1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.265 2005/11/21 10:09:12 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."
155 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
156 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
159 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
160 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
162 PH/03 Typo: missing ".o" in src/pcre/Makefile.
164 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
165 header line, restrict the check to what is listed in RFCs 2369 and 2929.
166 Also, for "Auto-Submitted", treat anything other than "no" as
167 non-personal, in accordance with RFC 3834. (Previously it treated
168 anything starting "auto-" as non-personal.)
170 TF/01 The control=submission/name=... option had a problem with syntax
171 errors if the name included a slash character. The /name= option
172 now slurps the rest of the string, so it can include any characters
173 but it must come last in the list of options (after /sender_retain
176 PH/05 Some modifications to the interface to the fake nameserver for the new
179 PH/06 Add ${if def:sender_address {(envelope-from <$sender_address>)\n\t}}
180 into the default Received: header string.
187 TK/01 Added the "success_on_redirect" address verification option. See
188 NewStuff for rationale and an example.
190 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
192 PH/02 Patch to exigrep to allow it to work on syslog lines.
194 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
195 fread() to skip over the body file's header line, because in Cygwin the
196 header line is locked and is inaccessible.
198 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
199 co-exist for some time) to make it clear that it is the Exim ID that is
200 referenced, not the Message-ID: header line.
202 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
203 string_format() function, because snprintf() does not exist on all
206 PH/06 The use of forbid_filter_existstest now also locks out the use of the
207 ${stat: expansion item.
209 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
210 protocol synchronization error", to keep the pedants happy.
212 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
213 well as for IRIX systems, when gcc is being used. See the host.c source
216 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
218 PH/10 Named domain lists were not working if used in a queue_smtp_domains
221 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
222 transport and to the smtp transport in LMTP mode.
224 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
226 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
227 run a filter in a subprocess. This could lead to confusion in subsequent
228 lookups in the parent process. There should also be a search_tidyup() at
229 the end of the subprocess.
231 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
232 only if the host matched helo_try_verify_hosts, which caused the
233 verification to occur when the EHLO/HELO command was issued. The ACL just
234 tested the remembered result. Now, if a previous verification attempt has
235 not happened, "verify = helo" does it there and then.
237 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
239 TK/03 Fix log output including CR from clamd.
241 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
242 exist provoked a memory error which could cause a segfault.
244 PH/15 Installed PCRE 6.2
246 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
248 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
249 of the problem. Specifically, suggested +O2 rather than +O1 for the
252 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
254 PH/20 If a delivery was routed to a non-standard port by means of an SRV
255 record, the port was not correctly logged when the outgoing_port log
256 selector was set (it logged the transort's default port).
258 PH/21 Added support for host-specific ports to manualroute, queryprogram,
259 fallback_hosts, and "hosts" in the smtp transport.
261 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
262 host errors such as "Connection refused".
264 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
265 authentication with radiusclient 0.4.9:
267 - Error returned from rc_read_config was caught wrongly
268 - Username/password not passed on to radius server due to wrong length.
270 The presumption is that some radiusclient API changes for 4.51/PH/17
271 were not taken care of correctly. The code is still untested by me (my
272 Linux distribution still has 0.3.2 of radiusclient), but it was
273 contributed by a Radius user.
275 PH/24 When doing a callout, the value of $domain wasn't set correctly when
276 expanding the "port" option of the smtp transport.
278 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
279 while reading a MIME header. Thanks to Tom Hughes for a patch.
281 PH/24 Include config.h inside local_scan.h so that configuration settings are
284 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
285 that in an ACL for RCPT (for example), you can examine exactly what was
288 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
289 commands, but it was not correctly comparing the address with the actual
290 client host address. Thus, it would show the EHLO address in Received:
291 header lines when this was not necessary.
293 PH/27 Added the % operator to ${eval:}.
295 PH/28 Exim tries to create and chdir to its spool directory when it starts;
296 it should be ignoring failures (because with -C, for example, it has lost
297 privilege). It wasn't ignoring creation failures other than "already
300 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
303 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
304 because an input line was too long, either on its own, or by virtue of
305 too many continuations, the temporary file was not being removed, and the
306 return code was incorrect.
308 PH/31 Missing "BOOL" in function definition in filtertest.c.
310 PH/32 Applied Sieve patches from the maintainer.
312 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
314 PH/33 Added "verify = not_blind".
316 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
317 Local/Makefile (with some defaults set). These are used in built scripts
318 such as exicyclog, but they have never been used in the exim_install
319 script (though there are many overriding facilities there). I have
320 arranged that the exim_install script now takes note of these two
323 PH/35 Installed configuration files for Dragonfly.
325 PH/36 When a locally submitted message by a trusted user did not contain a
326 From: header, and the sender address was obtained from -f or from an SMTP
327 MAIL command, and the trusted user did not use -F to supply a sender
328 name, $originator_name was incorrectly used when constructing a From:
329 header. Furthermore, $originator_name was used for submission mode
330 messages from external hosts without From: headers in a similar way,
331 which is clearly wrong.
333 PH/37 Added control=suppress_local_fixups.
335 PH/38 When log_selector = +received_sender was set, and the addition of the
336 sender made the log line's construction buffer exactly full, or one byte
337 less than full, an overflow happened when the terminating "\n" was
340 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
341 when the result of a list match is failure because a DNS lookup failed.
343 PH/40 RM_COMMAND is now used in the building process.
345 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
346 the "build-* directories that it finds.
348 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
349 address in a domain literal was a prefix of an interface address.
351 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
352 when verifying a sender address, unless rewrite_headers is false.
354 PH/44 Wrote a long comment about why errors_to addresses are verified as
355 recipients, not senders.
357 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
358 the ratelimit ACL was added.
360 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
362 PH/46 Added extra information about PostgreSQL errors to the error string.
364 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
365 avoiding the need to install special zones in a real server. This is
366 backwards compatible; if it can't find the fake resolver, it drops back.
367 Thus, both old and new test suites can be run.
369 TF/02 Added util/ratelimit.pl
371 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
372 clock is set back in time.
374 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
375 Candler <B.Candler@pobox.com>.
377 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
378 OK for addresses that are the result of redirections.
380 PH/48 A number of further additions for the benefit of the new test suite,
381 including a fake gethostbyname() that interfaces to the fake DNS resolver
384 TF/06 The fix for widen_domains has also been applied to qualify_single and
385 search_parents which are the other dnslookup options that can cause
388 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
391 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
393 TF/07 Exim produced the error message "an SRV record indicated no SMTP
394 service" if it encountered an MX record with an empty target hostname.
395 The message is now "an MX or SRV record indicated no SMTP service".
397 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
398 if the DNS of the sending site is misconfigured. This is quite a
399 common situation. This change restores the behaviour of treating a
400 helo verification defer as a failure.
402 PH/51 If self=fail was set on a router, the bounce message did not include the
403 actual error message.
409 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
411 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
412 command itself ended in error, the underlying message about the transport
413 filter timeout was being overwritten with the pipe command error. Now the
414 underlying error message should be appended to the second error message.
416 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
418 PH/02 Reduce the amount of output that "make" produces by default. Full output
419 can still be requested.
421 PH/03 The warning log line about a condition test deferring for a "warn" verb
422 was being output only once per connection, rather than after each
423 occurrence (because it was using the same function as for successful
424 "warn" verbs). This seems wrong, so I have changed it.
426 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
427 it should not have, which might have caused a crash in the right
428 circumstances, but probably never did.
430 PH/04 Installed a modified version of Tony Finch's patch to make submission
431 mode fix the return path as well as the Sender: header line, and to
432 add a /name= option so that you can make the user's friendly name appear
435 TF/03 Added the control = fakedefer ACL modifier.
437 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
438 Mark Lowes for thorough testing.
440 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
442 TK/03 Merged latest SRS patch from Miles Wilton.
444 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
445 with the definition in sysexits.h (which is #included earlier).
446 Fortunately, Exim does not actually use EX_OK. The code used to try to
447 preserve the sysexits.h value, by assumimg that macro definitions were
448 scanned for macro replacements. I have been disabused of this notion,
449 so now the code just undefines EX_OK before #including unistd.h.
451 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
452 in the smtp transport. When a block could not be written in a single
453 write() function, the timeout was being re-applied to each part-write.
454 This seems wrong - if the receiver was accepting one byte at a time it
455 would take for ever. The timeout is now adjusted when this happens. It
456 doesn't have to be particularly precise.
458 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
459 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
461 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
462 without a domain if the check to <postmaster@domain> fails.
464 SC/01 Eximstats: added -xls and the ability to specify output files
465 (patch written by Frank Heydlauf).
467 SC/02 Eximstats: use FileHandles for outputing results.
469 SC/03 Eximstats: allow any combination of xls, txt, and html output.
471 SC/04 Eximstats: fixed display of large numbers with -nvr option
473 SC/05 Eximstats: fixed merging of reports with empty tables.
475 SC/06 Eximstats: added the -include_original_destination flag
477 SC/07 Eximstats: removed tabs and trailing whitespace.
479 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
481 TK/06 MBOX spool code: Add real "From " MBOX separator line
482 so the .eml file is really in mbox format (even though
483 most programs do not really care). Patch from Alex Miller.
485 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
486 The latter is generated from $received_to and is only set if the
487 message has one envelope recipient. SA can use these headers,
488 obviously out-of-the-box. Patch from Alex Miller.
490 PH/08 The ${def test on a variable was returning false if the variable's
491 value was "0", contrary to what the specification has always said!
