
387 TF/06 The fix for widen_domains has also been applied to qualify_single and
388 search_parents which are the other dnslookup options that can cause
391 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
394 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
396 TF/07 Exim produced the error message "an SRV record indicated no SMTP
397 service" if it encountered an MX record with an empty target hostname.
398 The message is now "an MX or SRV record indicated no SMTP service".
400 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
401 if the DNS of the sending site is misconfigured. This is quite a
402 common situation. This change restores the behaviour of treating a
403 helo verification defer as a failure.
405 PH/51 If self=fail was set on a router, the bounce message did not include the
406 actual error message.
412 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
414 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
415 command itself ended in error, the underlying message about the transport
416 filter timeout was being overwritten with the pipe command error. Now the
417 underlying error message should be appended to the second error message.
419 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
421 PH/02 Reduce the amount of output that "make" produces by default. Full output
422 can still be requested.
424 PH/03 The warning log line about a condition test deferring for a "warn" verb
425 was being output only once per connection, rather than after each
426 occurrence (because it was using the same function as for successful
427 "warn" verbs). This seems wrong, so I have changed it.
429 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
430 it should not have, which might have caused a crash in the right
431 circumstances, but probably never did.
433 PH/04 Installed a modified version of Tony Finch's patch to make submission
434 mode fix the return path as well as the Sender: header line, and to
435 add a /name= option so that you can make the user's friendly name appear
438 TF/03 Added the control = fakedefer ACL modifier.
440 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
441 Mark Lowes for thorough testing.
443 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
445 TK/03 Merged latest SRS patch from Miles Wilton.
447 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
448 with the definition in sysexits.h (which is #included earlier).
449 Fortunately, Exim does not actually use EX_OK. The code used to try to
450 preserve the sysexits.h value, by assumimg that macro definitions were
451 scanned for macro replacements. I have been disabused of this notion,
452 so now the code just undefines EX_OK before #including unistd.h.
454 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
455 in the smtp transport. When a block could not be written in a single
456 write() function, the timeout was being re-applied to each part-write.
457 This seems wrong - if the receiver was accepting one byte at a time it
458 would take for ever. The timeout is now adjusted when this happens. It
459 doesn't have to be particularly precise.
461 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
462 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
464 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
465 without a domain if the check to <postmaster@domain> fails.
467 SC/01 Eximstats: added -xls and the ability to specify output files
468 (patch written by Frank Heydlauf).
470 SC/02 Eximstats: use FileHandles for outputing results.
472 SC/03 Eximstats: allow any combination of xls, txt, and html output.
474 SC/04 Eximstats: fixed display of large numbers with -nvr option
476 SC/05 Eximstats: fixed merging of reports with empty tables.
478 SC/06 Eximstats: added the -include_original_destination flag
480 SC/07 Eximstats: removed tabs and trailing whitespace.
482 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
484 TK/06 MBOX spool code: Add real "From " MBOX separator line
485 so the .eml file is really in mbox format (even though
486 most programs do not really care). Patch from Alex Miller.
488 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
489 The latter is generated from $received_to and is only set if the
490 message has one envelope recipient. SA can use these headers,
491 obviously out-of-the-box. Patch from Alex Miller.
493 PH/08 The ${def test on a variable was returning false if the variable's
494 value was "0", contrary to what the specification has always said!
495 The result should be true unless the variable is empty.
497 PH/09 The syntax error of a character other than { following "${if
498 def:variable_name" (after optional whitespace) was not being diagnosed.
499 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
500 accidental colon was present, for example, could give incorrect results.
502 PH/10 Tidied the code in a number of places where the st_size field of a stat()
503 result is used (not including appendfile, where other changes are about
506 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
507 This involved changing a lot of size variables from int to off_t. It
508 should work with maildirs and everything.
510 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
511 spamd dying while we are connected to it.
513 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
516 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
517 function for BDB 4.3.
519 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
521 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
522 the sources for PCRE have been re-arranged and more files are now
525 PH/15 The code I had for printing potentially long long variables in PH/11
526 above was not the best (it lost precision). The length of off_t variables
527 is now inspected at build time, and an appropriate printing format (%ld
528 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
529 to be "long long int" or "long int". This is needed for the internal
530 formatting function string_vformat().
532 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
533 the configuration file to be ":syslog", then the script "guesses" where
534 the logs files are, rather than using the compiled in default. In our
535 case the guess is not the same as the compiled default, so the script
536 suddenly stopped working when I started to use syslog. The patch checks
537 to see if log_file_path is "". If so, it attempts to read it from exim
538 with no configuration file to get the compiled in version, before it
539 falls back to the previous guessing code."
541 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
542 implementing BATV in an Exim configuration. See NewStuff for the gory
545 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
546 Makefile that are specific to HP-UX.
548 PH/18 If the "use_postmaster" option was set for a recipient callout together
549 with the "random" option, the postmaster address was used as the MAIL
550 FROM address for the random test, but not for the subsequent recipient
551 test. It is now used for both.
553 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
554 patch removes a few documentation additions to RFC 3028, because the
555 latest draft now contains them. It adds the new en;ascii-case comparator
556 and a new error check for 8bit text in MIME parts. Comparator and
557 require names are now matched exactly. I enabled the subaddress
558 extension, but it is not well tested yet (read: it works for me)."
560 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
561 rework some of the code of TK/09 above to avoid the hardwired use of
562 "%lld" and "long long". Replaced the call to snprintf() with a call to
565 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
566 records point to non-existent hosts", "retry timeout exceeded", and
567 "retry time not reached for any host after a long failure period".
569 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
570 experimental DomainKeys support:
572 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
573 (2) On an error such as an illegally used "control", the wrong name for
574 the control was given.
576 These problems did NOT occur unless DomainKeys support was compiled.
578 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
580 PH/24 Added ${if match_ip condition.
582 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
583 databases so that it will be absolutely obvious if a crash occurs in the
584 DB library. This is a regular occurrence (often caused by mis-matched
587 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
588 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
589 picked up on a user's system that detects such things. There doesn't seem
590 to be a gcc warning option for this - only an attribute that has to be
591 put on the function's prototype. It seems that in Fedora Core 4 they have
592 set this on a number of new functions. No doubt there will be more in due
595 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
596 specify a transport. However, if an address that was verified by such a
597 router was the subject of a callout, Exim crashed because it tried to
598 read the rcpt_include_affixes from the non-existent transport. Now it
599 just assumes that the setting of that option is false. This bug was
600 introduced by 4.51/PH/31.
602 PH/28 Changed -d+all to exclude +memory, because that information is very
603 rarely of interest, but it makes the output a lot bigger. People tend to
604 do -d+all out of habit.
606 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
607 code in os-type was giving problems when libc.so lives in lib64, like on
610 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
611 aren't the modern standard, and it seems that some systems' include files
612 don't always have them. Exim was already checking for some of the newer
613 ones like T_AAAA, and defining it itself. I've added checks for all the
614 record types that Exim uses.
616 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
617 not automatically generating a new one, as it is supposed to. This
618 prevented TLS from working. If the file did exist, but contained invalid
619 data, a new version was generated, as expected. It was only the case of a
620 non-existent file that was broken.
622 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
623 with a change in libdomainkeys > 0.64.
625 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
626 from DNS. If the selector record carries the flag, it now has
627 precedence over the domain-wide flag.
629 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
631 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
632 the use of a port name that isn't defined in /etc/services) occurred, the
633 message was deferred as in a normal delivery, and thus remained on the
634 spool, instead of being failed because of the mua_wrapper setting. This
635 is now fixed, and I tidied up some of the mua_wrapper messages at the
638 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
639 the messages in a hash of arrays rather than using individual hashes.
640 This is a bit cleaner and results in dramatic memory savings, albeit
641 at a slight CPU cost.
643 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
644 as requested by Marc Sherman.
646 SC/10 Eximstats: added histograms for user specified patterns as requested
649 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
651 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
652 fopen() in the content-scanning modules that did not already have it.
658 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
659 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
661 TK/02 Fix ACL "control" statement not being available in MIME ACL.
663 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
665 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
666 to test Sieve filters that use "vacation".
668 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
669 that changes the way the GnuTLS parameters are stored in the cache file.
670 The new format can be generated externally. For backward compatibility,
671 if the data in the cache doesn't make sense, Exim assumes it has read an
672 old-format file, and it generates new data and writes a new file. This
673 means that you can't go back to an older release without removing the
676 PH/03 A redirect router that has both "unseen" and "one_time" set does not
677 work if there are any delivery delays because "one_time" forces the
678 parent to be marked "delivered", so its unseen clone is never tried
679 again. For this reason, Exim now forbids the simultaneous setting of
682 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
683 redirected to themselves ("homonym" addresses). Read the long ChangeLog
684 entry if you want to know the details. The fix, however, neglected to
685 consider the case when local delivery batching is involved. The test for
686 "previously delivered" was not happening when checking to see if an
687 address could be batched with a previous (undelivered) one; under
688 certain circumstances this could lead to multiple deliveries to the same
691 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
692 in its include files, and this causes problems building Exim.
694 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
695 header_syntax) but Exim was just ignoring anything given after a slash.
696 In particular, this caused confusion with an attempt to use "verify =
697 reverse_host_lookup/defer_ok". An error is now given when options are
698 supplied for verify items that do not have them. (Maybe reverse_host_
699 lookup should have a defer_ok option, but that's a different point.)
701 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
702 defined by RFC 821) to 2048, because there were problems with some AUTH
703 commands, and RFC 1869 says the size should be increased for extended
704 SMTP commands that take arguments.
706 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
709 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
710 "unknown" error; now it says that the functionality isn't in the binary.
712 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
713 an address' error message when a string expansion fails (syntax or
714 whatever). Otherwise the password may appear in the log. Following change
715 PH/42 below, there is no longer a chance of it appearing in a bounce
718 PH/11 Installed exipick version 20050225.0 from John Jetmore.
720 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
721 of its addresses was ever tried. (Bugzilla bug #2.)
723 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
724 the result incorrectly in the debug output. (It correctly added a newline
725 to what was transported.)
727 TF/01 Added $received_time.
729 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
730 commented out examples of how to interface to a virus scanner and to
731 SpamAssassin. Also added commented examples of av_scanner and
732 spamd_address settings.
734 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
735 and controls are allowed in which ACLs. There were a couple of minor
736 errors. Some of the entries in the conditions table (which is a table of
737 where they are NOT allowed) were getting very unwieldy; rewrote them as a
738 negation of where the condition IS allowed.
740 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
742 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
743 header file does not have a version number, so I've had to invent a new
744 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
745 API. The code is untested by me (my Linux distribution still has 0.3.2 of
746 radiusclient), but it was contributed by a Radius user.
748 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
749 files or directories, for OpenSSL.
751 PH/19 When an Exim process that is running as root has to create an Exim log
752 file, it does so in a subprocess that runs as exim:exim so as to get the
753 ownership right at creation (otherwise, other Exim processes might see
754 the file with the wrong ownership). There was no test for failure of this
755 fork() call, which would lead to the process getting stuck as it waited
756 for a non-existent subprocess. Forks do occasionally fail when resources
757 run out. I reviewed all the other calls to fork(); they all seem to check
760 PH/20 When checking for unexpected SMTP input at connect time (before writing
761 the banner), Exim was not dealing correctly with a non-positive return
762 from the read() function. If the client had disconnected by this time,
763 the result was a log entry for a synchronization error with an empty
764 string after "input=" when read() returned zero. If read() returned -1
765 (an event I could not check), uninitialized data bytes were printed.
766 There were reports of junk text (parts of files, etc) appearing after
769 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
771 PH/22 Added support for macro redefinition, and (re)definition in between
772 driver and ACL definitions.
774 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
775 forgetting to use the resulting value; it was using the unexpanded value.
777 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
778 hadn't been configured. The fix is from Juergen Kreileder, who
779 understands it better than I do:
781 "Here's what I see happening with three configured cyrus_sasl
782 authenticators configured (plain, login, cram-md5):
784 On startup auth_cyrus_sasl_init() gets called for each of these.
785 This means three calls to sasl_listmech() without a specified mech_list.
786 => SASL tests which mechs of all available mechs actually work
787 => three warnings about OTP not working
788 => the returned list contains: plain, login, cram-md5, digest-md5, ...
790 With the patch, sasl_listmech() also gets called three times. But now
791 SASL's mech_list option is set to the server_mech specified in the the
792 authenticator. Or in other words, the answer from sasl_listmech()
793 gets limited to just the mech you're testing for (which is different
795 => the return list contains just 'plain' or 'login', 'cram-md5' or
796 nothing depending on the value of ob->server_mech.
798 I've just tested the patch: Authentication still works fine,
799 unavailable mechs specified in the exim configuration are still
800 caught, and the auth.log warnings about OTP are gone."
802 PH/25 When debugging is enabled, the contents of the command line are added
803 to the debugging output, even when log_selector=+arguments is not
806 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
807 answer is "GNU", and only if the return is "GNU/something" is the answer
810 PH/27 $acl_verify_message is now set immediately after the failure of a
811 verification in an ACL, and so is available in subsequent modifiers. In
812 particular, the message can be preserved by coding like this:
814 warn !verify = sender
815 set acl_m0 = $acl_verify_message
817 Previously, $acl_verify_message was set only while expanding "message"
818 and "log_message" when a very denied access.
820 PH/28 Modified OS/os.c-Linux with
822 -#ifndef OS_LOAD_AVERAGE
823 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
825 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
826 nomenclature these days.)
828 PH/29 Installed patch from the Sieve maintainer that adds the options
829 sieve_useraddress and sieve_subaddress to the redirect router.