492 The result should be true unless the variable is empty.
494 PH/09 The syntax error of a character other than { following "${if
495 def:variable_name" (after optional whitespace) was not being diagnosed.
496 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
497 accidental colon was present, for example, could give incorrect results.
499 PH/10 Tidied the code in a number of places where the st_size field of a stat()
500 result is used (not including appendfile, where other changes are about
503 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
504 This involved changing a lot of size variables from int to off_t. It
505 should work with maildirs and everything.
507 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
508 spamd dying while we are connected to it.
510 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
513 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
514 function for BDB 4.3.
516 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
518 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
519 the sources for PCRE have been re-arranged and more files are now
522 PH/15 The code I had for printing potentially long long variables in PH/11
523 above was not the best (it lost precision). The length of off_t variables
524 is now inspected at build time, and an appropriate printing format (%ld
525 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
526 to be "long long int" or "long int". This is needed for the internal
527 formatting function string_vformat().
529 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
530 the configuration file to be ":syslog", then the script "guesses" where
531 the logs files are, rather than using the compiled in default. In our
532 case the guess is not the same as the compiled default, so the script
533 suddenly stopped working when I started to use syslog. The patch checks
534 to see if log_file_path is "". If so, it attempts to read it from exim
535 with no configuration file to get the compiled in version, before it
536 falls back to the previous guessing code."
538 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
539 implementing BATV in an Exim configuration. See NewStuff for the gory
542 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
543 Makefile that are specific to HP-UX.
545 PH/18 If the "use_postmaster" option was set for a recipient callout together
546 with the "random" option, the postmaster address was used as the MAIL
547 FROM address for the random test, but not for the subsequent recipient
548 test. It is now used for both.
550 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
551 patch removes a few documentation additions to RFC 3028, because the
552 latest draft now contains them. It adds the new en;ascii-case comparator
553 and a new error check for 8bit text in MIME parts. Comparator and
554 require names are now matched exactly. I enabled the subaddress
555 extension, but it is not well tested yet (read: it works for me)."
557 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
558 rework some of the code of TK/09 above to avoid the hardwired use of
559 "%lld" and "long long". Replaced the call to snprintf() with a call to
562 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
563 records point to non-existent hosts", "retry timeout exceeded", and
564 "retry time not reached for any host after a long failure period".
566 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
567 experimental DomainKeys support:
569 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
570 (2) On an error such as an illegally used "control", the wrong name for
571 the control was given.
573 These problems did NOT occur unless DomainKeys support was compiled.
575 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
577 PH/24 Added ${if match_ip condition.
579 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
580 databases so that it will be absolutely obvious if a crash occurs in the
581 DB library. This is a regular occurrence (often caused by mis-matched
584 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
585 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
586 picked up on a user's system that detects such things. There doesn't seem
587 to be a gcc warning option for this - only an attribute that has to be
588 put on the function's prototype. It seems that in Fedora Core 4 they have
589 set this on a number of new functions. No doubt there will be more in due
592 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
593 specify a transport. However, if an address that was verified by such a
594 router was the subject of a callout, Exim crashed because it tried to
595 read the rcpt_include_affixes from the non-existent transport. Now it
596 just assumes that the setting of that option is false. This bug was
597 introduced by 4.51/PH/31.
599 PH/28 Changed -d+all to exclude +memory, because that information is very
600 rarely of interest, but it makes the output a lot bigger. People tend to
601 do -d+all out of habit.
603 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
604 code in os-type was giving problems when libc.so lives in lib64, like on
607 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
608 aren't the modern standard, and it seems that some systems' include files
609 don't always have them. Exim was already checking for some of the newer
610 ones like T_AAAA, and defining it itself. I've added checks for all the
611 record types that Exim uses.
613 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
614 not automatically generating a new one, as it is supposed to. This
615 prevented TLS from working. If the file did exist, but contained invalid
616 data, a new version was generated, as expected. It was only the case of a
617 non-existent file that was broken.
619 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
620 with a change in libdomainkeys > 0.64.
622 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
623 from DNS. If the selector record carries the flag, it now has
624 precedence over the domain-wide flag.
626 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
628 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
629 the use of a port name that isn't defined in /etc/services) occurred, the
630 message was deferred as in a normal delivery, and thus remained on the
631 spool, instead of being failed because of the mua_wrapper setting. This
632 is now fixed, and I tidied up some of the mua_wrapper messages at the
635 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
636 the messages in a hash of arrays rather than using individual hashes.
637 This is a bit cleaner and results in dramatic memory savings, albeit
638 at a slight CPU cost.
640 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
641 as requested by Marc Sherman.
643 SC/10 Eximstats: added histograms for user specified patterns as requested
646 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
648 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
649 fopen() in the content-scanning modules that did not already have it.
655 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
656 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
658 TK/02 Fix ACL "control" statement not being available in MIME ACL.
660 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
662 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
663 to test Sieve filters that use "vacation".
665 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
666 that changes the way the GnuTLS parameters are stored in the cache file.
667 The new format can be generated externally. For backward compatibility,
668 if the data in the cache doesn't make sense, Exim assumes it has read an
669 old-format file, and it generates new data and writes a new file. This
670 means that you can't go back to an older release without removing the
673 PH/03 A redirect router that has both "unseen" and "one_time" set does not
674 work if there are any delivery delays because "one_time" forces the
675 parent to be marked "delivered", so its unseen clone is never tried
676 again. For this reason, Exim now forbids the simultaneous setting of
679 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
680 redirected to themselves ("homonym" addresses). Read the long ChangeLog
681 entry if you want to know the details. The fix, however, neglected to
682 consider the case when local delivery batching is involved. The test for
683 "previously delivered" was not happening when checking to see if an
684 address could be batched with a previous (undelivered) one; under
685 certain circumstances this could lead to multiple deliveries to the same
688 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
689 in its include files, and this causes problems building Exim.
691 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
692 header_syntax) but Exim was just ignoring anything given after a slash.
693 In particular, this caused confusion with an attempt to use "verify =
694 reverse_host_lookup/defer_ok". An error is now given when options are
695 supplied for verify items that do not have them. (Maybe reverse_host_
696 lookup should have a defer_ok option, but that's a different point.)
698 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
699 defined by RFC 821) to 2048, because there were problems with some AUTH
700 commands, and RFC 1869 says the size should be increased for extended
701 SMTP commands that take arguments.
703 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
706 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
707 "unknown" error; now it says that the functionality isn't in the binary.
709 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
710 an address' error message when a string expansion fails (syntax or
711 whatever). Otherwise the password may appear in the log. Following change
712 PH/42 below, there is no longer a chance of it appearing in a bounce
715 PH/11 Installed exipick version 20050225.0 from John Jetmore.
717 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
718 of its addresses was ever tried. (Bugzilla bug #2.)
720 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
721 the result incorrectly in the debug output. (It correctly added a newline
722 to what was transported.)
724 TF/01 Added $received_time.
726 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
727 commented out examples of how to interface to a virus scanner and to
728 SpamAssassin. Also added commented examples of av_scanner and
729 spamd_address settings.
731 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
732 and controls are allowed in which ACLs. There were a couple of minor
733 errors. Some of the entries in the conditions table (which is a table of
734 where they are NOT allowed) were getting very unwieldy; rewrote them as a
735 negation of where the condition IS allowed.
737 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
739 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
740 header file does not have a version number, so I've had to invent a new
741 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
742 API. The code is untested by me (my Linux distribution still has 0.3.2 of
743 radiusclient), but it was contributed by a Radius user.
745 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
746 files or directories, for OpenSSL.
748 PH/19 When an Exim process that is running as root has to create an Exim log
749 file, it does so in a subprocess that runs as exim:exim so as to get the
750 ownership right at creation (otherwise, other Exim processes might see
751 the file with the wrong ownership). There was no test for failure of this
752 fork() call, which would lead to the process getting stuck as it waited
753 for a non-existent subprocess. Forks do occasionally fail when resources
754 run out. I reviewed all the other calls to fork(); they all seem to check
757 PH/20 When checking for unexpected SMTP input at connect time (before writing
758 the banner), Exim was not dealing correctly with a non-positive return
759 from the read() function. If the client had disconnected by this time,
760 the result was a log entry for a synchronization error with an empty
761 string after "input=" when read() returned zero. If read() returned -1
762 (an event I could not check), uninitialized data bytes were printed.
763 There were reports of junk text (parts of files, etc) appearing after
766 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
768 PH/22 Added support for macro redefinition, and (re)definition in between
769 driver and ACL definitions.
771 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
772 forgetting to use the resulting value; it was using the unexpanded value.
774 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
775 hadn't been configured. The fix is from Juergen Kreileder, who
776 understands it better than I do:
778 "Here's what I see happening with three configured cyrus_sasl
779 authenticators configured (plain, login, cram-md5):
781 On startup auth_cyrus_sasl_init() gets called for each of these.
782 This means three calls to sasl_listmech() without a specified mech_list.
783 => SASL tests which mechs of all available mechs actually work
784 => three warnings about OTP not working
785 => the returned list contains: plain, login, cram-md5, digest-md5, ...
787 With the patch, sasl_listmech() also gets called three times. But now
788 SASL's mech_list option is set to the server_mech specified in the the
789 authenticator. Or in other words, the answer from sasl_listmech()
790 gets limited to just the mech you're testing for (which is different
792 => the return list contains just 'plain' or 'login', 'cram-md5' or
793 nothing depending on the value of ob->server_mech.
795 I've just tested the patch: Authentication still works fine,
796 unavailable mechs specified in the exim configuration are still
797 caught, and the auth.log warnings about OTP are gone."
799 PH/25 When debugging is enabled, the contents of the command line are added
800 to the debugging output, even when log_selector=+arguments is not
803 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
804 answer is "GNU", and only if the return is "GNU/something" is the answer
807 PH/27 $acl_verify_message is now set immediately after the failure of a
808 verification in an ACL, and so is available in subsequent modifiers. In
809 particular, the message can be preserved by coding like this:
811 warn !verify = sender
812 set acl_m0 = $acl_verify_message
814 Previously, $acl_verify_message was set only while expanding "message"
815 and "log_message" when a very denied access.
817 PH/28 Modified OS/os.c-Linux with
819 -#ifndef OS_LOAD_AVERAGE
820 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
822 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
823 nomenclature these days.)
825 PH/29 Installed patch from the Sieve maintainer that adds the options
826 sieve_useraddress and sieve_subaddress to the redirect router.
828 PH/30 In these circumstances:
829 . Two addresses routed to the same list of hosts;
830 . First host does not offer TLS;
831 . First host accepts first address;
832 . First host gives temporary error to second address;
833 . Second host offers TLS and a TLS session is established;
834 . Second host accepts second address.
835 Exim incorrectly logged both deliveries with the TLS parameters (cipher
836 and peerdn, if requested) that were in fact used only for the second
839 PH/31 When doing a callout as part of verifying an address, Exim was not paying
840 attention to any local part prefix or suffix that was matched by the
841 router that accepted the address. It now behaves in the same way as it
842 does for delivery: the affixes are removed from the local part unless
843 rcpt_include_affixes is set on the transport.
845 PH/32 Add the sender address, as F=<...>, to the log line when logging a
846 timeout during the DATA phase of an incoming message.
848 PH/33 Sieve envelope tests were broken for match types other than :is. I have
849 applied a patch sanctioned by the Sieve maintainer.
851 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
852 the uid or gid is negative. A case of a negative gid caused this to be
853 noticed. The fix allows for either to be negative.
855 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
856 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
859 PH/36 The change PH/12 above was broken. Fixed it.
861 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
862 the grounds that routing the same address twice would always produce the
863 same answer. This might have been true once, but it is certainly no
864 longer true now. Routing a child address may depend on the previous
865 routing that produced that child. Some complicated redirection strategies
866 went wrong when messages had multiple recipients, and made Exim's
867 behaviour dependent on the order in which the addresses were given.
869 I have moved the duplicate checking until after the routing is complete.
870 Exim scans the addresses that are assigned to local and remote
871 transports, and removes any duplicates. This means that more work will be
872 done, as duplicates will always all be routed, but duplicates are
873 presumably rare, so I don't expect this is of any significance.
875 For deliveries to pipes, files, and autoreplies, the duplicate checking
876 still happens during the routing process, since they are not going to be
879 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
880 It corrects a timeout issue with spamd. This is Ian's comment: "The
881 background is that sometimes spamd either never reads data from a
882 connection it has accepted, or it never writes response data. The exiscan
883 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
884 blindly assumes that writes won't block so it may never time out."
886 PH/39 Allow G after quota size as well as K and M.
888 PH/40 The value set for $authenticated_id in an authenticator may not contain
889 binary zeroes or newlines because the value is written to log lines and
890 to spool files. There was no check on this. Now the value is run through
891 the string_printing() function so that such characters are converted to
892 printable escape sequences.
894 PH/41 $message_linecount is a new variable that contains the total number of
895 lines in the message. Compare $body_linecount, which is the count for the
898 PH/42 Exim no longer gives details of delivery errors for specific addresses in
899 bounce and delay warning messages, except in certain special cases, which
902 (a) An SMTP error message from a remote host;
903 (b) A message specified in a :fail: redirection;
904 (c) A message specified in a "fail" command in a system filter;
905 (d) A message specified in a FAIL return from the queryprogram router;
906 (e) A message specified by the cannot_route_message router option.
908 In these cases only, Exim does include the error details in bounce and
909 warning messages. There are also a few cases where bland messages such
910 as "unrouteable address" or "local delivery error" are given.
912 PH/43 $value is now also set for the "else" part of a ${run expansion.
914 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
915 being worked on, but at least Exim now implements the latest version to
918 PH/45 In a pipe transport, although a timeout while waiting for the pipe
919 process to complete was treated as a delivery failure, a timeout while
920 writing the message to the pipe was logged, but erroneously treated as a
921 successful delivery. Such timeouts include transport filter timeouts. For
922 consistency with the overall process timeout, these timeouts are now
923 treated as errors, giving rise to delivery failures by default. However,
924 there is now a new Boolean option for the pipe transport called
925 timeout_defer, which, if set TRUE, converts the failures into defers for
926 both kinds of timeout. A transport filter timeout is now identified in
929 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
930 systems where "make" and "gmake" are different, calling "gmake" at top
931 level broke things. I've arranged for the value of $(MAKE) to be passed
932 from the Makefile to this script so that it can call the same version of
936 A note about Exim versions 4.44 and 4.50
937 ----------------------------------------
939 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
940 changes of various kinds. As a consequence, a big documentation update was
941 needed. This delayed the release for rather longer than seemed good, especially
942 in the light of a couple of (minor) security issues. Therefore, the changes
943 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
944 release. So 4.44 and 4.50 are in effect two different branches that both start
947 I have left the 4.50 change log unchanged; it contains all the changes since
948 4.43. The change log for 4.44 is below; many of its items are identical to
949 those for 4.50. This seems to be the most sensible way to preserve the
950 historical information.
956 1. Minor wording change to the doc/README.SIEVE file.
958 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
959 computation of the current number of files was incorrect.
961 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
962 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
965 4. Give more explanation in the error message when the command for a transport
966 filter fails to execute.
968 5. There are several places where Exim runs a non-Exim command in a
969 subprocess. The SIGUSR1 signal should be disabled for these processes. This
970 was being done only for the command run by the queryprogram router. It is
971 now done for all such subprocesses. The other cases are: ${run, transport
972 filters, and the commands run by the lmtp and pipe transports.
974 6. Added CONFIGURE_GROUP build-time option.
976 7. Some older OS have a limit of 256 on the maximum number of file
977 descriptors. Exim was using setrlimit() to set 1000 as a large value
978 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
979 systems. I've change it so that if it can't get 1000, it tries for 256.
981 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
982 was an oversight, and furthermore, ever since the addition of extra
983 controls (e.g. 4.43/32), the checks on when to allow different forms of
984 "control" were broken. There should now be diagnostics for all cases when a
985 control that does not make sense is encountered.
987 9. Added the /retain_sender option to "control=submission".
989 10. $recipients is now available in the predata ACL (oversight).
991 11. Tidy the search cache before the fork to do a delivery from a message
992 received from the command line. Otherwise the child will trigger a lookup
993 failure and thereby defer the delivery if it tries to use (for example) a
994 cached ldap connection that the parent has called unbind on.
996 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
997 of $address_data from the recipient verification was clobbered by the
1000 13. The value of address_data from a sender verification is now available in
1001 $sender_address_data in subsequent conditions in the ACL statement.
1003 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1005 15. Added a new option "connect=<time>" to callout options, to set a different
1008 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1009 was its contents. (It was OK if the option was not defined at all.)
1011 17. A "Completed" log line is now written for messages that are removed from
1012 the spool by the -Mrm option.
1014 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1015 information about exactly what failed.
1017 19. Added -dd to debug only the daemon process.
1019 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1020 handles timeouts, both on the server side and network timeouts. Renamed the
1021 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1023 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1024 It is now set to "smtps".
1026 22. $host_address is now set to the target address during the checking of
1027 ignore_target_hosts.
1029 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1030 being passed; this would have caused $sender_host_name to have been used if
1031 matching the list had actually called for a host name (not very likely,
1032 since this list is usually IP addresses). A host name is now passed as
1035 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1036 code that specifies a non-restarting handler (typically sigaction() in
1037 modern systems) in an attempt to fix a rare and obscure crash bug.
1039 25. Narrowed the window for a race in the daemon that could cause it to ignore
1040 SIGCHLD signals. This is not a major problem, because they are used only to
1041 wake it up if nothing else does.
1043 26. A malformed maildirsize file could cause Exim to calculate negative values
1044 for the mailbox size or file count. Odd effects could occur as a result.
1045 The maildirsize information is now recalculated if the size or filecount
1048 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1049 support for a long time. Removed HAVE_SYS_VFS_H.
1051 28. Installed the latest version of exipick from John Jetmore.
1053 29. In an address list, if the pattern was not a regular expression, an empty
1054 subject address (from a bounce message) matched only if the pattern was an
1055 empty string. Non-empty patterns were not even tested. This was the wrong
1056 because it is perfectly reasonable to use an empty address as part of a
1057 database query. An empty address is now tested by patterns that are
1058 lookups. However, all the other forms of pattern expect the subject to
1059 contain a local part and a domain, and therefore, for them, an empty
1060 address still always fails if the pattern is not itself empty.
1062 30. Exim went into a mad DNS loop when attempting to do a callout where the
1063 host was specified on an smtp transport, and looking it up yielded more
1064 than one IP address.
1066 31. Re-factored the code for checking spool and log partition space into a
1067 function that finds that data and another that does the check. The former
1068 is then used to implement four new variables: $spool_space, $log_space,
1069 $spool_inodes, and $log_inodes.