831 PH/30 In these circumstances:
832 . Two addresses routed to the same list of hosts;
833 . First host does not offer TLS;
834 . First host accepts first address;
835 . First host gives temporary error to second address;
836 . Second host offers TLS and a TLS session is established;
837 . Second host accepts second address.
838 Exim incorrectly logged both deliveries with the TLS parameters (cipher
839 and peerdn, if requested) that were in fact used only for the second
842 PH/31 When doing a callout as part of verifying an address, Exim was not paying
843 attention to any local part prefix or suffix that was matched by the
844 router that accepted the address. It now behaves in the same way as it
845 does for delivery: the affixes are removed from the local part unless
846 rcpt_include_affixes is set on the transport.
848 PH/32 Add the sender address, as F=<...>, to the log line when logging a
849 timeout during the DATA phase of an incoming message.
851 PH/33 Sieve envelope tests were broken for match types other than :is. I have
852 applied a patch sanctioned by the Sieve maintainer.
854 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
855 the uid or gid is negative. A case of a negative gid caused this to be
856 noticed. The fix allows for either to be negative.
858 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
859 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
862 PH/36 The change PH/12 above was broken. Fixed it.
864 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
865 the grounds that routing the same address twice would always produce the
866 same answer. This might have been true once, but it is certainly no
867 longer true now. Routing a child address may depend on the previous
868 routing that produced that child. Some complicated redirection strategies
869 went wrong when messages had multiple recipients, and made Exim's
870 behaviour dependent on the order in which the addresses were given.
872 I have moved the duplicate checking until after the routing is complete.
873 Exim scans the addresses that are assigned to local and remote
874 transports, and removes any duplicates. This means that more work will be
875 done, as duplicates will always all be routed, but duplicates are
876 presumably rare, so I don't expect this is of any significance.
878 For deliveries to pipes, files, and autoreplies, the duplicate checking
879 still happens during the routing process, since they are not going to be
882 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
883 It corrects a timeout issue with spamd. This is Ian's comment: "The
884 background is that sometimes spamd either never reads data from a
885 connection it has accepted, or it never writes response data. The exiscan
886 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
887 blindly assumes that writes won't block so it may never time out."
889 PH/39 Allow G after quota size as well as K and M.
891 PH/40 The value set for $authenticated_id in an authenticator may not contain
892 binary zeroes or newlines because the value is written to log lines and
893 to spool files. There was no check on this. Now the value is run through
894 the string_printing() function so that such characters are converted to
895 printable escape sequences.
897 PH/41 $message_linecount is a new variable that contains the total number of
898 lines in the message. Compare $body_linecount, which is the count for the
901 PH/42 Exim no longer gives details of delivery errors for specific addresses in
902 bounce and delay warning messages, except in certain special cases, which
905 (a) An SMTP error message from a remote host;
906 (b) A message specified in a :fail: redirection;
907 (c) A message specified in a "fail" command in a system filter;
908 (d) A message specified in a FAIL return from the queryprogram router;
909 (e) A message specified by the cannot_route_message router option.
911 In these cases only, Exim does include the error details in bounce and
912 warning messages. There are also a few cases where bland messages such
913 as "unrouteable address" or "local delivery error" are given.
915 PH/43 $value is now also set for the "else" part of a ${run expansion.
917 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
918 being worked on, but at least Exim now implements the latest version to
921 PH/45 In a pipe transport, although a timeout while waiting for the pipe
922 process to complete was treated as a delivery failure, a timeout while
923 writing the message to the pipe was logged, but erroneously treated as a
924 successful delivery. Such timeouts include transport filter timeouts. For
925 consistency with the overall process timeout, these timeouts are now
926 treated as errors, giving rise to delivery failures by default. However,
927 there is now a new Boolean option for the pipe transport called
928 timeout_defer, which, if set TRUE, converts the failures into defers for
929 both kinds of timeout. A transport filter timeout is now identified in
932 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
933 systems where "make" and "gmake" are different, calling "gmake" at top
934 level broke things. I've arranged for the value of $(MAKE) to be passed
935 from the Makefile to this script so that it can call the same version of
939 A note about Exim versions 4.44 and 4.50
940 ----------------------------------------
942 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
943 changes of various kinds. As a consequence, a big documentation update was
944 needed. This delayed the release for rather longer than seemed good, especially
945 in the light of a couple of (minor) security issues. Therefore, the changes
946 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
947 release. So 4.44 and 4.50 are in effect two different branches that both start
950 I have left the 4.50 change log unchanged; it contains all the changes since
951 4.43. The change log for 4.44 is below; many of its items are identical to
952 those for 4.50. This seems to be the most sensible way to preserve the
953 historical information.
959 1. Minor wording change to the doc/README.SIEVE file.
961 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
962 computation of the current number of files was incorrect.
964 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
965 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
968 4. Give more explanation in the error message when the command for a transport
969 filter fails to execute.
971 5. There are several places where Exim runs a non-Exim command in a
972 subprocess. The SIGUSR1 signal should be disabled for these processes. This
973 was being done only for the command run by the queryprogram router. It is
974 now done for all such subprocesses. The other cases are: ${run, transport
975 filters, and the commands run by the lmtp and pipe transports.
977 6. Added CONFIGURE_GROUP build-time option.
979 7. Some older OS have a limit of 256 on the maximum number of file
980 descriptors. Exim was using setrlimit() to set 1000 as a large value
981 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
982 systems. I've change it so that if it can't get 1000, it tries for 256.
984 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
985 was an oversight, and furthermore, ever since the addition of extra
986 controls (e.g. 4.43/32), the checks on when to allow different forms of
987 "control" were broken. There should now be diagnostics for all cases when a
988 control that does not make sense is encountered.
990 9. Added the /retain_sender option to "control=submission".
992 10. $recipients is now available in the predata ACL (oversight).
994 11. Tidy the search cache before the fork to do a delivery from a message
995 received from the command line. Otherwise the child will trigger a lookup
996 failure and thereby defer the delivery if it tries to use (for example) a
997 cached ldap connection that the parent has called unbind on.
999 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1000 of $address_data from the recipient verification was clobbered by the
1001 sender verification.
1003 13. The value of address_data from a sender verification is now available in
1004 $sender_address_data in subsequent conditions in the ACL statement.
1006 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1008 15. Added a new option "connect=<time>" to callout options, to set a different
1011 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1012 was its contents. (It was OK if the option was not defined at all.)
1014 17. A "Completed" log line is now written for messages that are removed from
1015 the spool by the -Mrm option.
1017 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1018 information about exactly what failed.
1020 19. Added -dd to debug only the daemon process.
1022 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1023 handles timeouts, both on the server side and network timeouts. Renamed the
1024 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1026 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1027 It is now set to "smtps".
1029 22. $host_address is now set to the target address during the checking of
1030 ignore_target_hosts.
1032 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1033 being passed; this would have caused $sender_host_name to have been used if
1034 matching the list had actually called for a host name (not very likely,
1035 since this list is usually IP addresses). A host name is now passed as
1038 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1039 code that specifies a non-restarting handler (typically sigaction() in
1040 modern systems) in an attempt to fix a rare and obscure crash bug.
1042 25. Narrowed the window for a race in the daemon that could cause it to ignore
1043 SIGCHLD signals. This is not a major problem, because they are used only to
1044 wake it up if nothing else does.
1046 26. A malformed maildirsize file could cause Exim to calculate negative values
1047 for the mailbox size or file count. Odd effects could occur as a result.
1048 The maildirsize information is now recalculated if the size or filecount
1051 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1052 support for a long time. Removed HAVE_SYS_VFS_H.
1054 28. Installed the latest version of exipick from John Jetmore.
1056 29. In an address list, if the pattern was not a regular expression, an empty
1057 subject address (from a bounce message) matched only if the pattern was an
1058 empty string. Non-empty patterns were not even tested. This was the wrong
1059 because it is perfectly reasonable to use an empty address as part of a
1060 database query. An empty address is now tested by patterns that are
1061 lookups. However, all the other forms of pattern expect the subject to
1062 contain a local part and a domain, and therefore, for them, an empty
1063 address still always fails if the pattern is not itself empty.
1065 30. Exim went into a mad DNS loop when attempting to do a callout where the
1066 host was specified on an smtp transport, and looking it up yielded more
1067 than one IP address.
1069 31. Re-factored the code for checking spool and log partition space into a
1070 function that finds that data and another that does the check. The former
1071 is then used to implement four new variables: $spool_space, $log_space,
1072 $spool_inodes, and $log_inodes.
1074 32. The RFC2047 encoding function was originally intended for short strings
1075 such as real names; it was not keeping to the 75-character limit for
1076 encoded words that the RFC imposes. It now respects the limit, and
1077 generates multiple encoded words if necessary. To be on the safe side, I
1078 have increased the buffer size for the ${rfc2047: expansion operator from
1081 33. It is now permitted to omit both strings after an "if" condition; if the
1082 condition is true, the result is "true". As before, when the second string
1083 is omitted, a false condition yields an empty string. This makes it less
1084 cumbersome to write custom ACL and router conditions.
1086 34. Failure to deliver a bounce message always caused it to be frozen, even if
1087 there was an errors_to setting on the router. The errors_to setting is now
1090 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1091 canonical form (fully expanded) before being placed in
1092 $sender_host_address.
1094 36. The table in the code that translates DNS record types into text (T_A to
1095 "A" for instance) was missing entries for NS and CNAME. It is just possible
1096 that this could have caused confusion if both these types were looked up
1097 for the same domain, because the text type is used as part of Exim's
1098 per-process caching. But the chance of anyone hitting this buglet seems
1101 37. The dnsdb lookup has been extended in a number of ways.
1103 (1) There is a new type, "zns", which walks up the domain tree until it
1104 finds some nameserver records. It should be used with care.
1106 (2) There is a new type, "mxh", which is like "mx" except that it returns
1107 just the host names, not the priorities.
1109 (3) It is now possible to give a list of domains (or IP addresses) to be
1110 looked up. The behaviour when one of the lookups defers can be
1111 controlled by a keyword.
1113 (4) It is now possible to specify the separator character for use when
1114 multiple records are returned.
1116 38. The dnslists ACL condition has been extended: it is now possible to supply
1117 a list of IP addresses and/or domains to be looked up in a particular DNS
1120 39. Added log_selector=+queue_time_overall.
1122 40. When running the queue in the test harness, wait just a tad after forking a
1123 delivery process, to get repeatability of debugging output.
1125 41. Include certificate and key file names in error message when GnuTLS fails
1126 to set them up, because the GnuTLS error message doesn't include the name
1127 of the failing file when there is a problem reading it.
1129 42. Allow both -bf and -bF in the same test run.
1131 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1133 44. The "Exiscan patch" is now merged into the mainline Exim source.
1135 45. Sometimes the final signoff response after QUIT could fail to get
1136 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1137 < 0 before doing a fflush(). This bug looks as though it goes back to the
1138 introduction of TLS in release 3.20, but "sometimes" must have been rare
1139 because the tests only now provoked it.
1141 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1142 (this can affect the format of dates).
1144 47. exim_tidydb, when checking for the continued existence of a message for
1145 which it has found a message-specific retry record, was not finding
1146 messages that were in split spool directories. Consequently, it was
1147 deleting retry records that should have stayed in existence.
1149 48. Steve fixed some bugs in eximstats.
1151 49. The SPA authentication driver was not abandoning authentication and moving
1152 on to the next authenticator when an expansion was forced to fail,
1153 contradicting the general specification for all authenticators. Instead it
1154 was generating a temporary error. It now behaves as specified.
1156 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1157 (the order specifies the preference for clients). The order is now AES256,
1158 AES128, 3DES, ARCFOUR128.
1160 51. Small patch to Sieve code - explicitly set From: when generating an
1163 52. Exim crashed if a remote delivery caused a very long error message to be
1164 recorded - for instance if somebody sent an entire SpamAssassin report back
1165 as a large number of 550 error lines. This bug was coincidentally fixed by
1166 increasing the size of one of Exim's internal buffers (big_buffer) that
1167 happened as part of the Exiscan merge. However, to be on the safe side, I
1168 have made the code more robust (and fixed the comments that describe what
1171 53. Now that there can be additional text after "Completed" in log lines (if
1172 the queue_time_overall log selector is set), a one-byte patch to exigrep
1173 was needed to allow it to recognize "Completed" as not the last thing in
1176 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1177 patch that reportedly fixes this has been added. I am not expert enough to
1178 create a test for it. This is what the patch creator wrote:
1180 "I found a little strange behaviour of ldap code when working with
1181 Windows 2003 AD Domain, where users was placed in more than one
1182 Organization Units. When I tried to give exim partial DN, the exit code
1183 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1184 But simultaneously result of request was absolutely normal ldap result,
1185 so I produce this patch..."
1187 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1188 REFERENCE, so I have modified the code to exclude the patch when that macro
1191 55. Some experimental protocols are using DNS PTR records for new purposes. The
1192 keys for these records are domain names, not reversed IP addresses. The
1193 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1194 leaves it alone. Component reversal etc. now happens only for IP addresses.
1197 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1199 57. Double the size of the debug message buffer (to 2048) so that more of very
1200 long debug lines gets shown.
1202 58. The exicyclog utility now does better if the number of log files to keep
1203 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1205 59. Two changes related to the smtp_active_hostname option:
1207 (1) $smtp_active_hostname is now available as a variable.
1208 (2) The default for smtp_banner uses $smtp_active_hostname instead
1209 of $primary_hostname.