1071 32. The RFC2047 encoding function was originally intended for short strings
1072 such as real names; it was not keeping to the 75-character limit for
1073 encoded words that the RFC imposes. It now respects the limit, and
1074 generates multiple encoded words if necessary. To be on the safe side, I
1075 have increased the buffer size for the ${rfc2047: expansion operator from
1078 33. It is now permitted to omit both strings after an "if" condition; if the
1079 condition is true, the result is "true". As before, when the second string
1080 is omitted, a false condition yields an empty string. This makes it less
1081 cumbersome to write custom ACL and router conditions.
1083 34. Failure to deliver a bounce message always caused it to be frozen, even if
1084 there was an errors_to setting on the router. The errors_to setting is now
1087 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1088 canonical form (fully expanded) before being placed in
1089 $sender_host_address.
1091 36. The table in the code that translates DNS record types into text (T_A to
1092 "A" for instance) was missing entries for NS and CNAME. It is just possible
1093 that this could have caused confusion if both these types were looked up
1094 for the same domain, because the text type is used as part of Exim's
1095 per-process caching. But the chance of anyone hitting this buglet seems
1098 37. The dnsdb lookup has been extended in a number of ways.
1100 (1) There is a new type, "zns", which walks up the domain tree until it
1101 finds some nameserver records. It should be used with care.
1103 (2) There is a new type, "mxh", which is like "mx" except that it returns
1104 just the host names, not the priorities.
1106 (3) It is now possible to give a list of domains (or IP addresses) to be
1107 looked up. The behaviour when one of the lookups defers can be
1108 controlled by a keyword.
1110 (4) It is now possible to specify the separator character for use when
1111 multiple records are returned.
1113 38. The dnslists ACL condition has been extended: it is now possible to supply
1114 a list of IP addresses and/or domains to be looked up in a particular DNS
1117 39. Added log_selector=+queue_time_overall.
1119 40. When running the queue in the test harness, wait just a tad after forking a
1120 delivery process, to get repeatability of debugging output.
1122 41. Include certificate and key file names in error message when GnuTLS fails
1123 to set them up, because the GnuTLS error message doesn't include the name
1124 of the failing file when there is a problem reading it.
1126 42. Allow both -bf and -bF in the same test run.
1128 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1130 44. The "Exiscan patch" is now merged into the mainline Exim source.
1132 45. Sometimes the final signoff response after QUIT could fail to get
1133 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1134 < 0 before doing a fflush(). This bug looks as though it goes back to the
1135 introduction of TLS in release 3.20, but "sometimes" must have been rare
1136 because the tests only now provoked it.
1138 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1139 (this can affect the format of dates).
1141 47. exim_tidydb, when checking for the continued existence of a message for
1142 which it has found a message-specific retry record, was not finding
1143 messages that were in split spool directories. Consequently, it was
1144 deleting retry records that should have stayed in existence.
1146 48. Steve fixed some bugs in eximstats.
1148 49. The SPA authentication driver was not abandoning authentication and moving
1149 on to the next authenticator when an expansion was forced to fail,
1150 contradicting the general specification for all authenticators. Instead it
1151 was generating a temporary error. It now behaves as specified.
1153 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1154 (the order specifies the preference for clients). The order is now AES256,
1155 AES128, 3DES, ARCFOUR128.
1157 51. Small patch to Sieve code - explicitly set From: when generating an
1160 52. Exim crashed if a remote delivery caused a very long error message to be
1161 recorded - for instance if somebody sent an entire SpamAssassin report back
1162 as a large number of 550 error lines. This bug was coincidentally fixed by
1163 increasing the size of one of Exim's internal buffers (big_buffer) that
1164 happened as part of the Exiscan merge. However, to be on the safe side, I
1165 have made the code more robust (and fixed the comments that describe what
1168 53. Now that there can be additional text after "Completed" in log lines (if
1169 the queue_time_overall log selector is set), a one-byte patch to exigrep
1170 was needed to allow it to recognize "Completed" as not the last thing in
1173 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1174 patch that reportedly fixes this has been added. I am not expert enough to
1175 create a test for it. This is what the patch creator wrote:
1177 "I found a little strange behaviour of ldap code when working with
1178 Windows 2003 AD Domain, where users was placed in more than one
1179 Organization Units. When I tried to give exim partial DN, the exit code
1180 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1181 But simultaneously result of request was absolutely normal ldap result,
1182 so I produce this patch..."
1184 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1185 REFERENCE, so I have modified the code to exclude the patch when that macro
1188 55. Some experimental protocols are using DNS PTR records for new purposes. The
1189 keys for these records are domain names, not reversed IP addresses. The
1190 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1191 leaves it alone. Component reversal etc. now happens only for IP addresses.
1194 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1196 57. Double the size of the debug message buffer (to 2048) so that more of very
1197 long debug lines gets shown.
1199 58. The exicyclog utility now does better if the number of log files to keep
1200 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1202 59. Two changes related to the smtp_active_hostname option:
1204 (1) $smtp_active_hostname is now available as a variable.
1205 (2) The default for smtp_banner uses $smtp_active_hostname instead
1206 of $primary_hostname.
1208 60. The host_aton() function is supposed to be passed a string that is known
1209 to be a valid IP address. However, in the case of IPv6 addresses, it was
1210 not checking this. This is a hostage to fortune. Exim now panics and dies
1211 if the condition is not met. A case was found where this could be provoked
1212 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1213 components; fortuitously, this particular loophole had already been fixed
1214 by change 4.50/55 above.
1216 If there are any other similar loopholes, the new check in host_aton()
1217 itself should stop them being exploited. The report I received stated that
1218 data on the command line could provoke the exploit when Exim was running as
1219 exim, but did not say which command line option was involved. All I could
1220 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1221 running as the user.
1224 61. There was a buffer overflow vulnerability in the SPA authentication code
1225 (which came originally from the Samba project). I have added a test to the
1226 spa_base64_to_bits() function which I hope fixes it.
1229 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1230 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1232 63. The daemon start-up calls getloadavg() while still root for those OS that
1233 need the first call to be done as root, but it missed one case: when
1234 deliver_queue_load_max is set with deliver_drop_privilege. This is
1235 necessary for the benefit of the queue runner, because there is no re-exec
1236 when deliver_drop_privilege is set.
1238 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1239 This has been fixed.
1241 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1242 was in use, was not putting the data itself into the right store pool;
1243 consequently, it could be overwritten for a subsequent message in the same
1244 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1247 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1249 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1250 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1251 the function were treating the return as a boolean value, which happened to
1252 work because 0=false and not-0=true, but is not correct code.
1254 68. The host_aton() function was not handling scoped IPv6 addresses (those
1255 with, for example, "%eth0" on the end) correctly.
1257 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1258 negated items (that is, ~something) in unsigned ints. Some compilers
1259 apparently mutter when there is no cast.
1261 70. If an address verification called from an ACL failed, and did not produce a
1262 user-specific message (i.e. there was only a "system" message), nothing was
1263 put in $acl_verify_message. In this situation, it now puts the system
1266 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1267 session; change 4.31/43 added the unwanted input to the log line - except
1268 that it did not do this in the start of session case. It now does.
1270 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1271 This is wrong and can cause the other end to generate a synchronization
1272 error if it is another Exim or anything else that does the synchronization
1273 check. A QUIT command is no longer sent after a timeout.
1275 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1276 during host lookups.
1278 74. The defer_ok option of callout verification was not working if it was used
1279 when verifying addresses in header lines, that is, for this case:
1281 verify = header_sender/callout=defer_ok
1283 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1284 those file descriptors could be used for SMTP connections. If anything
1285 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1286 could be sent to the SMTP client, causing chaos. The daemon now opens
1287 stdin, stdout, and stderr to /dev/null when it puts itself into the
1290 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1291 log by default. The user can override this with suitable Perl magic.
1293 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1294 the log message when discard triggers, was not working for the DATA ACL or
1295 for the non-SMTP ACL.
1297 78. Error message wording change in sieve.c.
1299 79. If smtp_accept_max_per_host was set, the number of connections could be
1300 restricted to fewer than expected, because the daemon was trying to set up
1301 a new connection before checking whether the processes handling previous
1302 connections had finished. The check for completed processes is now done
1303 earlier. On busy systems, this bug wouldn't be noticed because something
1304 else would have woken the daemon, and it would have reaped the completed
1307 80. If a message was submitted locally by a user whose login name contained one
1308 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1309 It caused a spool format error. I have fixed the spool reading code. A
1310 related problem was that the "from" clause in the Received: line became
1311 illegal because of the space(s). It is now covered by ${quote_local_part.
1313 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1316 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1318 83. Patch from Sieve maintainer to fix unterminated string problem in
1319 "vacation" handling.
1321 84. Some minor changes to the Linux configuration files to help with other
1322 OS variants using glibc.
1324 85. One more patch for Sieve to update vacation handling to latest spec.
1327 ----------------------------------------------------
1328 See the note above about the 4.44 and 4.50 releases.
1329 ----------------------------------------------------
1335 1. Change 4.43/35 introduced a bug that caused file counts to be
1336 incorrectly computed when quota_filecount was set in an appendfile
1339 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1340 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1343 3. Give more explanation in the error message when the command for a transport
1344 filter fails to execute.
1346 4. There are several places where Exim runs a non-Exim command in a
1347 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1348 was being done only for the command run by the queryprogram router. It is
1349 now done for all such subprocesses. The other cases are: ${run, transport
1350 filters, and the commands run by the lmtp and pipe transports.
1352 5. Some older OS have a limit of 256 on the maximum number of file
1353 descriptors. Exim was using setrlimit() to set 1000 as a large value
1354 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1355 systems. I've change it so that if it can't get 1000, it tries for 256.
1357 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1358 was an oversight, and furthermore, ever since the addition of extra
1359 controls (e.g. 4.43/32), the checks on when to allow different forms of
1360 "control" were broken. There should now be diagnostics for all cases when a
1361 control that does not make sense is encountered.
1363 7. $recipients is now available in the predata ACL (oversight).
1365 8. Tidy the search cache before the fork to do a delivery from a message
1366 received from the command line. Otherwise the child will trigger a lookup
1367 failure and thereby defer the delivery if it tries to use (for example) a
1368 cached ldap connection that the parent has called unbind on.
1370 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1371 of $address_data from the recipient verification was clobbered by the
1372 sender verification.
1374 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1375 was its contents. (It was OK if the option was not defined at all.)
1377 11. A "Completed" log line is now written for messages that are removed from
1378 the spool by the -Mrm option.
1380 12. $host_address is now set to the target address during the checking of
1381 ignore_target_hosts.
1383 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1384 being passed; this would have caused $sender_host_name to have been used if
1385 matching the list had actually called for a host name (not very likely,
1386 since this list is usually IP addresses). A host name is now passed as
1389 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1390 code that specifies a non-restarting handler (typically sigaction() in
1391 modern systems) in an attempt to fix a rare and obscure crash bug.
1393 15. Narrowed the window for a race in the daemon that could cause it to ignore
1394 SIGCHLD signals. This is not a major problem, because they are used only to
1395 wake it up if nothing else does.
1397 16. A malformed maildirsize file could cause Exim to calculate negative values
1398 for the mailbox size or file count. Odd effects could occur as a result.
1399 The maildirsize information is now recalculated if the size or filecount
1402 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1403 support for a long time. Removed HAVE_SYS_VFS_H.
1405 18. Updated exipick to current release from John Jetmore.
1407 19. Allow an empty sender to be matched against a lookup in an address list.
1408 Previously the only cases considered were a regular expression, or an
1411 20. Exim went into a mad DNS lookup loop when doing a callout where the
1412 host was specified on the transport, if the DNS lookup yielded more than
1415 21. The RFC2047 encoding function was originally intended for short strings
1416 such as real names; it was not keeping to the 75-character limit for
1417 encoded words that the RFC imposes. It now respects the limit, and
1418 generates multiple encoded words if necessary. To be on the safe side, I
1419 have increased the buffer size for the ${rfc2047: expansion operator from
1422 22. Failure to deliver a bounce message always caused it to be frozen, even if
1423 there was an errors_to setting on the router. The errors_to setting is now
1426 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1427 canonical form (fully expanded) before being placed in
1428 $sender_host_address.
1430 24. Updated eximstats to version 1.33
1432 25. Include certificate and key file names in error message when GnuTLS fails
1433 to set them up, because the GnuTLS error message doesn't include the name
1434 of the failing file when there is a problem reading it.
1436 26. Expand error message when OpenSSL has problems setting up cert/key files.
1439 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1440 (this can affect the format of dates).
1442 28. exim_tidydb, when checking for the continued existence of a message for
1443 which it has found a message-specific retry record, was not finding
1444 messages that were in split spool directories. Consequently, it was
1445 deleting retry records that should have stayed in existence.
1447 29. eximstats updated to version 1.35
1448 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1449 1.35 - bugfix such that pie charts by volume are generated correctly
1451 30. The SPA authentication driver was not abandoning authentication and moving
1452 on to the next authenticator when an expansion was forced to fail,
1453 contradicting the general specification for all authenticators. Instead it
1454 was generating a temporary error. It now behaves as specified.
1456 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1457 (the order specifies the preference for clients). The order is now AES256,
1458 AES128, 3DES, ARCFOUR128.
1460 31. Small patch to Sieve code - explicitly set From: when generating an
1463 32. Exim crashed if a remote delivery caused a very long error message to be
1464 recorded - for instance if somebody sent an entire SpamAssassin report back
1465 as a large number of 550 error lines. This bug was coincidentally fixed by
1466 increasing the size of one of Exim's internal buffers (big_buffer) that
1467 happened as part of the Exiscan merge. However, to be on the safe side, I
1468 have made the code more robust (and fixed the comments that describe what
1471 33. Some experimental protocols are using DNS PTR records for new purposes. The
1472 keys for these records are domain names, not reversed IP addresses. The
1473 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1474 leaves it alone. Component reversal etc. now happens only for IP addresses.
1477 34. The host_aton() function is supposed to be passed a string that is known
1478 to be a valid IP address. However, in the case of IPv6 addresses, it was
1479 not checking this. This is a hostage to fortune. Exim now panics and dies
1480 if the condition is not met. A case was found where this could be provoked
1481 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1482 components; fortuitously, this particular loophole had already been fixed
1483 by change 4.50/55 or 4.44/33 above.
1485 If there are any other similar loopholes, the new check in host_aton()
1486 itself should stop them being exploited. The report I received stated that
1487 data on the command line could provoke the exploit when Exim was running as
1488 exim, but did not say which command line option was involved. All I could
1489 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1490 running as the user.
1493 35. There was a buffer overflow vulnerability in the SPA authentication code
1494 (which came originally from the Samba project). I have added a test to the
1495 spa_base64_to_bits() function which I hope fixes it.
1498 36. The daemon start-up calls getloadavg() while still root for those OS that
1499 need the first call to be done as root, but it missed one case: when
1500 deliver_queue_load_max is set with deliver_drop_privilege. This is
1501 necessary for the benefit of the queue runner, because there is no re-exec
1502 when deliver_drop_privilege is set.
1504 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1505 was in use, was not putting the data itself into the right store pool;
1506 consequently, it could be overwritten for a subsequent message in the same
1507 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1510 38. Sometimes the final signoff response after QUIT could fail to get
1511 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1512 < 0 before doing a fflush(). This bug looks as though it goes back to the
1513 introduction of TLS in release 3.20, but "sometimes" must have been rare
1514 because the tests only now provoked it.
1520 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1521 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1522 Now it returns an integer. A number of calls were still expecting a T/F
1523 return. Fortuitously, in all cases, the tests worked in OK situations,
1524 which is the norm. However, things would have gone wrong on any write
1525 failures on the smtp file descriptor. This function is used when sending
1526 messages over SMTP and also when doing verify callouts.
1528 2. When Exim is called to do synchronous delivery of a locally submitted
1529 message (the -odf or -odi options), it no longer closes stderr before doing
1532 3. Implemented the mua_wrapper option.
1534 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1536 5. Implemented the functions header_remove(), header_testname(),
1537 header_add_at_position(), and receive_remove_recipient(), and exported them
1540 6. If an ACL "warn" statement specified the addition of headers, Exim already
1541 inserted X-ACL-Warn: at the start if there was no header name. However, it
1542 was not making this test for the second and subsequent header lines if
1543 there were newlines in the string. This meant that an invalid header could
1544 be inserted if Exim was badly configured.
1546 7. Allow an ACL "warn" statement to add header lines at the start or after all
1547 the Received: headers, as well as at the end.
1549 8. Added the rcpt_4xx retry error code.
1551 9. Added postmaster_mailfrom=xxx to callout verification option.
1553 10. Added mailfrom=xxxx to the callout verification option, for verify=
1556 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1557 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1559 12. Inserted some casts to stop certain compilers warning when using pointer
1560 differences as field lengths or precisions in printf-type calls (mostly
1561 affecting debugging statements).
1563 13. Added optional readline() support for -be (dynamically loaded).
1565 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1566 same clock tick as a message's arrival, so that its received time was the
1567 same as the "first fail" time on the retry record, and that message
1568 remained on the queue past the ultimate address timeout, every queue runner
1569 would try a delivery (because it was past the ultimate address timeout) but
1570 after another failure, the ultimate address timeout, which should have then
1571 bounced the address, did not kick in. This was a "< instead of <=" error;
1572 in most cases the first failure would have been in the next clock tick
1573 after the received time, and all would be well.
1575 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1576 being recognized when the domain list was tested by the match_domain
1577 condition in an expansion string.
1579 16. Added the ${str2b64: operator.
1581 17. Exim was always calling setrlimit() to set a large limit for the number of
1582 processes, without checking whether the existing limit was already
1583 adequate. (It did check for the limit on file descriptors.) Furthermore,
1584 errors from getrlimit() and setrlimit() were being ignored. Now they are
1585 logged to the main and panic logs, but Exim does carry on, to try to do its
1586 job under whatever limits there are.
1588 18. Imported PCRE 5.0.
1590 19. Trivial typo in log message " temporarily refused connection" (the leading
1593 20. If the log selector return_path_on_delivery was set and an address was
1594 redirected to /dev/null, the delivery process crashed because it assumed
1595 that a return path would always be set for a "successful" delivery. In this
1596 case, the whole delivery is bypassed as an optimization, and therefore no
1599 21. Internal re-arrangement: the function for sending a challenge and reading
1600 a response while authentication was assuming a zero-terminated challenge
1601 string. It's now changed to take a pointer and a length, to allow for
1602 binary data in such strings.
1604 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1606 23. Exim was not respecting finduser_retries when seeking the login of the
1607 uid under which it was called; it was always trying 10 times. (The default
1608 setting of finduser_retries is zero.) Also, it was sleeping after the final
1609 failure, which is pointless.
1611 24. Implemented tls_on_connect_ports.