1211 60. The host_aton() function is supposed to be passed a string that is known
1212 to be a valid IP address. However, in the case of IPv6 addresses, it was
1213 not checking this. This is a hostage to fortune. Exim now panics and dies
1214 if the condition is not met. A case was found where this could be provoked
1215 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1216 components; fortuitously, this particular loophole had already been fixed
1217 by change 4.50/55 above.
1219 If there are any other similar loopholes, the new check in host_aton()
1220 itself should stop them being exploited. The report I received stated that
1221 data on the command line could provoke the exploit when Exim was running as
1222 exim, but did not say which command line option was involved. All I could
1223 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1224 running as the user.
1227 61. There was a buffer overflow vulnerability in the SPA authentication code
1228 (which came originally from the Samba project). I have added a test to the
1229 spa_base64_to_bits() function which I hope fixes it.
1232 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1233 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1235 63. The daemon start-up calls getloadavg() while still root for those OS that
1236 need the first call to be done as root, but it missed one case: when
1237 deliver_queue_load_max is set with deliver_drop_privilege. This is
1238 necessary for the benefit of the queue runner, because there is no re-exec
1239 when deliver_drop_privilege is set.
1241 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1242 This has been fixed.
1244 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1245 was in use, was not putting the data itself into the right store pool;
1246 consequently, it could be overwritten for a subsequent message in the same
1247 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1250 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1252 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1253 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1254 the function were treating the return as a boolean value, which happened to
1255 work because 0=false and not-0=true, but is not correct code.
1257 68. The host_aton() function was not handling scoped IPv6 addresses (those
1258 with, for example, "%eth0" on the end) correctly.
1260 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1261 negated items (that is, ~something) in unsigned ints. Some compilers
1262 apparently mutter when there is no cast.
1264 70. If an address verification called from an ACL failed, and did not produce a
1265 user-specific message (i.e. there was only a "system" message), nothing was
1266 put in $acl_verify_message. In this situation, it now puts the system
1269 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1270 session; change 4.31/43 added the unwanted input to the log line - except
1271 that it did not do this in the start of session case. It now does.
1273 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1274 This is wrong and can cause the other end to generate a synchronization
1275 error if it is another Exim or anything else that does the synchronization
1276 check. A QUIT command is no longer sent after a timeout.
1278 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1279 during host lookups.
1281 74. The defer_ok option of callout verification was not working if it was used
1282 when verifying addresses in header lines, that is, for this case:
1284 verify = header_sender/callout=defer_ok
1286 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1287 those file descriptors could be used for SMTP connections. If anything
1288 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1289 could be sent to the SMTP client, causing chaos. The daemon now opens
1290 stdin, stdout, and stderr to /dev/null when it puts itself into the
1293 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1294 log by default. The user can override this with suitable Perl magic.
1296 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1297 the log message when discard triggers, was not working for the DATA ACL or
1298 for the non-SMTP ACL.
1300 78. Error message wording change in sieve.c.
1302 79. If smtp_accept_max_per_host was set, the number of connections could be
1303 restricted to fewer than expected, because the daemon was trying to set up
1304 a new connection before checking whether the processes handling previous
1305 connections had finished. The check for completed processes is now done
1306 earlier. On busy systems, this bug wouldn't be noticed because something
1307 else would have woken the daemon, and it would have reaped the completed
1310 80. If a message was submitted locally by a user whose login name contained one
1311 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1312 It caused a spool format error. I have fixed the spool reading code. A
1313 related problem was that the "from" clause in the Received: line became
1314 illegal because of the space(s). It is now covered by ${quote_local_part.
1316 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1319 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1321 83. Patch from Sieve maintainer to fix unterminated string problem in
1322 "vacation" handling.
1324 84. Some minor changes to the Linux configuration files to help with other
1325 OS variants using glibc.
1327 85. One more patch for Sieve to update vacation handling to latest spec.
1330 ----------------------------------------------------
1331 See the note above about the 4.44 and 4.50 releases.
1332 ----------------------------------------------------
1338 1. Change 4.43/35 introduced a bug that caused file counts to be
1339 incorrectly computed when quota_filecount was set in an appendfile
1342 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1343 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1346 3. Give more explanation in the error message when the command for a transport
1347 filter fails to execute.
1349 4. There are several places where Exim runs a non-Exim command in a
1350 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1351 was being done only for the command run by the queryprogram router. It is
1352 now done for all such subprocesses. The other cases are: ${run, transport
1353 filters, and the commands run by the lmtp and pipe transports.
1355 5. Some older OS have a limit of 256 on the maximum number of file
1356 descriptors. Exim was using setrlimit() to set 1000 as a large value
1357 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1358 systems. I've change it so that if it can't get 1000, it tries for 256.
1360 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1361 was an oversight, and furthermore, ever since the addition of extra
1362 controls (e.g. 4.43/32), the checks on when to allow different forms of
1363 "control" were broken. There should now be diagnostics for all cases when a
1364 control that does not make sense is encountered.
1366 7. $recipients is now available in the predata ACL (oversight).
1368 8. Tidy the search cache before the fork to do a delivery from a message
1369 received from the command line. Otherwise the child will trigger a lookup
1370 failure and thereby defer the delivery if it tries to use (for example) a
1371 cached ldap connection that the parent has called unbind on.
1373 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1374 of $address_data from the recipient verification was clobbered by the
1375 sender verification.
1377 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1378 was its contents. (It was OK if the option was not defined at all.)
1380 11. A "Completed" log line is now written for messages that are removed from
1381 the spool by the -Mrm option.
1383 12. $host_address is now set to the target address during the checking of
1384 ignore_target_hosts.
1386 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1387 being passed; this would have caused $sender_host_name to have been used if
1388 matching the list had actually called for a host name (not very likely,
1389 since this list is usually IP addresses). A host name is now passed as
1392 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1393 code that specifies a non-restarting handler (typically sigaction() in
1394 modern systems) in an attempt to fix a rare and obscure crash bug.
1396 15. Narrowed the window for a race in the daemon that could cause it to ignore
1397 SIGCHLD signals. This is not a major problem, because they are used only to
1398 wake it up if nothing else does.
1400 16. A malformed maildirsize file could cause Exim to calculate negative values
1401 for the mailbox size or file count. Odd effects could occur as a result.
1402 The maildirsize information is now recalculated if the size or filecount
1405 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1406 support for a long time. Removed HAVE_SYS_VFS_H.
1408 18. Updated exipick to current release from John Jetmore.
1410 19. Allow an empty sender to be matched against a lookup in an address list.
1411 Previously the only cases considered were a regular expression, or an
1414 20. Exim went into a mad DNS lookup loop when doing a callout where the
1415 host was specified on the transport, if the DNS lookup yielded more than
1418 21. The RFC2047 encoding function was originally intended for short strings
1419 such as real names; it was not keeping to the 75-character limit for
1420 encoded words that the RFC imposes. It now respects the limit, and
1421 generates multiple encoded words if necessary. To be on the safe side, I
1422 have increased the buffer size for the ${rfc2047: expansion operator from
1425 22. Failure to deliver a bounce message always caused it to be frozen, even if
1426 there was an errors_to setting on the router. The errors_to setting is now
1429 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1430 canonical form (fully expanded) before being placed in
1431 $sender_host_address.
1433 24. Updated eximstats to version 1.33
1435 25. Include certificate and key file names in error message when GnuTLS fails
1436 to set them up, because the GnuTLS error message doesn't include the name
1437 of the failing file when there is a problem reading it.
1439 26. Expand error message when OpenSSL has problems setting up cert/key files.
1442 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1443 (this can affect the format of dates).
1445 28. exim_tidydb, when checking for the continued existence of a message for
1446 which it has found a message-specific retry record, was not finding
1447 messages that were in split spool directories. Consequently, it was
1448 deleting retry records that should have stayed in existence.
1450 29. eximstats updated to version 1.35
1451 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1452 1.35 - bugfix such that pie charts by volume are generated correctly
1454 30. The SPA authentication driver was not abandoning authentication and moving
1455 on to the next authenticator when an expansion was forced to fail,
1456 contradicting the general specification for all authenticators. Instead it
1457 was generating a temporary error. It now behaves as specified.
1459 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1460 (the order specifies the preference for clients). The order is now AES256,
1461 AES128, 3DES, ARCFOUR128.
1463 31. Small patch to Sieve code - explicitly set From: when generating an
1466 32. Exim crashed if a remote delivery caused a very long error message to be
1467 recorded - for instance if somebody sent an entire SpamAssassin report back
1468 as a large number of 550 error lines. This bug was coincidentally fixed by
1469 increasing the size of one of Exim's internal buffers (big_buffer) that
1470 happened as part of the Exiscan merge. However, to be on the safe side, I
1471 have made the code more robust (and fixed the comments that describe what
1474 33. Some experimental protocols are using DNS PTR records for new purposes. The
1475 keys for these records are domain names, not reversed IP addresses. The
1476 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1477 leaves it alone. Component reversal etc. now happens only for IP addresses.
1480 34. The host_aton() function is supposed to be passed a string that is known
1481 to be a valid IP address. However, in the case of IPv6 addresses, it was
1482 not checking this. This is a hostage to fortune. Exim now panics and dies
1483 if the condition is not met. A case was found where this could be provoked
1484 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1485 components; fortuitously, this particular loophole had already been fixed
1486 by change 4.50/55 or 4.44/33 above.
1488 If there are any other similar loopholes, the new check in host_aton()
1489 itself should stop them being exploited. The report I received stated that
1490 data on the command line could provoke the exploit when Exim was running as
1491 exim, but did not say which command line option was involved. All I could
1492 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1493 running as the user.
1496 35. There was a buffer overflow vulnerability in the SPA authentication code
1497 (which came originally from the Samba project). I have added a test to the
1498 spa_base64_to_bits() function which I hope fixes it.
1501 36. The daemon start-up calls getloadavg() while still root for those OS that
1502 need the first call to be done as root, but it missed one case: when
1503 deliver_queue_load_max is set with deliver_drop_privilege. This is
1504 necessary for the benefit of the queue runner, because there is no re-exec
1505 when deliver_drop_privilege is set.
1507 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1508 was in use, was not putting the data itself into the right store pool;
1509 consequently, it could be overwritten for a subsequent message in the same
1510 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1513 38. Sometimes the final signoff response after QUIT could fail to get
1514 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1515 < 0 before doing a fflush(). This bug looks as though it goes back to the
1516 introduction of TLS in release 3.20, but "sometimes" must have been rare
1517 because the tests only now provoked it.
1523 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1524 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1525 Now it returns an integer. A number of calls were still expecting a T/F
1526 return. Fortuitously, in all cases, the tests worked in OK situations,
1527 which is the norm. However, things would have gone wrong on any write
1528 failures on the smtp file descriptor. This function is used when sending
1529 messages over SMTP and also when doing verify callouts.
1531 2. When Exim is called to do synchronous delivery of a locally submitted
1532 message (the -odf or -odi options), it no longer closes stderr before doing
1535 3. Implemented the mua_wrapper option.
1537 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1539 5. Implemented the functions header_remove(), header_testname(),
1540 header_add_at_position(), and receive_remove_recipient(), and exported them
1543 6. If an ACL "warn" statement specified the addition of headers, Exim already
1544 inserted X-ACL-Warn: at the start if there was no header name. However, it
1545 was not making this test for the second and subsequent header lines if
1546 there were newlines in the string. This meant that an invalid header could
1547 be inserted if Exim was badly configured.
1549 7. Allow an ACL "warn" statement to add header lines at the start or after all
1550 the Received: headers, as well as at the end.
1552 8. Added the rcpt_4xx retry error code.
1554 9. Added postmaster_mailfrom=xxx to callout verification option.
1556 10. Added mailfrom=xxxx to the callout verification option, for verify=
1559 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1560 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1562 12. Inserted some casts to stop certain compilers warning when using pointer
1563 differences as field lengths or precisions in printf-type calls (mostly
1564 affecting debugging statements).
1566 13. Added optional readline() support for -be (dynamically loaded).
1568 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1569 same clock tick as a message's arrival, so that its received time was the
1570 same as the "first fail" time on the retry record, and that message
1571 remained on the queue past the ultimate address timeout, every queue runner
1572 would try a delivery (because it was past the ultimate address timeout) but
1573 after another failure, the ultimate address timeout, which should have then
1574 bounced the address, did not kick in. This was a "< instead of <=" error;
1575 in most cases the first failure would have been in the next clock tick
1576 after the received time, and all would be well.
1578 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1579 being recognized when the domain list was tested by the match_domain
1580 condition in an expansion string.
1582 16. Added the ${str2b64: operator.
1584 17. Exim was always calling setrlimit() to set a large limit for the number of
1585 processes, without checking whether the existing limit was already
1586 adequate. (It did check for the limit on file descriptors.) Furthermore,
1587 errors from getrlimit() and setrlimit() were being ignored. Now they are
1588 logged to the main and panic logs, but Exim does carry on, to try to do its
1589 job under whatever limits there are.
1591 18. Imported PCRE 5.0.
1593 19. Trivial typo in log message " temporarily refused connection" (the leading
1596 20. If the log selector return_path_on_delivery was set and an address was
1597 redirected to /dev/null, the delivery process crashed because it assumed
1598 that a return path would always be set for a "successful" delivery. In this
1599 case, the whole delivery is bypassed as an optimization, and therefore no
1602 21. Internal re-arrangement: the function for sending a challenge and reading
1603 a response while authentication was assuming a zero-terminated challenge
1604 string. It's now changed to take a pointer and a length, to allow for
1605 binary data in such strings.