1613 25. Implemented acl_smtp_predata.
1615 26. If the domain in control=submission is set empty, Exim assumes that the
1616 authenticated id is a complete email address when it generates From: or
1617 Sender: header lines.
1619 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1620 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1621 chown and chgrp in /bin and hostname in /usr/bin.
1623 28. Exim was keeping the "process log" file open after each use, just as it
1624 does for the main log. This opens the possibility of it remaining open for
1625 long periods when the USR1 signal hits a daemon. Occasional processlog
1626 errors were reported, that could have been caused by this. Anyway, it seems
1627 much more sensible not to leave this file open at all, so that is what now
1630 29. The long-running daemon process does not normally write to the log once it
1631 has entered its main loop, and it closes the log before doing so. This is
1632 so that log files can straightforwardly be renamed and moved. However,
1633 there are a couple of unusual error situations where the daemon does write
1634 log entries, and I had neglected to close the log afterwards.
1636 30. The text of an SMTP error response that was received during a remote
1637 delivery was being truncated at 512 bytes. This is too short for some of
1638 the long messages that one sometimes sees. I've increased the limit to
1641 31. It is now possible to make retry rules that apply only when a message has a
1642 specific sender, in particular, an empty sender.
1644 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1645 it possible to be selective about when SMTP synchronization is enforced.
1647 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1649 32. Implemented hosts_connection_nolog.
1651 33. Added an ACL for QUIT.
1653 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1656 35. Added mailbox_size and mailbox_filecount to appendfile.
1658 36. Added control = no_multiline_responses to ACLs.
1660 37. There was a bug in the logic of the code that waits for the clock to tick
1661 in the case where the clock went backwards by a substantial amount such
1662 that the microsecond fraction of "now" was more than the microsecond
1663 fraction of "then" (but the whole seconds number was less).
1665 38. Added support for the libradius Radius client library this is found on
1666 FreeBSD (previously only the radiusclient library was supported).
1672 1. When certain lookups returned multiple values in the form name=value, the
1673 quoting of the values was not always being done properly. Specifically:
1674 (a) If the value started with a double quote, but contained no whitespace,
1676 (b) If the value contained whitespace other than a space character (i.e.
1677 tabs or newlines or carriage returns) it was not quoted.
1678 This fix has been applied to the mysql and pgsql lookups by writing a
1679 separate quoting function and calling it from the lookup code. The fix
1680 should probably also be applied to nisplus, ibase and oracle lookups, but
1681 since I cannot test any of those, I have not disturbed their existing code.
1683 2. A hit in the callout cache for a specific address caused a log line with no
1684 reason for rejecting RCPT. Now it says "Previous (cached) callout
1685 verification failure".
1687 3. There was an off-by-one bug in the queryprogram router. An over-long
1688 return line was truncated at 256 instead of 255 characters, thereby
1689 overflowing its buffer with the terminating zero. As well as fixing this, I
1690 have increased the buffer size to 1024 (and made a note to document this).
1692 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1693 when Exim is waiting for an SMTP response from a remote server, Exim
1694 restarts its select() call on the socket, thereby resetting its timeout.
1695 This is not a problem when such interrupts are rare. Somebody set up a cron
1696 job to run exiwhat every 2 minutes, which is less than the normal select()
1697 timeout (5 or 10 minutes). This meant that the select() timeout never
1698 kicked in because it was always reset. I have fixed this by comparing the
1699 time when an interrupt arrives with the time at the start of the first call
1700 to select(). If more time than the timeout has elapsed, the interrupt is
1701 treated as a timeout.
1703 5. Some internal re-factoring in preparation for the addition of Sieve
1704 extensions (by MH). In particular, the "personal" test is moved to a
1705 separate function, and given an option for scanning Cc: and Bcc: (which is
1706 not set for Exim filters).
1708 6. When Exim created an email address using the login of the caller as the
1709 local part (e.g. when creating a From: or Sender: header line), it was not
1710 quoting the local part when it contained special characters such as @.
1712 7. Installed new OpenBSD configuration files.
1714 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1715 try to make them clearer.
1717 9. Callout options, other than the timeout value, were being ignored when
1718 verifying sender addresses in header lines. For example, when using
1720 verify = header_sender/callout=no_cache
1722 the cache was (incorrectly) being used.
1724 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1725 only the Cygwin environment.
1727 11. When return_path_on_delivery was set as a log selector, if different remote
1728 addresses in the same message used different return paths and parallel
1729 remote delivery occurred, the wrong values would sometimes be logged.
1730 (Whenever a remote delivery process finished, the return path value from
1731 the most recently started remote delivery process was logged.)
1733 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1734 lines when AUTH and/or TLS are in use. This is the "received protocol"
1735 field. Exim used to use "asmtp" for authenticated SMTP, without any
1736 indication (in the protocol name) for TLS use. Now it follows the RFC and
1737 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1738 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1739 names appear in log lines as well as in Received: header lines.
1741 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1742 extensions, and comparison tests, and to fix some bugs.
1744 14. Changes to the "personal" filter test:
1746 (1) The test was buggy in that it was just doing the equivalent of
1747 "contains" tests on header lines. For example, if a user's address was
1748 anne@some.where, the "personal" test would incorrectly be true for
1750 To: susanne@some.where
1752 This test is now done by extracting each address from the header in turn,
1753 and checking the entire address. Other tests that are part of "personal"
1754 are now done using regular expressions (for example, to check local parts
1755 of addresses in From: header lines).
1757 (2) The list of non-personal local parts in From: addresses has been
1758 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1759 taken from the Sieve specification recommendations.
1761 (3) If the message contains any header line starting with "List-" it is
1762 treated as non-personal.
1764 (4) The test for "circular" in the Subject: header line has been removed
1765 because it now seems ill-conceived.
1767 15. Minor typos in src/EDITME comments corrected.
1769 16. Installed latest exipick from John Jetmore.
1771 17. If headers_add on a router specified a text string that was too long for
1772 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1773 of string_sprintf() is now avoided.
1775 18. $message_body_size was not set (it was always zero) when running the DATA
1776 ACL and the local_scan() function.
1778 19. For the "mail" command in an Exim filter, no default was being set for
1779 the once_repeat time, causing a random time value to be used if "once" was
1780 specified. (If the value happened to be <= 0, no repeat happened.) The
1781 default is now 0s, meaning "never repeat". The "vacation" command was OK
1782 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1783 (I found it when inspecting the code).
1785 20. There is now an overall timeout for performing a callout verification. It
1786 defaults to 4 times the callout timeout, which applies to individual SMTP
1787 commands during the callout. The overall timeout applies when there is more
1788 than one host that can be tried. The timeout is checked before trying the
1789 next host. This prevents very long delays if there are a large number of
1790 hosts and all are timing out (e.g. when the network connections are timing
1791 out). The value of the overall timeout can be changed by specifying an
1792 additional sub-option for "callout", called "maxwait". For example:
1794 verify = sender/callout=5s,maxwait=20s
1796 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1797 to the end before writing, but this should make it even safer).
1799 22. Exim was forgetting that it had advertised PIPELINING for the second and
1800 subsequent messages on an SMTP connection. It was also not resetting its
1801 memory on STARTTLS and an internal HELO.
1803 23. When Exim logs an SMTP synchronization error within a session, it now
1804 records whether PIPELINING has been advertised or not.
1806 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1807 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1808 rather than long int.
1810 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1812 26. Added the never_mail option to autoreply.
1818 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1819 crash if the getsockname() call failed; this can happen if a connection is
1820 closed very soon after it is established. The problem was simply in the
1821 order in which certain operations were done, causing Exim to try to write
1822 to the SMTP stream before it had set up the file descriptor. The bug has
1823 been fixed by making things happen in the correct order.
1829 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1830 before the connection was closed, thus losing the rejection response.
1832 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1833 some early Solaris releases, but causes trouble in current releases where
1834 socklen_t is defined.
1836 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1839 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1842 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1843 that releasing the top of it at the end releases what was used for sub-
1844 expansions (unless the block got too big). However, discard this block if
1845 the first thing is a variable or header, so that we can use its block when
1846 it is dynamic (useful for very large $message_headers, for example).
1848 6. Lookups now cache *every* query, not just the most recent. A new, separate
1849 store pool is used for this. It can be recovered when all lookup caches are
1850 flushed. Lookups now release memory at the end of their result strings.
1851 This has involved some general refactoring of the lookup sources.
1853 7. Some code has been added to the store_xxx() functions to reduce the amount
1854 of flapping under certain conditions.
1856 8. log_incoming_interface used to affect only the <= reception log lines. Now
1857 it causes the local interface and port to be added to several more SMTP log
1858 lines, for example "SMTP connection from", and rejection lines.
1860 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1862 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1864 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1865 could be overwritten at the end of the current message (or the start of a
1866 new message if it was set in a HELO ACL). The value is now preserved for
1867 the duration of the SMTP connection.
1869 12. If a transport had a headers_rewrite setting, and a matching header line
1870 contained an unqualified address, that address was qualified, even if it
1871 did not match any rewriting rules. The underlying bug was that the values
1872 of the flags that permit the existence of unqualified sender and recipient
1873 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1874 non-local messages, and by -bnq for local messages) were not being
1875 preserved with the message after it was received.
1877 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1878 "next input=" as part of the text comprising the host identity instead of
1879 the correct text. The code was using the same buffer for two different
1880 strings. However, depending on which order the printing function evaluated
1881 its arguments, the bug did not always show up. Under Linux, for example, my
1882 test suite worked just fine.