1607 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1609 23. Exim was not respecting finduser_retries when seeking the login of the
1610 uid under which it was called; it was always trying 10 times. (The default
1611 setting of finduser_retries is zero.) Also, it was sleeping after the final
1612 failure, which is pointless.
1614 24. Implemented tls_on_connect_ports.
1616 25. Implemented acl_smtp_predata.
1618 26. If the domain in control=submission is set empty, Exim assumes that the
1619 authenticated id is a complete email address when it generates From: or
1620 Sender: header lines.
1622 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1623 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1624 chown and chgrp in /bin and hostname in /usr/bin.
1626 28. Exim was keeping the "process log" file open after each use, just as it
1627 does for the main log. This opens the possibility of it remaining open for
1628 long periods when the USR1 signal hits a daemon. Occasional processlog
1629 errors were reported, that could have been caused by this. Anyway, it seems
1630 much more sensible not to leave this file open at all, so that is what now
1633 29. The long-running daemon process does not normally write to the log once it
1634 has entered its main loop, and it closes the log before doing so. This is
1635 so that log files can straightforwardly be renamed and moved. However,
1636 there are a couple of unusual error situations where the daemon does write
1637 log entries, and I had neglected to close the log afterwards.
1639 30. The text of an SMTP error response that was received during a remote
1640 delivery was being truncated at 512 bytes. This is too short for some of
1641 the long messages that one sometimes sees. I've increased the limit to
1644 31. It is now possible to make retry rules that apply only when a message has a
1645 specific sender, in particular, an empty sender.
1647 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1648 it possible to be selective about when SMTP synchronization is enforced.
1650 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1652 32. Implemented hosts_connection_nolog.
1654 33. Added an ACL for QUIT.
1656 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1659 35. Added mailbox_size and mailbox_filecount to appendfile.
1661 36. Added control = no_multiline_responses to ACLs.
1663 37. There was a bug in the logic of the code that waits for the clock to tick
1664 in the case where the clock went backwards by a substantial amount such
1665 that the microsecond fraction of "now" was more than the microsecond
1666 fraction of "then" (but the whole seconds number was less).
1668 38. Added support for the libradius Radius client library this is found on
1669 FreeBSD (previously only the radiusclient library was supported).
1675 1. When certain lookups returned multiple values in the form name=value, the
1676 quoting of the values was not always being done properly. Specifically:
1677 (a) If the value started with a double quote, but contained no whitespace,
1679 (b) If the value contained whitespace other than a space character (i.e.
1680 tabs or newlines or carriage returns) it was not quoted.
1681 This fix has been applied to the mysql and pgsql lookups by writing a
1682 separate quoting function and calling it from the lookup code. The fix
1683 should probably also be applied to nisplus, ibase and oracle lookups, but
1684 since I cannot test any of those, I have not disturbed their existing code.
1686 2. A hit in the callout cache for a specific address caused a log line with no
1687 reason for rejecting RCPT. Now it says "Previous (cached) callout
1688 verification failure".
1690 3. There was an off-by-one bug in the queryprogram router. An over-long
1691 return line was truncated at 256 instead of 255 characters, thereby
1692 overflowing its buffer with the terminating zero. As well as fixing this, I
1693 have increased the buffer size to 1024 (and made a note to document this).
1695 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1696 when Exim is waiting for an SMTP response from a remote server, Exim
1697 restarts its select() call on the socket, thereby resetting its timeout.
1698 This is not a problem when such interrupts are rare. Somebody set up a cron
1699 job to run exiwhat every 2 minutes, which is less than the normal select()
1700 timeout (5 or 10 minutes). This meant that the select() timeout never
1701 kicked in because it was always reset. I have fixed this by comparing the
1702 time when an interrupt arrives with the time at the start of the first call
1703 to select(). If more time than the timeout has elapsed, the interrupt is
1704 treated as a timeout.
1706 5. Some internal re-factoring in preparation for the addition of Sieve
1707 extensions (by MH). In particular, the "personal" test is moved to a
1708 separate function, and given an option for scanning Cc: and Bcc: (which is
1709 not set for Exim filters).
1711 6. When Exim created an email address using the login of the caller as the
1712 local part (e.g. when creating a From: or Sender: header line), it was not
1713 quoting the local part when it contained special characters such as @.
1715 7. Installed new OpenBSD configuration files.
1717 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1718 try to make them clearer.
1720 9. Callout options, other than the timeout value, were being ignored when
1721 verifying sender addresses in header lines. For example, when using
1723 verify = header_sender/callout=no_cache
1725 the cache was (incorrectly) being used.
1727 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1728 only the Cygwin environment.
1730 11. When return_path_on_delivery was set as a log selector, if different remote
1731 addresses in the same message used different return paths and parallel
1732 remote delivery occurred, the wrong values would sometimes be logged.
1733 (Whenever a remote delivery process finished, the return path value from
1734 the most recently started remote delivery process was logged.)
1736 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1737 lines when AUTH and/or TLS are in use. This is the "received protocol"
1738 field. Exim used to use "asmtp" for authenticated SMTP, without any
1739 indication (in the protocol name) for TLS use. Now it follows the RFC and
1740 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1741 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1742 names appear in log lines as well as in Received: header lines.
1744 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1745 extensions, and comparison tests, and to fix some bugs.
1747 14. Changes to the "personal" filter test:
1749 (1) The test was buggy in that it was just doing the equivalent of
1750 "contains" tests on header lines. For example, if a user's address was
1751 anne@some.where, the "personal" test would incorrectly be true for
1753 To: susanne@some.where
1755 This test is now done by extracting each address from the header in turn,
1756 and checking the entire address. Other tests that are part of "personal"
1757 are now done using regular expressions (for example, to check local parts
1758 of addresses in From: header lines).
1760 (2) The list of non-personal local parts in From: addresses has been
1761 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1762 taken from the Sieve specification recommendations.
1764 (3) If the message contains any header line starting with "List-" it is
1765 treated as non-personal.
1767 (4) The test for "circular" in the Subject: header line has been removed
1768 because it now seems ill-conceived.
1770 15. Minor typos in src/EDITME comments corrected.
1772 16. Installed latest exipick from John Jetmore.
1774 17. If headers_add on a router specified a text string that was too long for
1775 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1776 of string_sprintf() is now avoided.
1778 18. $message_body_size was not set (it was always zero) when running the DATA
1779 ACL and the local_scan() function.
1781 19. For the "mail" command in an Exim filter, no default was being set for
1782 the once_repeat time, causing a random time value to be used if "once" was
1783 specified. (If the value happened to be <= 0, no repeat happened.) The
1784 default is now 0s, meaning "never repeat". The "vacation" command was OK
1785 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1786 (I found it when inspecting the code).
1788 20. There is now an overall timeout for performing a callout verification. It
1789 defaults to 4 times the callout timeout, which applies to individual SMTP
1790 commands during the callout. The overall timeout applies when there is more
1791 than one host that can be tried. The timeout is checked before trying the
1792 next host. This prevents very long delays if there are a large number of
1793 hosts and all are timing out (e.g. when the network connections are timing
1794 out). The value of the overall timeout can be changed by specifying an
1795 additional sub-option for "callout", called "maxwait". For example:
1797 verify = sender/callout=5s,maxwait=20s
1799 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1800 to the end before writing, but this should make it even safer).
1802 22. Exim was forgetting that it had advertised PIPELINING for the second and
1803 subsequent messages on an SMTP connection. It was also not resetting its
1804 memory on STARTTLS and an internal HELO.
1806 23. When Exim logs an SMTP synchronization error within a session, it now
1807 records whether PIPELINING has been advertised or not.
1809 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1810 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1811 rather than long int.
1813 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1815 26. Added the never_mail option to autoreply.
1821 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1822 crash if the getsockname() call failed; this can happen if a connection is
1823 closed very soon after it is established. The problem was simply in the
1824 order in which certain operations were done, causing Exim to try to write
1825 to the SMTP stream before it had set up the file descriptor. The bug has
1826 been fixed by making things happen in the correct order.
1832 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1833 before the connection was closed, thus losing the rejection response.
1835 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1836 some early Solaris releases, but causes trouble in current releases where
1837 socklen_t is defined.
1839 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1842 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1845 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1846 that releasing the top of it at the end releases what was used for sub-
1847 expansions (unless the block got too big). However, discard this block if
1848 the first thing is a variable or header, so that we can use its block when
1849 it is dynamic (useful for very large $message_headers, for example).
1851 6. Lookups now cache *every* query, not just the most recent. A new, separate
1852 store pool is used for this. It can be recovered when all lookup caches are
1853 flushed. Lookups now release memory at the end of their result strings.
1854 This has involved some general refactoring of the lookup sources.
1856 7. Some code has been added to the store_xxx() functions to reduce the amount
1857 of flapping under certain conditions.
1859 8. log_incoming_interface used to affect only the <= reception log lines. Now
1860 it causes the local interface and port to be added to several more SMTP log
1861 lines, for example "SMTP connection from", and rejection lines.
1863 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1865 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1867 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1868 could be overwritten at the end of the current message (or the start of a
1869 new message if it was set in a HELO ACL). The value is now preserved for
1870 the duration of the SMTP connection.
1872 12. If a transport had a headers_rewrite setting, and a matching header line
1873 contained an unqualified address, that address was qualified, even if it
1874 did not match any rewriting rules. The underlying bug was that the values
1875 of the flags that permit the existence of unqualified sender and recipient
1876 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1877 non-local messages, and by -bnq for local messages) were not being
1878 preserved with the message after it was received.
1880 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1881 "next input=" as part of the text comprising the host identity instead of
1882 the correct text. The code was using the same buffer for two different
1883 strings. However, depending on which order the printing function evaluated
1884 its arguments, the bug did not always show up. Under Linux, for example, my
1885 test suite worked just fine.
1887 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1888 doesn't work with some older versions of Perl. It has been changed to "my",
1889 which in any case is probably the better facility to use.
1891 15. A really picky compiler found some instances of statements for creating
1892 error messages that either had too many or two few arguments for the format
1895 16. The size of the buffer for calls to the DNS resolver has been increased
1896 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1897 for addresses that have a lot of PTR records. This alleviates a problem; it
1898 does not fully solve it.
1900 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1901 buffer now truncates the list and logs the incident, which is the same
1902 action as happens when Exim is looking up a host name and its aliases.
1903 Previously in this situation something unpredictable would happen;
1904 sometimes it was "internal error: store_reset failed".
1906 18. If a server dropped the connection unexpectedly when an Exim client was
1907 using GnuTLS and trying to read a response, the client delivery process
1908 crashed while trying to generate an error log message.
1910 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1911 string, for example:
1913 warn message = H1: something\nH2: something
1915 the text was added as a single header line from Exim's point of view
1916 though it ended up OK in the delivered message. However, searching for the
1917 second and subsequent header lines using $h_h2: did not work. This has been
1918 fixed. Similarly, if a system filter added multiple headers in this way,
1919 the routers could not see them.
1921 20. Expanded the error message when iplsearch is called with an invalid key to
1922 suggest using net-iplsearch in a host list.
1924 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1925 ACLs are no longer actually imposed (and a message to that effect is
1928 22. If a "gecos" field in a passwd entry contained escaped characters, in
1929 particular, if it contained a \" sequence, Exim got it wrong when building
1930 a From: or a Sender: header from that name. A second bug also caused
1931 incorrect handling when an unquoted " was present following a character
1932 that needed quoting.
1934 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1935 was not being matched caselessly.
1937 24. Arranged for all hyphens in the exim.8 source to be escaped with
1940 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1941 properly. Recipient callout cache records were still being keyed to include
1942 the sender, even when use_sender was set false. This led to far more
1943 callouts that were necessary. The sender is no longer included in the key
1944 when use_sender is false.
1946 26. Added "control = submission" modifier to ACLs.
1948 27. Added the ${base62d: operator to decode base 62 numbers.
1950 28. dnsdb lookups can now access SRV records.
1952 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1953 the configuration file.
1955 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1956 (-v) mode. This makes the output for a verbose queue run more intelligible.
1958 31. Added a use_postmaster feature to recipient callouts.
1960 32. Added the $body_zerocount variable, containing the number of binary zero
1961 bytes in the message body.
1963 33. The time of last modification of the "new" subdirectory is now used as the
1964 "mailbox time last read" when there is a quota error for a maildir
1967 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1969 35. Added +ignore_unknown as a special item in host lists.
1971 36. Code for decoding IPv6 addresses in host lists is now included, even if
1972 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1973 address was recognized as an IP address, but was then not correctly decoded
1974 into binary, causing unexpected and incorrect effects when compared with
1981 1. Very minor rewording of debugging text in manualroute to say "list of
1982 hosts" instead of "hostlist".
1984 2. If verify=header_syntax was set, and a header line with an unqualified
1985 address (no domain) and a large number of spaces between the end of the
1986 name and the colon was received, the reception process suffered a buffer
1987 overflow, and (when I tested it) crashed. This was caused by some obsolete
1988 code that should have been removed. The fix is to remove it!
1990 3. When running in the test harness, delay a bit after writing a bounce
1991 message to get a bit more predictability in the log output.
1993 4. Added a call to search_tidyup() just before forking a reception process. In
1994 theory, someone could use a lookup in the expansion of smtp_accept_max_
1995 per_host which, without the tidyup, could leave open a database connection.
1997 5. Added the variables $recipient_data and $sender_data which get set from a
1998 lookup success in an ACL "recipients" or "senders" condition, or a router
1999 "senders" option, similar to $domain_data and $local_part_data.
2001 6. Moved the writing of debug_print from before to after the "senders" test
2004 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2005 problems for message scanning, either using a data ACL, or using
2006 local_scan() because the Received: header was not generated till after they
2007 were called (in order to set the time as the time of reception completion).
2008 I have revised the way this works. The header is now generated after the
2009 body is received, but before the ACL or local_scan() are called. After they
2010 are run, the timestamp in the header is updated.
2016 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2017 before starting a queue runner without re-exec. This happened only when
2018 deliver_drop_privilege was set or when the Exim user was set to root. The
2019 effect of the bug was that timeouts during subsequent deliveries caused
2020 crashes instead of being properly handled. The handler is now left at its
2021 default (and expected) setting.
2023 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2024 message, again when deliver_drop_privilege is set or Exim is run as root.
2025 The bug described in (1) was not present in this case, but the tidying up
2026 of the other signals was missing. I have made the two cases consistent.
2028 3. The ignore_target_hosts setting on a manualroute router was being ignored
2029 for hosts that were looked up using the /MX notation.
2031 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2034 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2035 operated on the sender address. After changing the $sender_address to <>
2036 for the sender address verify, Exim was re-instated it as the original
2037 (before rewriting) address, but remembering that it had rewritten it, so it
2038 wasn't rewriting it again. This bug also had the effect of breaking the
2039 sender address verification caching when the sender address was rewritten.
2041 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2042 This has been changed so that if the ip literal address matches
2043 ignore_target_hosts, the router declines.
2045 7. Added expansion conditions match_domain, match_address, and match_local_
2046 part (NOT match_host).
2048 8. The placeholder for the Received: header didn't have a length field set.
2050 9. Added code to Exim itself and to exim_lock to test for a specific race
2051 condition that could lead to file corruption when using MBX delivery. The
2052 issue is with the lockfile that is created in /tmp. If this file is removed
2053 after a process has opened it but before that process has acquired a lock,
2054 there is the potential for a second process to recreate the file and also
2055 acquire a lock. This could lead to two Exim processes writing to the file
2056 at the same time. The added code performs the same test as UW imapd; it
2057 checks after acquiring the lock that its file descriptor still refers to
2058 the same named file.
2060 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2061 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2064 11. Added the smtp_active_hostname option. If used, this will typically be made
2065 to depend on the incoming interface address. Because $interface_address is
2066 not set up until the daemon has forked a reception process, error responses
2067 that can happen earlier (such as "too many connections") no longer contain
2070 12. If an expansion in a condition on a "warn" statement fails because a lookup
2071 defers, the "warn" statement is abandoned, and the next ACL statement is
2072 processed. Previously this caused the whole ACL to be aborted.
2074 13. Added the iplsearch lookup type.
2076 14. Added ident_timeout as a log selector.
2078 15. Added tls_certificate_verified as a log selector.
2080 16. Added a global option tls_require_ciphers (compare the smtp transport
2081 option of the same name). This controls incoming TLS connections.
2083 17. I finally figured out how to make tls_require_ciphers do a similar thing
2084 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2085 before starting the TLS session.
2087 18. Tabs are now shown as \t in -bP output.
2089 19. If the log selector return_path_on_delivery was set, Exim crashed when
2090 bouncing a message because it had too many Received: header lines.
2092 20. If two routers both had headers_remove settings, and the first one included
2093 a superfluous trailing colon, the final name in the first list and the
2094 first name in the second list were incorrectly joined into one item (with a
2095 colon in the middle).
2101 1. Added -C and -D options to the exinext utility, mainly to make it easier
2102 to include in the automated testing, but these could be helpful when
2103 multiple configurations are in use.
2105 2. The exinext utility was not formatting the output nicely when there was
2106 an alternate port involved in the retry record key, nor when there was a
2107 message id as well (for retries that were specific to a specific message
2108 and a specific host). It was also confused by IPv6 addresses, because of
2109 the additional colons they contain. I have fixed the IPv4 problem, and
2110 patched it up to do a reasonable job for IPv6.
2112 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2113 delivery, the log line now contains "pipelined" if PIPELINING was used.
2115 4. An SMTP transport process used to panic and die if the bind() call to set
2116 an explicit outgoing interface failed. This has been changed; it is now
2117 treated in the same way as a connect() failure.
2119 5. A reference to $sender_host_name in the part of a conditional expansion
2120 that was being skipped was still causing a DNS lookup. This no longer
2123 6. The def: expansion condition was not recognizing references to header lines
2124 that used bh_ and bheader_.
2126 7. Added the _cache feature to named lists.
2128 8. The code for checking quota_filecount in the appendfile transport was
2129 allowing one more file than it should have been.
2131 9. For compatibility with Sendmail, the command line option
2139 and sets the incoming protocol and host name (for trusted callers). The
2140 host name and its colon can be omitted when only the protocol is to be set.
2141 Note the Exim already has two private options, -pd and -ps, that refer to
2142 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2143 "s", but I don't think that's a major issue.
2145 10. A number of refactoring changes to the code, none of which should affect
2148 (a) The number of logging options was getting close to filling up the
2149 32-bit word that was used as a bit map. I have split them into two classes:
2150 those that are passed in the argument to log_write(), and those that are
2151 only ever tested independently outside of that function. These are now in
2152 separate 32-bit words, so there is plenty of room for expansion again.
2153 There is no change in the user interface or the logging behaviour.
2155 (b) When building, for example, log lines, the code previously used a
2156 macro that called string_cat() twice, in order to add two strings. This is
2157 not really sufficiently general. Furthermore, there was one instance where
2158 it was actually wrong because one of the argument was used twice, and in
2159 one call a function was used. (As it happened, calling the function twice
2160 did not affect the overall behaviour.) The macro has been replaced by a
2161 function that can join an arbitrary number of extra strings onto a growing
2164 (c) The code for expansion conditions now uses a table and a binary chop
2165 instead of a serial search (which was left over from when there were very
2166 few conditions). Also, it now recognizes conditions like "pam" even when
2167 the relevant support is not compiled in: a suitably worded error message is
2168 given if an attempt is made to use such a condition.
2170 11. Added ${time_interval:xxxxx}.
2172 12. A bug was causing one of the ddress fields not to be passed back correctly
2173 from remote delivery subprocesses. The field in question was not being
2174 subsequently used, so this caused to problems in practice.
2176 13. Added new log selectors queue_time and deliver_time.
2178 14. Might have fixed a bug in maildirsizefile handling that threw up
2179 "unexpected character" debug warnings, and recalculated the data
2180 unnecessarily. In any case, I expanded the warning message to give more
2183 15. Added the message "Restricted characters in address" to the statements in
2184 the default ACL that block characters like @ and % in local parts.
2186 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2187 Three changes have been made:
2189 (a) There was a serious bug; a negative response to MAIL caused the whole
2190 recipient domain to be cached as invalid, thereby blocking all messages
2191 to all local parts at the same domain, from all senders. This bug has
2192 been fixed. The domain is no longer cached after a negative response to
2193 MAIL if the sender used is not empty.
2195 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2198 (c) A new callout option, "use_sender" has been added for people who want
2199 the modified behaviour.
2205 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2208 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2209 indeed breaks things for older releases.
2211 3. Added additional logging to the case where there is a problem reading data
2212 from a filter that is running in a subprocess using a pipe, in order to
2213 try to track down a specific problem.
2215 4. Testing facility fudge: when running in the test harness and attempting
2216 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2217 getting "No route to host". Convert this to a timeout.
2219 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2222 6. Some OS don't have socklen_t but use size_t instead. This affects the
2223 fifth argument of getsockopt() amongst other things. This is now
2224 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2225 can be set for individual OS. I have set it for SunOS5, OSF1, and
2226 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2227 some earlier ones do not.
2229 7. Change 4.30/15 was not doing the test caselessly.
2231 8. The standard form for an IPv6 address literal was being rejected by address
2232 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2233 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2234 this, as well as the form without the "IPv6" on the front (but only when
2235 address literals are enabled, of course).
2237 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2239 10. Exim crashed if a message with an empty sender address specified by -f
2240 encountered a router with an errors_to setting. This could be provoked only
2241 by a command such as
2245 where an empty string was supplied; "<>" did not hit this bug.
2247 11. Installed PCRE release 4.5.
2249 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2250 remained set. It is now erased.
2252 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2253 times from message ids (which are base 36 rather than the normal 62).
2255 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2256 were being counted as actual protocol errors, and logged if the log
2257 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2258 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2259 and DATA following a set of rejected RCPTs do not count as protocol errors.
2260 In other words, Exim assumes they were pipelined, though this may not
2261 actually be the case. Of course, in all cases the client gets an
2262 appropriate error code.
2264 15. If a lookup fails in an ACL condition, a message about the failure may
2265 be available; it is used if testing the ACL cannot continue, because most
2266 such messages specify what the cause of the deferral is. However, some
2267 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2268 that caused an old message to be retained and used if a later statement
2269 caused a defer, replacing the real cause of the deferral.
2271 16. If an IP address had so many PTR records that the DNS lookup buffer
2272 was not large enough to hold them, Exim could crash while trying to process
2273 the truncated data. It now detects and logs this case.
2275 17. Further to 4.21/58, another change has been made: if (and only if) the
2276 first line of a message (the first header line) ends with CRLF, a bare LF
2277 in a subsequent header line has a space inserted after it, so as not to
2278 terminate the header.
2280 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2281 unnecessarily, used atoi() instead of strtol(), and didn't check the
2282 termination when getting file sizes from file names by regex.
2284 19. Completely re-implemented the support for maildirsize files, in the light
2285 of a number of problems with the previous contributed implementation
2286 (4.30/29). In particular:
2288 . If the quota is zero, the maildirsize file is maintained, but no quota is
2291 . If the maildir directory does not exist, it is created before any attempt
2292 to write a maildirsize file.
2294 . The quota value in the file is just a cache; if the quota is changed in
2295 the transport, the new value overrides.
2297 . A regular expression is available for excluding directories from the
2300 20. The autoreply transport checks the characters in options that define the
2301 message's headers; it allows continued headers, but it was checking with
2302 isspace() after an embedded newline instead of explicitly looking for a
2305 21. If all the "regular" hosts to which an address was routed had passed their
2306 expiry times, and had not reached their retry times, the address was
2307 bounced, even if fallback hosts were defined. Now Exim should go on to try
2310 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2311 equivalent code in the SMTP transport. Some hosts send humungous responses
2312 to HELO/EHLO, more than 1024 it seems.
2314 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2315 gives compiler warnings in some environments. I've now done it "properly",
2318 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2319 (because of problems with the built-in one) was declared to return uschar *
2320 instead of char *, causing compiler failure.
2322 25. Fixed a file descriptor leak when processing alias/forward files.
2324 26. Fixed a minor format string issue in dbfn.c.
2326 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2328 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2329 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2332 29. When a sender address is verified, it is cached, to save repeating the test
2333 when there is more than one recipient in a message. However, when the
2334 verification involves a callout, it is possible for different callout
2335 options to be set for different recipients. It is too complicated to keep
2336 track of this in the cache, so now Exim always runs a verification when a
2337 callout is required, relying on the callout cache for the optimization.
2338 The overhead is duplication of the address routing, but this should not be
2341 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2342 to be verified with callout=postmaster, and the main callout worked but the
2343 postmaster check failed, the verification correctly failed. However, if a
2344 subsequent RCPT command asked for sender verification *without* the
2345 postmaster check, incorrect caching caused this verification also to fail,
2348 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2349 it was not caching the DNS options (qualify_single, search_parents) that
2350 were used when the lookup failed. A subsequent lookup with different
2351 options therefore always gave the same answer, though there were cases
2352 where it should not have. (Example: a "domains = !$mx_any" option on a
2353 dnslookup router: the "domains" option is always processed without any
2354 widening, but the router might have qualify_single set.) Now Exim uses the
2355 cached value only when the same options are set.
2357 32. Added John Jetmore's "exipick" utility to the distribution.
2359 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2360 than a timeout (e.g. a certificate is required, and is not provided), an
2361 Exim server now closes the connection immediately. Previously it waited for
2362 the client to close - but if the client is SSL, it seems that they each
2363 wait for each other, leading to a delay before one of them times out.
2365 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2366 maintained 0.8.x compatibility because I don't think many are using it, and
2367 it is clearly obsolete.
2369 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2372 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2373 client certificate was expired. A simple patch fixes this, though I don't
2374 understand the full logic of why the verify callback is called multiple
2377 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2378 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2379 which causes problems with some clients (such as the Certicom SSL Plus
2380 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2381 disables the coutermeasure allowing Eudora to connect."
2383 38. Exim was not checking that a write() to a log file succeeded. This could
2384 lead to Bad Things if a log got too big, in particular if it hit a file
2385 size limit. Exim now panics and dies if it cannot write to a log file, just
2386 as it does if it cannot open a log file.
2388 39. Modified OS/Makefile-Linux so that it now contains
2390 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2392 The two -D definitions ensure that Exim is compiled with large file
2393 support, which makes it possible to handle log files that are bigger than
2396 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2397 instance) a domain was checked against a named list that involved a lookup,
2398 causing $domain_data to be set, then another domain was checked against the
2399 same list, then the first domain was re-checked, the value of $domain_data
2400 after the final check could be wrong. In particular, if the second check
2401 failed, it could be set empty. This bug probably also applied to
2404 41. The strip_trailing_dot option was not being applied to the address given
2405 with the -f command-line option.
2407 42. The code for reading a message's header from the spool was incrementing
2408 $received_count, but never initializing it. This meant that the value was
2409 incorrect (doubled) while delivering a message in the same process in which
2410 it was received. In the most common configuration of Exim, this never
2411 happens - a fresh exec is done - but it can happen when
2412 deliver_drop_privilege is set.