1884 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1885 doesn't work with some older versions of Perl. It has been changed to "my",
1886 which in any case is probably the better facility to use.
1888 15. A really picky compiler found some instances of statements for creating
1889 error messages that either had too many or two few arguments for the format
1892 16. The size of the buffer for calls to the DNS resolver has been increased
1893 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1894 for addresses that have a lot of PTR records. This alleviates a problem; it
1895 does not fully solve it.
1897 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1898 buffer now truncates the list and logs the incident, which is the same
1899 action as happens when Exim is looking up a host name and its aliases.
1900 Previously in this situation something unpredictable would happen;
1901 sometimes it was "internal error: store_reset failed".
1903 18. If a server dropped the connection unexpectedly when an Exim client was
1904 using GnuTLS and trying to read a response, the client delivery process
1905 crashed while trying to generate an error log message.
1907 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1908 string, for example:
1910 warn message = H1: something\nH2: something
1912 the text was added as a single header line from Exim's point of view
1913 though it ended up OK in the delivered message. However, searching for the
1914 second and subsequent header lines using $h_h2: did not work. This has been
1915 fixed. Similarly, if a system filter added multiple headers in this way,
1916 the routers could not see them.
1918 20. Expanded the error message when iplsearch is called with an invalid key to
1919 suggest using net-iplsearch in a host list.
1921 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1922 ACLs are no longer actually imposed (and a message to that effect is
1925 22. If a "gecos" field in a passwd entry contained escaped characters, in
1926 particular, if it contained a \" sequence, Exim got it wrong when building
1927 a From: or a Sender: header from that name. A second bug also caused
1928 incorrect handling when an unquoted " was present following a character
1929 that needed quoting.
1931 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1932 was not being matched caselessly.
1934 24. Arranged for all hyphens in the exim.8 source to be escaped with
1937 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1938 properly. Recipient callout cache records were still being keyed to include
1939 the sender, even when use_sender was set false. This led to far more
1940 callouts that were necessary. The sender is no longer included in the key
1941 when use_sender is false.
1943 26. Added "control = submission" modifier to ACLs.
1945 27. Added the ${base62d: operator to decode base 62 numbers.
1947 28. dnsdb lookups can now access SRV records.
1949 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1950 the configuration file.
1952 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1953 (-v) mode. This makes the output for a verbose queue run more intelligible.
1955 31. Added a use_postmaster feature to recipient callouts.
1957 32. Added the $body_zerocount variable, containing the number of binary zero
1958 bytes in the message body.
1960 33. The time of last modification of the "new" subdirectory is now used as the
1961 "mailbox time last read" when there is a quota error for a maildir
1964 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1966 35. Added +ignore_unknown as a special item in host lists.
1968 36. Code for decoding IPv6 addresses in host lists is now included, even if
1969 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1970 address was recognized as an IP address, but was then not correctly decoded
1971 into binary, causing unexpected and incorrect effects when compared with
1978 1. Very minor rewording of debugging text in manualroute to say "list of
1979 hosts" instead of "hostlist".
1981 2. If verify=header_syntax was set, and a header line with an unqualified
1982 address (no domain) and a large number of spaces between the end of the
1983 name and the colon was received, the reception process suffered a buffer
1984 overflow, and (when I tested it) crashed. This was caused by some obsolete
1985 code that should have been removed. The fix is to remove it!
1987 3. When running in the test harness, delay a bit after writing a bounce
1988 message to get a bit more predictability in the log output.
1990 4. Added a call to search_tidyup() just before forking a reception process. In
1991 theory, someone could use a lookup in the expansion of smtp_accept_max_
1992 per_host which, without the tidyup, could leave open a database connection.
1994 5. Added the variables $recipient_data and $sender_data which get set from a
1995 lookup success in an ACL "recipients" or "senders" condition, or a router
1996 "senders" option, similar to $domain_data and $local_part_data.
1998 6. Moved the writing of debug_print from before to after the "senders" test
2001 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2002 problems for message scanning, either using a data ACL, or using
2003 local_scan() because the Received: header was not generated till after they
2004 were called (in order to set the time as the time of reception completion).
2005 I have revised the way this works. The header is now generated after the
2006 body is received, but before the ACL or local_scan() are called. After they
2007 are run, the timestamp in the header is updated.
2013 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2014 before starting a queue runner without re-exec. This happened only when
2015 deliver_drop_privilege was set or when the Exim user was set to root. The
2016 effect of the bug was that timeouts during subsequent deliveries caused
2017 crashes instead of being properly handled. The handler is now left at its
2018 default (and expected) setting.
2020 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2021 message, again when deliver_drop_privilege is set or Exim is run as root.
2022 The bug described in (1) was not present in this case, but the tidying up
2023 of the other signals was missing. I have made the two cases consistent.
2025 3. The ignore_target_hosts setting on a manualroute router was being ignored
2026 for hosts that were looked up using the /MX notation.
2028 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2031 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2032 operated on the sender address. After changing the $sender_address to <>
2033 for the sender address verify, Exim was re-instated it as the original
2034 (before rewriting) address, but remembering that it had rewritten it, so it
2035 wasn't rewriting it again. This bug also had the effect of breaking the
2036 sender address verification caching when the sender address was rewritten.
2038 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2039 This has been changed so that if the ip literal address matches
2040 ignore_target_hosts, the router declines.
2042 7. Added expansion conditions match_domain, match_address, and match_local_
2043 part (NOT match_host).
2045 8. The placeholder for the Received: header didn't have a length field set.
2047 9. Added code to Exim itself and to exim_lock to test for a specific race
2048 condition that could lead to file corruption when using MBX delivery. The
2049 issue is with the lockfile that is created in /tmp. If this file is removed
2050 after a process has opened it but before that process has acquired a lock,
2051 there is the potential for a second process to recreate the file and also
2052 acquire a lock. This could lead to two Exim processes writing to the file
2053 at the same time. The added code performs the same test as UW imapd; it
2054 checks after acquiring the lock that its file descriptor still refers to
2055 the same named file.
2057 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2058 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2061 11. Added the smtp_active_hostname option. If used, this will typically be made
2062 to depend on the incoming interface address. Because $interface_address is
2063 not set up until the daemon has forked a reception process, error responses
2064 that can happen earlier (such as "too many connections") no longer contain
2067 12. If an expansion in a condition on a "warn" statement fails because a lookup
2068 defers, the "warn" statement is abandoned, and the next ACL statement is
2069 processed. Previously this caused the whole ACL to be aborted.
2071 13. Added the iplsearch lookup type.
2073 14. Added ident_timeout as a log selector.
2075 15. Added tls_certificate_verified as a log selector.
2077 16. Added a global option tls_require_ciphers (compare the smtp transport
2078 option of the same name). This controls incoming TLS connections.
2080 17. I finally figured out how to make tls_require_ciphers do a similar thing
2081 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2082 before starting the TLS session.
2084 18. Tabs are now shown as \t in -bP output.
2086 19. If the log selector return_path_on_delivery was set, Exim crashed when
2087 bouncing a message because it had too many Received: header lines.
2089 20. If two routers both had headers_remove settings, and the first one included
2090 a superfluous trailing colon, the final name in the first list and the
2091 first name in the second list were incorrectly joined into one item (with a
2092 colon in the middle).
2098 1. Added -C and -D options to the exinext utility, mainly to make it easier
2099 to include in the automated testing, but these could be helpful when
2100 multiple configurations are in use.
2102 2. The exinext utility was not formatting the output nicely when there was
2103 an alternate port involved in the retry record key, nor when there was a
2104 message id as well (for retries that were specific to a specific message
2105 and a specific host). It was also confused by IPv6 addresses, because of
2106 the additional colons they contain. I have fixed the IPv4 problem, and
2107 patched it up to do a reasonable job for IPv6.
2109 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2110 delivery, the log line now contains "pipelined" if PIPELINING was used.
2112 4. An SMTP transport process used to panic and die if the bind() call to set
2113 an explicit outgoing interface failed. This has been changed; it is now
2114 treated in the same way as a connect() failure.
2116 5. A reference to $sender_host_name in the part of a conditional expansion
2117 that was being skipped was still causing a DNS lookup. This no longer
2120 6. The def: expansion condition was not recognizing references to header lines
2121 that used bh_ and bheader_.
2123 7. Added the _cache feature to named lists.
2125 8. The code for checking quota_filecount in the appendfile transport was
2126 allowing one more file than it should have been.
2128 9. For compatibility with Sendmail, the command line option
2136 and sets the incoming protocol and host name (for trusted callers). The
2137 host name and its colon can be omitted when only the protocol is to be set.
2138 Note the Exim already has two private options, -pd and -ps, that refer to
2139 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2140 "s", but I don't think that's a major issue.
2142 10. A number of refactoring changes to the code, none of which should affect
2145 (a) The number of logging options was getting close to filling up the
2146 32-bit word that was used as a bit map. I have split them into two classes:
2147 those that are passed in the argument to log_write(), and those that are
2148 only ever tested independently outside of that function. These are now in
2149 separate 32-bit words, so there is plenty of room for expansion again.
2150 There is no change in the user interface or the logging behaviour.
2152 (b) When building, for example, log lines, the code previously used a
2153 macro that called string_cat() twice, in order to add two strings. This is
2154 not really sufficiently general. Furthermore, there was one instance where
2155 it was actually wrong because one of the argument was used twice, and in
2156 one call a function was used. (As it happened, calling the function twice
2157 did not affect the overall behaviour.) The macro has been replaced by a
2158 function that can join an arbitrary number of extra strings onto a growing
2161 (c) The code for expansion conditions now uses a table and a binary chop
2162 instead of a serial search (which was left over from when there were very
2163 few conditions). Also, it now recognizes conditions like "pam" even when
2164 the relevant support is not compiled in: a suitably worded error message is
2165 given if an attempt is made to use such a condition.