2414 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2415 it now includes up to 150 characters of the unexpected data in the log
2418 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2419 and building data strings. The size of both of these buffers was 10 000
2420 bytes - far larger than anybody would *ever* want, thought I. Needless to
2421 say, somebody hit the limit. I have increased the maximum line length to
2422 20 000 and the maximum data length of concatenated lines to 100 000. I have
2423 also fixed two bugs, because there was no checking on these buffers. Tsk,
2424 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2425 buffer is too small.
2427 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2428 lsearch lookups. Now it does.
2430 46. When parsing a route_list item in a manualroute router, a fixed-length
2431 buffer was used for the list of hosts. I made this 1024 bytes long,
2432 thinking that nobody would ever have a list of hosts that long. Wrong.
2433 Somebody had a whole pile of complicated expansion conditions, and the
2434 string was silently truncated, leading to an expansion error. It turns out
2435 that it is easier to change to an unlimited length (owing to other changes
2436 that have happened since this code was originally written) than to build
2437 structure for giving a limitation error. The length of the item that
2438 expands into the list of hosts is now unlimited.
2440 47. The lsearch lookup could not handle data where the length of text line was
2441 more than 4095 characters. Such lines were truncated, leading to shortened
2442 data being returned. It should now handle lines of any length.
2444 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2445 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2448 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2449 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2450 no longer happens in comment lines. A list of replacements is now placed
2451 at the head of all of the source files, except those whose only change is
2452 to replace PERL_COMMAND in the very first #! line.
2454 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2455 messages on the queue, with a bottom-up merge sort, using code contributed
2456 by Michael Haardt. This should make operations like -bp somewhat faster on
2457 large queues. It won't affect queue runners, except when queue_run_in_order
2460 51. Installed eximstats 1.31 in the distribution.
2462 52. Added support for SRV lookups to the dnslookup router.
2464 53. If an ACL referred to $message_body or $message_body_end, the value was not
2465 reset for any messages that followed in the same SMTP session.
2467 54. The store-handling optimization for building very long strings was not
2468 differentiating between the different store pools. I don't think this
2469 actually made any difference in practice, but I've tidied it.
2471 55. While running the routers to verify a sender address, $sender_address
2472 was still set to the sender address. This is wrong, because when routing to
2473 send a bounce to the sender, it would be empty. Therefore, I have changed
2474 it so that, while verifying a sender address, $sender_address is set to <>.
2475 (There is no change to what happens when verifying a recipient address.)
2477 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2478 target A or AAAA records (if not already returned) without resetting the
2479 qualify_single or search_parents options of the DNS resolver. These are
2480 inappropriate in this case because the targets of MX and SRV records must
2481 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2482 target that, when qualified, matched something in the local domain. These
2483 two options are now turned off when doing these lookups.
2485 57. It seems that at least some releases of Reiserfs (which does not have the
2486 concept of a fixed number of inodes) returns zero and not -1 for the
2487 number of available inodes. This interacted badly with check_spool_inodes,
2488 which assumed that -1 was the "no such thing" setting. What I have done is
2489 to check that the total number of inodes is greater than zero before doing
2490 the test of how many are available.
2492 58. When a "warn" ACL statement has a log_message modifier, the message is
2493 remembered, and not repeated. This is to avoid a lot of repetition when a
2494 message has many recipients that cause the same warning to be written.
2495 Howewer, Exim was preserving the list of already written lines for an
2496 entire SMTP session, which doesn't seem right. The memory is now reset if a
2497 new message is started.
2499 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2500 debugging output unless log_rewrite was also set.
2502 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2503 of (int)(handle) when we know that handle contains (void *)(-1).
2505 61. The Exim daemon panic-logs an error return when it closes the incoming
2506 connection. However "connection reset by peer" seems to be common, and
2507 isn't really an error worthy of noting specially, so that particular error
2510 62. When Exim is trying to find all the local interfaces, it used to panic and
2511 die if the ioctl to get the interface flags failed. However, it seems that
2512 on at least one OS (Solaris 9) it is possible to have an interface that is
2513 included in the list of interfaces, but for which you get a failure error
2514 for this call. This happens when the interface is not "plumbed" into a
2515 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2516 failure of the "get flags" call assumes that the interface is down.
2518 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2519 makes life easier for people who are doing arithmetic on fields extracted
2520 from dates, where you often get leading zeros that should not be
2521 interpreted as octal.
2523 64. Added qualify_domain to the redirect router, to override the global
2526 65. If a pathologically long header line contained very many addresses (the
2527 report of this problem mentioned 10 000) and each of them was rewritten,
2528 Exim could use up a very large amount of memory. (It kept on making new
2529 copies of the header line as it rewrote, and never released the old ones.)
2530 At the expense of a bit more processing, the header rewriting function has
2531 been changed so that it no longer eats memory in this way.
2533 66. The generation of the Received: header has been moved from the time that a
2534 message starts to be received, to the time that it finishes. The timestamp
2535 in the Received: header should now be very close to that of the <= log
2536 line. There are two side-effects of this change:
2538 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2539 logged header lines no longer include the local Received: line, because
2540 it has not yet been created. The same applies to a copy of the message
2541 that is returned to a non-SMTP sender when a message is rejected.
2543 (b) When a filter file is tested using -bf, no additional Received: header
2544 is added to the test message. After some thought, I decided that this
2547 This change does not affect the value of $received_for. It is still set
2548 after address rewriting, but before local_scan() is called.
2550 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2552 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2553 gave an unhelpful panic error message, and a defer error. I have managed to
2554 change this behaviour so that it now rejects any supplied certificate,
2555 which seems right, as the list of acceptable certificates is empty.
2557 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2558 gave an unhelpful defer error. I have not managed to make this reject any
2559 supplied certificates, but the error message it gives is "no certificate
2560 supplied", which is not helpful.
2562 70. exigrep's output now also includes lines that are not associated with any
2563 message, but which match the given pattern. Implemented by a patch from
2564 Martin Sluka, which also tidied up the Perl a bit.
2566 71. Recipient callout verification, like sender verification, was using <> in
2567 the MAIL FROM command. This isn't really the right thing, since the actual
2568 sender may affect whether the remote host accepts the recipient or not. I
2569 have changed it to use the actual sender in the callout; this means that
2570 the cache record is now keyed on a recipient/sender pair, not just the
2571 recipient address. There doesn't seem to be a real danger of callout loops,
2572 since a callout by the remote host to check the sender would use <>.
2573 [SEE ABOVE: changed after hitting problems.]
2575 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2576 temporary errors. However, in the case of such a code being given after
2577 the end of a data transmission (i.e. after ".") Exim was failing to write
2578 a retry record for the message. (Yes, there was some broken host that was
2579 actually sending 8xx at this point.)
2581 73. An unknown lookup type in a host list could cause Exim to panic-die when
2582 the list was checked. (An example that provoked this was putting <; in the
2583 middle of a list instead of at the start.) If this happened during a DATA
2584 ACL check, a -D file could be left lying around. This kind of configuration
2585 error no longer causes Exim to die; instead it causes a defer errror. The
2586 incident is still logged to the main and panic logs.
2588 74. Buglet left over from Exim 3 conversion. The message "too many messages
2589 in one connection" was written to the rejectlog but not the mainlog, except
2590 when address rewriting (yes!) was being logged.
2592 75. Added write_rejectlog option.
2594 76. When a system filter was run not as root (that is, when system_filter_user
2595 was set), the values of the $n variables were not being returned to the
2596 main process; thus, they were not subsequently available in the $sn
2599 77. Added +return_path_on_delivery log selector.
2601 78. A connection timeout was being treated differently from recipients deferred
2602 when testing hosts_max_try with a message that was older than the host's
2603 retry timeout. (The host should not be counted, thus allowing all hosts to
2604 be tried at least once before bouncing.) This may have been the cause of an
2605 occasionally reported bug whereby a message would remain on the queue
2606 longer than the retry timeout, but would be bounced if a delivery was
2607 forced. I say "may" because I never totally pinned down the problem;
2608 setting up timeout/retry tests is difficult. See also the next item.
2610 79. The ultimate address timeout was not being applied to errors that involved
2611 a combination of host plus message (for example, a timeout on a MAIL
2612 command). When an address resolved to a number of possible hosts, and they
2613 were not all tried for each delivery (e.g. because of hosts_max_try), a
2614 message could remain on the queue longer than the retry timeout.
2616 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2619 81. Fixed an obscure SMTP outgoing bug which required at least the following
2620 conditions: (a) there was another message waiting for the same server;
2621 (b) the server returned 5xx to all RCPT commands in the first message so
2622 that the message was not completed; (c) the server dropped the connection
2623 or gave a negative response to the RSET that Exim sends to abort the
2624 transaction. The observed case was a dropped connection after DATA that had
2625 been sent in pipelining mode. That is, the server had advertised PIPELINING
2626 but was not implementing it correctly. The effect of the bug was incorrect
2627 behaviour, such as trying another host, and this could lead to a crash.
2633 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2634 and daemon.c were passed as pointers to ints; they should have been
2635 pointers to socklen_t variables (which are typically unsigned ints).
2637 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2640 3. Fixed a really odd bug that affected only the testing scheme; patching a
2641 certain fixed string in the binary changed the value of another string that
2642 happened to be identical to the end of the original first string.
2644 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2645 name", it returns that address as the IP address. On some operating
2646 systems (e.g. Solaris), it also passes back the IP address string as the
2647 "host name". However, on others (e.g. Linux), it passes back an empty
2648 string. Exim wasn't checking for this, and was changing the host name to an
2649 empty string, assuming it had been canonicized.
2651 5. Although rare, it is permitted to have more than one PTR record for a given
2652 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2653 all the names associated with an address, because they do in Solaris.
2654 However, it seems that they do not in Linux for data that comes from the
2655 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2656 I found this out when I moved to a new Linux workstation and tried to run
2657 the Exim test suite.
2659 To get round this problem I have changed the code so that it now does its
2660 own call to the DNS to look up PTR records when searching for a host name.
2661 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2662 addresses that are only in /etc/hosts are still found.
2664 This behaviour is, however, controlled by an option called host_lookup_
2665 order, which defaults to "bydns:byaddr". If people want to use the other
2666 order, or indeed, just use one or the other means of lookup, they can
2667 specify it in this variable.
2669 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2670 some operating systems, this comes back from gethostbyaddr() as an empty
2671 string, and this is what Exim used to test for. However, it seems that in
2672 other systems, "." is yielded. Exim now tests for this case too.
2674 7. The values of check_spool_space and check_log_space are now held internally
2675 as a number of kilobytes instead of an absolute number of bytes. If a
2676 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2677 kilobyte. This means that much larger values can be stored.
2679 8. Exim monitor: an attempt to get the action menu when not actually pointing
2680 at a message produces an empty menu entitled "No message selected". This
2681 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2682 no entries in it ("Shell widget menu has zero width and/or height"). So I
2683 have added a single, blank menu entry in this case.
2685 9. Added ${quote_local_part.
2687 10. MIME decoding is now applied to the contents of Subject: header lines when
2690 11. Now that a reference to $sender_host_address automatically causes a reverse
2691 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2692 host lookup before query-style lookups in lists that might use this
2693 variable. This has therefore been abolished, and the "net-" prefix is no
2694 longer necessary for query-style lookups.
2696 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2697 have been a typo for LFLAGS, so it has been changed.
2699 13. The install script calls Exim with "-C /dev/null" in order to find the
2700 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2701 to be output. Howeve, since Exim outputs its version number before the
2702 error, it didn't break the script. It just looked ugly. I fixed this by
2703 always allowing "-C /dev/null" if the caller is root.
2705 14. Ignore overlarge ACL variable number when reading spool file - insurance
2706 against a later release with more variables having written the file.
2708 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2709 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2710 this, as well as the form without the "IPv6" on the front.
2712 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2713 OS/Makefile-Darwin file.
2715 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2716 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2718 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2719 variable. However, some OS use ints for inodes. I've added cast to long int
2720 to get rid of the compiler warning.
2722 19. I had forgotten to lock out "/../" in configuration file names when
2723 ALT_CONFIG_PREFIX was set.
2725 20. Routers used for verification do not need to specify transports. However,
2726 if such a router generated a host list, and callout was configured, Exim
2727 crashed, because it could not find a port number from the (non-existent)
2728 transport. It now assumes port 25 in this circumstance.
2730 21. Added the -t option to exigrep.
2732 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2733 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2734 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2735 them is set, LOOKUP_LSEARCH is forced.
2737 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2739 24. Added sender and host information to the "rejected by local_scan()" log
2740 line; previously there was no indication of these.
2742 25. Added .include_if_exists.
2744 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2745 Linux. It turns out that not all file systems support this. Apparently some
2746 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2747 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2748 not supported on the file descriptor, is now ignored when Exim is trying to
2749 sync a directory. This applies only to Linux.
2751 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2753 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2754 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2757 29. Added support for maildirsize files from supplied patch (modified a bit).
2759 30. The use of :fail: followed by an empty string could lead Exim to respond to
2760 sender verification failures with (e.g.):
2762 550 Verification failed for <xxx>
2763 550 Sender verify failed
2765 where the first response line was missing the '-' that indicates it is not
2766 the final line of the response.
2768 31. The loop for finding the name of the user that called Exim had a hardwired
2769 limit of 10; it now uses the value of finduser_retries, which is used for
2770 all other user lookups.
2772 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2775 33. Exim was neglecting to zero errno before one call of strtol() when
2776 expanding a string and expecting an integer value. On some systems this
2777 resulted in spurious "integer overflow" errors. Also, it was casting the
2778 result into an int without checking.
2780 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2781 did not work. The code looks as if it has *never* worked, though it appears
2782 to have been documented since at least releast 1.62. I have made it work.
2784 35. The "timeout_DNS" error in retry rules, also documented since at least
2785 1.62, also never worked. As it isn't clear exactly what this means, and
2786 clearly it isn't a major issue, I have abolished the feature by treating it
2787 as "timeout", and writing a warning to the main and panic logs.
2789 36. The display of retry rules for -brt wasn't always showing the error code
2792 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2793 timeout_connect_A, timeout_connect_MX.
2795 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2796 to the empty sender.
2798 39. The daemon was not analysing the content of -oX till after it had closed
2799 stderr and disconnected from the controlling terminal. This meant that any
2800 syntax errors were only noted on the panic log, and the return code from
2801 the command was 0. By re-arranging the code a little, I've made the
2802 decoding happen first, so such errors now appear on stderr, and the return
2803 code is 1. However, the actual setting up of the sockets still happens in
2804 the disconnected process, so errors there are still only recorded on the
2807 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2808 connections (as happens on some IP stacks) was logged at start up time as
2809 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2810 it from "IPv6 and IPv4", which means that two separate sockets are being
2813 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2814 says whether AF_INET or AF_INET6 was passed as an argument.
2816 42. Exiwhat output was messed up when time zones were included in log
2819 43. Exiwhat now gives more information about the daemon's listening ports,
2820 and whether -tls-on-connect was used.
2822 44. The "port" option of the smtp transport is now expanded.
2824 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2825 silently ignored. Now an error message is written to the main and panic
2828 46. There's a new ACL modifier called "logwrite" which writes to a log file
2829 as soon as it is encountered.
2831 47. Added $local_user_uid and $local_user_gid at routing time.
2833 48. Exim crashed when trying to verify a sender address that was being
2836 49. Exim was recognizing only a space character after ".include". It now also
2837 recognizes a tab character.
2839 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2840 extracting the relevant information from the specification. The man page no
2841 longer contains scrambled data for the -d option, and I've added a section
2842 at the front about calling Exim under different names.
2844 51. Added "extra_headers" argument to the "mail" command in filter files.
2846 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2849 53. Installed eximstats 1.29.
2851 54. Added transport_filter_timeout as a generic transport option.
2853 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2854 Cc: header lines. This was required by RFC 822, but it not required by RFC
2857 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2858 incoming messages that did not have them. Now it does so only if the
2859 message originates locally, that is, if there is no associated remote host
2860 address. When Resent- header lines are present, this applies to the Resent-
2861 lines rather than the non-Resent- lines.
2863 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2864 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2866 58. Messages for configuration errors now include the name of the main
2867 configuration file - useful now that there may be more than one file in a
2868 list (.included file names were always shown).
2870 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2871 for those rare installations that do not start the daemon as root or run it
2872 setuid root. I've cut out the call to initgroups() if the daemon is not
2875 60. The Exim user and group can now be bound into the binary as text strings
2876 that are looked up at the start of Exim's processing.
2878 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2880 62. Added $mailstore_basename variable.
2882 63. Installed patch to sieve.c from Michael Haardt.
2884 64. When Exim failed to open the panic log after failing to open the main log,
2885 the original message it was trying to log was written to stderr and debug
2886 output, but if they were not available (the usual case in production), it
2887 was lost. Now it is written to syslog before the two lines that record the
2888 failures to open the logs.
2890 65. Users' Exim filters run in subprocesses under the user's uid. It is
2891 possible for a "deliver" command or an alias in a "personal" command to
2892 provoke an address rewrite. If logging of address rewriting is configured,
2893 this fails because the process is not running as root or exim. There may be
2894 a better way of dealing with this, but for the moment (because 4.30 needs
2895 to be released), I have disabled address rewrite logging when running a
2896 filter in a non-root, non-exim process.
2902 1. The buildconfig auxiliary program wasn't quoting the value set for
2903 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2904 not defined. This bug was masked in 4.22 by the effect that was fixed in
2907 2. Some messages that were rejected after a message id was allocated were
2908 shown as "incomplete" by exigrep. It no longer does this for messages that
2909 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2911 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2912 have allow_domain_literals set, the ID did not get logged in the <= line.
2913 Domain literals are now always recognized in Message-ID: header lines.
2915 4. The first argument for a ${extract expansion item is the key name or field
2916 number. Leading and trailing spaces in this item were not being ignored,
2917 causing some misleading effects.
2919 5. When deliver_drop_privilege was set, single queue runner processes started
2920 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2921 same command in the process it spins off) were not dropping privilege.
2923 6. When the daemon running as "exim" started a queue runner, it always
2924 re-executed Exim in the spun-off process. This is a waste of effort when
2925 deliver_drop_privilege is set. The new process now just calls the
2926 queue-runner function directly.
2932 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2935 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2936 ignored. Though the use of -oP was forcing the writing of a pid file, it
2937 was always written to the default place.
2939 3. If the message "no IP address found for host xxxx" is generated during
2940 incoming verification, it is now followed by identification of the incoming
2941 connection (so you can more easily find what provoked it).
2943 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2945 5. Added some features to "harden" Exim a bit more against certain attacks:
2947 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2948 be put in Local/Makefile. This is like the never_users runtime option,
2949 but it cannot be overridden. The default setting is "root".
2951 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2952 prefix string with which any file named in a -C command line option
2955 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2956 is retained for -C and -D only if the caller of Exim is root. Without
2957 it, the exim user may also use -C and -D and retain privilege.
2959 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2960 command line option is disabled.
2962 6. Macro names set by the -D option must start with an upper case letter, just
2963 like macro names defined in the configuration file.
2965 7. Added "dereference=" facility to LDAP.
2967 8. Two instances of the typo "uknown" in the source files are fixed.
2969 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2970 the Configure-Makefile script screwed up while processing it.
2972 10. Incorporated PCRE 4.4.
2974 11. The SMTP synchronization check was not operating right at the start of an
2975 SMTP session. For example, it could not catch a HELO sent before the client
2976 waited for the greeting. There is now a check for outstanding input at the
2977 point when the greeting is written. Because of the duplex, asynchronous
2978 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2979 way, but not yet received, when the check is performed.
2981 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2982 on TCP/IP sockets, because this apparently causes some broken clients to
2985 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2986 unchanged) from the Cygwin maintainer.
2988 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2989 was supported instead of testing for mbx. Effectively a typo.
2991 15. The spa authenticator server code was not checking that the input it
2992 received was valid base64.
2994 16. The debug output line for the "set" modifier in ACLs was not showing the
2995 name of the variable that was being set.
2997 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2999 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3000 lookup on its own. Something else was needed to trigger the lookup. For
3001 example, a match in host_lookup or the need for a host name in a host list.
3002 Now, if $sender_host_name is referenced and the host name has not yet been
3003 looked up, a lookup is performed. If the lookup fails, the variable remains
3004 empty, and $host_lookup_failed is set to "1".
3006 19. Added "eqi" as a case-independent comparison operator.
3008 20. The saslauthd authentication condition could segfault if neither service
3009 nor realm was specified.
3011 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3012 error message that was logged was misleading, and incoming SMTP
3013 connections were dropped. The message is now more accurate, and temporary
3014 errors are given to SMTP connections.
3016 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3017 (except RSET) with a 421 error, until QUIT is received. However, it was
3018 failing to send a response to QUIT.
3020 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3021 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3022 ACL is now run after the helo_try_verify_hosts code.
3024 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3025 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3026 case-independent, so other case variants are also recognized). Apparently
3027 some systems use these upper case variants.
3029 25. If more than two messages were waiting for the same host, and a transport
3030 filter was specified for the transport, Exim sent two messages over the
3031 same TCP/IP connection, and then failed with "socket operation on non-
3032 socket" when it tried to send the third.
3034 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3036 27. The extern definition of crypt16() in expand.c was not being excluded when
3037 the OS had its own crypt16() function.
3039 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3040 as a preferred synonym for return_size_limit, both as an option and as an
3043 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3045 30. Changed the default configuration ACL to relax the local part checking rule
3046 for addresses that are not in any local domains. For these addresses,
3047 slashes and pipe symbols are allowed within local parts, but the sequence
3048 /../ is explicitly forbidden.
3050 31. SPA server authentication was not clearing the challenge buffer before
3053 32. log_message in a "warn" ACL statement was writing to the reject log as
3054 well as to the main log, which contradicts the documentation and doesn't
3055 seem right (because no rejection is happening). So I have stopped it.
3057 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3058 However, I am unable to do any testing of this.
3060 34. Fixed an infelicity in the appendfile transport. When checking directories
3061 for a mailbox, to see if any needed to be created, it was accidentally
3062 using path names with one or more superfluous leading slashes; tracing
3063 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3065 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3066 contents are added to the log line that is written for every discarded
3067 recipient. (Previously a log_message setting was ignored.)
3069 36. The ${quote: operator now quotes the string if it is empty.
3071 37. The install script runs exim in order to find its version number. If for
3072 some reason other than non-existence or emptiness, which it checks, it
3073 could not run './exim', it was installing it with an empty version number,
3074 i.e. as "exim-". This error state is now caught, and the installation is
3077 38. An argument was missing from the function that creates an error message
3078 when Exim fails to connect to the socket for saslauthd authentication.
3079 This could cause Exim to crash, or give a corrupted message.
3081 39. Added isip, isip4, and isip6 to ${if conditions.
3083 40. The ACL variables $acl_xx are now saved with the message, and can be
3084 accessed later in routers, transports, and filters.
3086 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3087 strings in the file are not string-expanded.
3089 42. If a MAIL command specified a SIZE value that was too large to fit into an
3090 int variable, the check against message_size_limit failed. Such values are
3091 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3092 day this will have to be increased, but I don't think I want to be around
3093 when emails are that large.
3100 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3101 iconv() is not standard in FreeBSD.
3103 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3104 IPv6 enabled. The observed symptom was a segmentation fault on return from
3105 the function os_common_find_running_interfaces() in src/os.c.
3107 3. In the check_special_case() function in daemon.c I had used "errno" as an
3108 argument name, which causes warnings on some systems. This was basically a
3109 typo, since it was named "eno" in the comments!
3111 4. The code that waits for the clock to tick (at a resolution of some fraction
3112 of a second) so as to ensure message-id uniqueness was always waiting for
3113 at least one whole tick, when it could have waited for less. [This is
3114 almost certainly not relevant at current processor speeds, where it is
3115 unlikely to ever wait at all. But we try to future-proof.]
3117 5. The function that sleeps for a time interval that includes fractions of a
3118 second contained a race. It did not block SIGALRM between setting the
3119 timer, and suspending (a couple of lines later). If the interval was short
3120 and the sigsuspend() was delayed until after it had expired, the suspension
3121 never ended. On busy systems this could lead to processes getting stuck for
3124 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3125 process, before it forks any delivery processes. The open lookup caching
3126 mechanism meant that the open file or database connection was passed into
3127 the delivery process. The problem was that delivery processes always tidy
3128 up cached lookup data. This could cause a problem for the next delivery
3129 process started by the queue runner, because the external queue runner
3130 process does not know about the closure. So the next delivery process
3131 still has data in the lookup cache. In the case of a file lookup, there was
3132 no problem because closing a file descriptor in a subprocess doesn't affect
3133 the parent. However, if the lookup was caching a connection to a database,
3134 the connection was closed, and the second delivery process was likely to
3135 see errors such as "PGSQL: query failed: server closed the connection
3136 unexpectedly". The problem has been fixed by closing all cached lookups
3137 in a queue runner before running a delivery process.
3139 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3140 seem to have the "const" qualifier which it has on other OS. I've
3143 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3144 *of the same type* (client or server) with the same public name that an
3145 error should be diagnosed.
3147 9. When Exim looked up a host name for an IP address, but failed to find the
3148 original IP address when looking up the host name (a safety check), it
3149 output the message "<ip address> does not match any IP for NULL", which was
3150 confusing, to say the least. The bug was that the host name should have
3151 appeared instead of "NULL".
3153 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3154 user that is built into the binary, and the -C or -D options is used, root
3155 privilege is dropped before the configuration file is read. In addition,
3156 logging is switched to stderr instead of the normal log files. If the
3157 configuration then re-defines the Exim user, the unprivileged environment
3158 is probably not what is expected, so Exim logs a panic warning message (but
3161 However, if deliver_drop_privilege is set, the unprivileged state may well
3162 be exactly what is intended, so the warning has been cut out in that case,
3163 and Exim is allowed to try to write to its normal log files.
3169 1. smtp_return_error_details was not giving details for temporary sender
3170 or receiver verification errors.
3172 2. Diagnose a configuration error if two authenticators have the same public
3175 3. Exim used not to create the message log file for a message until the first
3176 delivery attempt. This could be confusing when incoming messages were held
3177 for policy or load reasons. The message log file is now created at the time
3178 the message is received, and an initial "Received" line is written to it.
3180 4. The automatically generated man page for command line options had a minor
3181 bug that caused no ill effects; however, a more serious problem was that
3182 the procedure for building the man page automatically didn't always
3183 operate. Consequently, release 4.20 contains an out-of-date version. This
3184 shouldn't happen again.
3186 5. When building Exim with embedded Perl support, the script that builds the
3187 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3188 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3190 6. The freeze_tell option was not being used for messages that were frozen on
3191 arrival, either by an ACL or by local_scan().
3193 7. Added the smtp_incomplete_transaction log selector.
3195 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3196 was accepting AUTH without a new EHLO.
3198 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3199 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3202 10. Logging of TCP/IP connections (when configured) now happens in the main
3203 daemon process instead of the child process, so that the TCP/IP connection
3204 count is more accurate (but it can never be perfect).
3206 11. The use of "drop" in a nested ACL was not being handled correctly in the
3207 outer ACL. Now, if condition failure induced by the nested "drop" causes
3208 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3209 or "require"), the connection is dropped.
3211 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3212 that yield "discard" can now be used with an "accept" or a "discard" verb,
3213 but an error is generated for any others (because I can't see a useful way
3214 to define what should happen).
3216 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3217 are now processed in the same way as lines in the Exim configuration file.
3218 In particular, continuation lines are supported.
3220 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3222 15. Added -ti meaning -t -i.
3224 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3225 domains, and warn by logging if others are found.
3227 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3228 returned from GIFCONF when Exim is trying to find the list of interfaces on
3229 a host. The code in os.c has been modified to copy each ifreq to an aligned
3230 structure in all cases.
3232 Also, in some cases, the returned ifreq's were being copied to a 'struct
3233 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3234 means the last couple of bytes of an IPv6 address could be chopped if the
3235 ifreq contained only a normal sockaddr (14 bytes storage).
3237 18. Named domain lists were not supported in the hosts_treat_as_local option.
3238 An entry such as +xxxx was not recognized, and was treated as a literal
3241 19. Ensure that header lines added by a DATA ACL are included in the reject log
3242 if the ACL subsequently rejects the message.
3244 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3245 MD5 (which is deprecated).
3247 21. When testing a filter file using -bf, Exim was writing a message when it
3248 took the sender from a "From " line in the message, but it was not doing so
3249 when it took $return_path from a Return-Path: header line. It now does.
3251 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3252 with a valid header line field name (a series of printing characters
3253 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3255 23. Changed "disc" in the source to "disk" to conform to the documentation and
3256 the book and for uniformity.
3258 24. Ignore Sendmail's -Ooption=value command line item.
3260 25. When execve() failed while trying to run a command in a pipe transport,
3261 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3262 could be confused with a return value of 69 from the command itself. This
3263 has been changed to 127, the value the shell returns if it is asked to run
3264 a non-existent command. The wording for the related log line suggests a
3265 non-existent command as the problem.
3267 26. If received_header_text expands to an empty string, do not add a Received:
3268 header line to the message. (Well, it adds a token one on the spool, but
3269 marks it "old" so that it doesn't get used or transmitted.)
3271 27. Installed eximstats 1.28 (addition of -nt option).
3273 28. There was no check for failure on the call to getsockname() in the daemon
3274 code. This can fail if there is a shortage of resources on the system, with
3275 ENOMEM, for example. A temporary error is now given on failure.
3277 29. Contrary to the C standard, it seems that in some environments, the
3278 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3279 program. Exim now does this explicitly; it affects the formatting of
3280 timestamps using strftime().
3282 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3283 complaints. I've now initialized all the variables, to avoid this.
3285 32. Header lines added by a system filter were not being "seen" during
3286 transport-time rewrites.
3288 33. The info_callback() function passed to OpenSSL is set up with type void
3289 (*)(SSL *, int, int), as described somewhere. However, when calling the
3290 function (actually a macro) that sets it up, the type void(*)() is
3291 expected. I've put in a cast to prevent warnings from picky compilers.
3293 34. If a DNS black list lookup found a CNAME record, but there were no A
3294 records associated with the domain it pointed at, Exim crashed.
3296 35. If a DNS black list lookup returned more than one A record, Exim ignored
3297 all but the first. It now scans all returned addresses if a particular IP
3298 value is being sought. In this situation, the contents of the
3299 $dnslist_value variable are a list of all the addresses, separated by a
3302 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3303 to accept a host name and all its aliases if the forward lookup for any of
3304 them yielded the IP address of the incoming connection. Now it accepts only
3305 those names whose forward lookup yields the correct IP address. Any other
3306 names are discarded. This closes a loophole whereby a rogue DNS
3307 administrator could create reverse DNS records to break through a
3308 wildcarded host restriction in an ACL.
3310 37. If a user filter or a system filter that ran in a subprocess used any of
3311 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3312 the wrong values were passed to the pipe command ($thisaddress had the
3313 value of $0, $0 had the value of $1, etc). This bug was introduced by
3314 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3316 38. Improved the line breaking for long SMTP error messages from ACLs.
3317 Previously, if there was no break point between 40 and 75 characters, Exim
3318 left the rest of the message alone. Two changes have been made: (a) I've
3319 reduced the minimum length to 35 characters; (b) if it can't find a break
3320 point between 35 and 75 characters, it looks ahead and uses the first one
3321 that it finds. This may give the occasional overlong line, but at least the
3322 remaining text gets split now.
3324 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3325 file descriptors might be low, and that setting 1000 would always raise it.
3326 It turns out that in some environments, the limit is already over 1000 and
3327 that lowering it causes trouble. So now Exim takes care not to decrease it.
3329 40. When delivering a message, the value of $return_path is set to $sender_
3330 address at the start of routing (routers may change the value). By an
3331 oversight, this default was not being set up when an address was tested by
3332 -bt or -bv, which affected the outcome if any router or filter referred to
3335 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3336 log only when "message" or "log_message" are set. However, if one of the
3337 conditions was an address verification, or a call to a nested ACL, the
3338 messages generated by the underlying test were being passed through. This
3339 no longer happens. The underlying message is available in $acl_verify_
3340 message for both "message" and "log_message" expansions, so it can be
3341 passed through if needed.
3343 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3344 new expansion $bh_ to give the encoded byte string without charset
3345 translation. Translation happens only if iconv() is available; HAVE_ICONV
3346 indicates this at build time. HEADERS_CHARSET gives the charset to
3347 translate to; headers_charset can change it in the configuration, and
3348 "headers charset" can change it in an individual filter file.
3350 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3351 that creates RFC 2047 encoded "words" labels them as that charset instead
3352 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3353 expansion operator; (ii) when Exim creates a From: line for a local
3354 message; (iii) when a header line is rewritten to include a "phrase" part.
3356 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3357 buggy, causing it to skip the first lines of messages whose message ID
3358 ended in 'D'. This would not have bitten before Exim release 4.14, because
3359 message IDs were unlikely to end in 'D' before then. The effect was to have
3360 incorrect size information for certain domains.
3362 45. #include "config.h" was missing at the start of the crypt16.c module. This
3363 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3366 46. If there was a timeout during a "random" callout check, Exim treated it as
3367 a failure of the random address, and carried on sending RSET and the real
3368 address. If the delay was just some slowness somewhere, the response to the
3369 original RCPT would be taken as a response to RSET and so on, causing
3370 mayhem of various kinds.
3372 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3373 when I implemented it. It didn't allow for the fact that some option values
3374 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3375 the right test for positive values.
3377 48. Domain names in DNS records are case-independent. Exim always looks them up
3378 in lower case. Some resolvers return domain names in exactly the case they
3379 appear in the zone file, that is, they may contain uppercase letters. Not
3380 all resolvers do this - some return always lower case. Exim was treating a
3381 change of case by a resolver as a change of domain, similar to a widening
3382 of a domain abbreviation. This triggered its re-routing code and so it was
3383 trying to route what was effectively the same domain again. This normally
3384 caused routing to fail (because the router wouldn't handle the domain
3385 twice). Now Exim checks for this case specially, and just changes the
3386 casing of the domain that it ultimately uses when it transmits the message
3389 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3392 50. If a filter generated a file delivery with a non-absolute name (possible if
3393 no home directory exists for the router), the forbid_file option was not
3396 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3397 the existing equality matching.
3399 52. Exim was using ints instead of ino_t variables in some places where it was
3400 dealing with inode numbers.
3402 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3403 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3404 called TMPDIR, and if it finds it is different, it changes its value.
3406 54. The smtp_printf() function is now made available to local_scan() so
3407 additional output lines can be written before returning. There is also an
3408 smtp_fflush() function to enable the detection of a dropped connection.
3409 The variables smtp_input and smtp_batched_input are exported to
3412 55. Changed the default runtime configuration: the message "Unknown user"
3413 has been removed from the ACL, and instead placed on the localuser router,
3414 using the cannot_route_message feature. This means that any verification
3415 failures that generate their own messages won't get overridden. Similarly,
3416 the "Unrouteable address" message that was in the ACL for unverifiable
3417 relay addresses has also been removed.
3419 56. Added hosts_avoid_esmtp to the smtp transport.
3421 57. The exicyclog script was not checking for the esoteric option
3422 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3423 will work only if exicyclog is run under the appropriate euid.
3425 58. Following a discussion on the list, the rules by which Exim recognises line
3426 endings on incoming messages have been changed. The -dropcr and drop_cr
3427 options are now no-ops, retained only for backwards compatibility. The
3428 following line terminators are recognized: LF CRLF CR. However, special
3429 processing applies to CR:
3431 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3432 nor a local message in the state where . is a terminator.
3434 (ii) If a bare CR is encountered in a header line, an extra space is added
3435 after the line terminator so as not to end the header. The reasoning
3436 behind this is that bare CRs in header lines are most likely either
3437 to be mistakes, or people trying to play silly games.
3439 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3440 was being incorrectly given as 18 bytes larger than it should have been.
3441 This is a VOB (very old bug).
3443 60. This may never have affected anything current, but just in case it has:
3444 When the local host is found other than at the start of a list of hosts,
3445 the local host, those with the same MX, and any that follow, are discarded.
3446 When the list in question was part of a longer list of hosts, the following
3447 hosts (not currently being processed) were also being discarded. This no
3448 longer happens. I'm not sure if this situation could ever has previously
3451 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3454 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3455 header. This is something that is recommended in a new Internet Draft, and
3456 is something that is documented as being done by Sendmail. There are two
3457 possible values. For messages generated by the autoreply transport, Exim
3460 Auto-Submitted: auto-replied
3462 whereas for all other generated messages (e.g. bounces) it adds
3464 Auto-Submitted: auto-generated
3466 63. The "personal" condition in filters now includes a test for the
3467 Auto-Submitted: header. If it contains the string "auto-" the message it
3468 not considered personal.
3470 64. Added rcpt_include_affixes as a generic transport option.
3472 65. Added queue_only_override (default true).
3474 66. Added the syslog_duplication option.
3476 67. If what should have been the first header line of a message consisted of
3477 a space followed by a colon, Exim was mis-interpreting it as a header line.
3478 It isn't of course - it is syntactically invalid and should therefore be
3479 treated as the start of the message body. The misbehaviour could have
3480 caused a number of strange effects, including loss of data in subsequent
3481 header lines, and spool format errors.
3483 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3484 client host had authenticated. This control can now be exercised by an ACL
3485 for more flexibility.
3487 69. By default, callouts do not happen when testing with -bh. There is now a
3488 variant, -bhc, which does actually run the callout code, including
3489 consulting and updating the callout cache.
3491 70. Added support for saslauthd authentication, courtesy of Alexander
3494 71. If statvfs() failed on the spool or log directories while checking their
3495 size for availability, Exim confusingly gave the error "space shortage".
3496 Furthermore, in debugging mode it crashed with a floating point exception.
3497 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3498 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3499 serious problem, Exim now writes to the main and panic logs when this
3500 happens, with details of the failure. It then refuses to accept the
3501 incoming message, giving the message "spool directory problem" or "log
3502 directory problem" with a 421 code for SMTP messages.
3504 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3505 0, 1, and 2 exist, because some OS complain for execs without them (see
3506 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3507 descriptors. However, the code omitted to check that the open succeeded,
3508 causing mysterious errors if for some reason the permissions on /dev/null
3509 got screwed. Now Exim writes a message to the main and panic logs, and
3510 bombs out if it can't open /dev/null.
3512 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3513 interact so that it is all more flexible. It is supposed to remain
3514 backwards compatible. Also added extra_local_interfaces.
3516 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3517 to bomb out with an assertion failure - to the client this appears as a
3518 connection drop. This problem occurs in the part of the code that was taken
3519 from the Samba project. Fortunately, the assertion is in a very simple
3520 function, so I have fixed this by reproducing the function inline in the
3521 one place where it is called, and arranging for authentication to fail
3522 instead of killing the process with assert().
3524 75. The SPA client code was not working when the server requested OEM rather
3525 than Unicode encoding.
3527 76. Added code to make require_files with a specific uid setting more usable in
3528 the case where statting the file as root fails - usually a non-root-mounted
3529 NFS file system. When this happens and the failure is EACCES, Exim now
3530 forks a subprocess and does the per-uid checking as the relevant uid.
3532 77. Added process_log_path.
3534 78. If log_file_path was not explicitly set, a setting of check_log_space or
3535 check_log_inodes was ignored.
3537 79. If a space check for the spool or log partitions fails, the incident is now
3538 logged. Of course, in the latter case the data may get lost...
3540 80. Added the %p formatting code to string_format() so that it can be used to
3541 print addresses in debug_print(). Adjusted all the address printing in the
3542 debugging in store.c to use %p rather than %d.
3544 81. There was a concern that a line of code in smtp_in.c could overflow a
3545 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3546 initially expressed, the concern was not well-founded, because trailing
3547 spaces are removed early. However, if the trailing spaces were followed by
3548 a NULL, they did not get removed, so the overflow was possible. Two fixes
3551 (a) I re-wrote the offending code in a cleaner fashion.
3552 (b) If an incoming SMTP command contains a NULL character, it is rejected
3555 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3556 runs initgroups(), so that if the Exim user is in any additional groups,
3557 they will be used during message reception.
3563 The change log for 4.20 and earlier releases has been archived.