2167 11. Added ${time_interval:xxxxx}.
2169 12. A bug was causing one of the ddress fields not to be passed back correctly
2170 from remote delivery subprocesses. The field in question was not being
2171 subsequently used, so this caused to problems in practice.
2173 13. Added new log selectors queue_time and deliver_time.
2175 14. Might have fixed a bug in maildirsizefile handling that threw up
2176 "unexpected character" debug warnings, and recalculated the data
2177 unnecessarily. In any case, I expanded the warning message to give more
2180 15. Added the message "Restricted characters in address" to the statements in
2181 the default ACL that block characters like @ and % in local parts.
2183 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2184 Three changes have been made:
2186 (a) There was a serious bug; a negative response to MAIL caused the whole
2187 recipient domain to be cached as invalid, thereby blocking all messages
2188 to all local parts at the same domain, from all senders. This bug has
2189 been fixed. The domain is no longer cached after a negative response to
2190 MAIL if the sender used is not empty.
2192 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2195 (c) A new callout option, "use_sender" has been added for people who want
2196 the modified behaviour.
2202 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2205 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2206 indeed breaks things for older releases.
2208 3. Added additional logging to the case where there is a problem reading data
2209 from a filter that is running in a subprocess using a pipe, in order to
2210 try to track down a specific problem.
2212 4. Testing facility fudge: when running in the test harness and attempting
2213 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2214 getting "No route to host". Convert this to a timeout.
2216 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2219 6. Some OS don't have socklen_t but use size_t instead. This affects the
2220 fifth argument of getsockopt() amongst other things. This is now
2221 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2222 can be set for individual OS. I have set it for SunOS5, OSF1, and
2223 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2224 some earlier ones do not.
2226 7. Change 4.30/15 was not doing the test caselessly.
2228 8. The standard form for an IPv6 address literal was being rejected by address
2229 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2230 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2231 this, as well as the form without the "IPv6" on the front (but only when
2232 address literals are enabled, of course).
2234 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2236 10. Exim crashed if a message with an empty sender address specified by -f
2237 encountered a router with an errors_to setting. This could be provoked only
2238 by a command such as
2242 where an empty string was supplied; "<>" did not hit this bug.
2244 11. Installed PCRE release 4.5.
2246 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2247 remained set. It is now erased.
2249 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2250 times from message ids (which are base 36 rather than the normal 62).
2252 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2253 were being counted as actual protocol errors, and logged if the log
2254 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2255 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2256 and DATA following a set of rejected RCPTs do not count as protocol errors.
2257 In other words, Exim assumes they were pipelined, though this may not
2258 actually be the case. Of course, in all cases the client gets an
2259 appropriate error code.
2261 15. If a lookup fails in an ACL condition, a message about the failure may
2262 be available; it is used if testing the ACL cannot continue, because most
2263 such messages specify what the cause of the deferral is. However, some
2264 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2265 that caused an old message to be retained and used if a later statement
2266 caused a defer, replacing the real cause of the deferral.
2268 16. If an IP address had so many PTR records that the DNS lookup buffer
2269 was not large enough to hold them, Exim could crash while trying to process
2270 the truncated data. It now detects and logs this case.
2272 17. Further to 4.21/58, another change has been made: if (and only if) the
2273 first line of a message (the first header line) ends with CRLF, a bare LF
2274 in a subsequent header line has a space inserted after it, so as not to
2275 terminate the header.
2277 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2278 unnecessarily, used atoi() instead of strtol(), and didn't check the
2279 termination when getting file sizes from file names by regex.
2281 19. Completely re-implemented the support for maildirsize files, in the light
2282 of a number of problems with the previous contributed implementation
2283 (4.30/29). In particular:
2285 . If the quota is zero, the maildirsize file is maintained, but no quota is
2288 . If the maildir directory does not exist, it is created before any attempt
2289 to write a maildirsize file.
2291 . The quota value in the file is just a cache; if the quota is changed in
2292 the transport, the new value overrides.
2294 . A regular expression is available for excluding directories from the
2297 20. The autoreply transport checks the characters in options that define the
2298 message's headers; it allows continued headers, but it was checking with
2299 isspace() after an embedded newline instead of explicitly looking for a
2302 21. If all the "regular" hosts to which an address was routed had passed their
2303 expiry times, and had not reached their retry times, the address was
2304 bounced, even if fallback hosts were defined. Now Exim should go on to try
2307 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2308 equivalent code in the SMTP transport. Some hosts send humungous responses
2309 to HELO/EHLO, more than 1024 it seems.
2311 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2312 gives compiler warnings in some environments. I've now done it "properly",
2315 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2316 (because of problems with the built-in one) was declared to return uschar *
2317 instead of char *, causing compiler failure.
2319 25. Fixed a file descriptor leak when processing alias/forward files.
2321 26. Fixed a minor format string issue in dbfn.c.
2323 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2325 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2326 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2329 29. When a sender address is verified, it is cached, to save repeating the test
2330 when there is more than one recipient in a message. However, when the
2331 verification involves a callout, it is possible for different callout
2332 options to be set for different recipients. It is too complicated to keep
2333 track of this in the cache, so now Exim always runs a verification when a
2334 callout is required, relying on the callout cache for the optimization.
2335 The overhead is duplication of the address routing, but this should not be
2338 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2339 to be verified with callout=postmaster, and the main callout worked but the
2340 postmaster check failed, the verification correctly failed. However, if a
2341 subsequent RCPT command asked for sender verification *without* the
2342 postmaster check, incorrect caching caused this verification also to fail,
2345 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2346 it was not caching the DNS options (qualify_single, search_parents) that
2347 were used when the lookup failed. A subsequent lookup with different
2348 options therefore always gave the same answer, though there were cases
2349 where it should not have. (Example: a "domains = !$mx_any" option on a
2350 dnslookup router: the "domains" option is always processed without any
2351 widening, but the router might have qualify_single set.) Now Exim uses the
2352 cached value only when the same options are set.
2354 32. Added John Jetmore's "exipick" utility to the distribution.
2356 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2357 than a timeout (e.g. a certificate is required, and is not provided), an
2358 Exim server now closes the connection immediately. Previously it waited for
2359 the client to close - but if the client is SSL, it seems that they each
2360 wait for each other, leading to a delay before one of them times out.
2362 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2363 maintained 0.8.x compatibility because I don't think many are using it, and
2364 it is clearly obsolete.
2366 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2369 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2370 client certificate was expired. A simple patch fixes this, though I don't
2371 understand the full logic of why the verify callback is called multiple
2374 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2375 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2376 which causes problems with some clients (such as the Certicom SSL Plus
2377 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2378 disables the coutermeasure allowing Eudora to connect."
2380 38. Exim was not checking that a write() to a log file succeeded. This could
2381 lead to Bad Things if a log got too big, in particular if it hit a file
2382 size limit. Exim now panics and dies if it cannot write to a log file, just
2383 as it does if it cannot open a log file.
2385 39. Modified OS/Makefile-Linux so that it now contains
2387 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2389 The two -D definitions ensure that Exim is compiled with large file
2390 support, which makes it possible to handle log files that are bigger than
2393 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2394 instance) a domain was checked against a named list that involved a lookup,
2395 causing $domain_data to be set, then another domain was checked against the
2396 same list, then the first domain was re-checked, the value of $domain_data
2397 after the final check could be wrong. In particular, if the second check
2398 failed, it could be set empty. This bug probably also applied to
2401 41. The strip_trailing_dot option was not being applied to the address given
2402 with the -f command-line option.
2404 42. The code for reading a message's header from the spool was incrementing
2405 $received_count, but never initializing it. This meant that the value was
2406 incorrect (doubled) while delivering a message in the same process in which
2407 it was received. In the most common configuration of Exim, this never
2408 happens - a fresh exec is done - but it can happen when
2409 deliver_drop_privilege is set.
2411 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2412 it now includes up to 150 characters of the unexpected data in the log
2415 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2416 and building data strings. The size of both of these buffers was 10 000
2417 bytes - far larger than anybody would *ever* want, thought I. Needless to
2418 say, somebody hit the limit. I have increased the maximum line length to
2419 20 000 and the maximum data length of concatenated lines to 100 000. I have
2420 also fixed two bugs, because there was no checking on these buffers. Tsk,
2421 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2422 buffer is too small.
2424 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2425 lsearch lookups. Now it does.
2427 46. When parsing a route_list item in a manualroute router, a fixed-length
2428 buffer was used for the list of hosts. I made this 1024 bytes long,
2429 thinking that nobody would ever have a list of hosts that long. Wrong.
2430 Somebody had a whole pile of complicated expansion conditions, and the
2431 string was silently truncated, leading to an expansion error. It turns out
2432 that it is easier to change to an unlimited length (owing to other changes
2433 that have happened since this code was originally written) than to build
2434 structure for giving a limitation error. The length of the item that
2435 expands into the list of hosts is now unlimited.
2437 47. The lsearch lookup could not handle data where the length of text line was
2438 more than 4095 characters. Such lines were truncated, leading to shortened
2439 data being returned. It should now handle lines of any length.
2441 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2442 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2445 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2446 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2447 no longer happens in comment lines. A list of replacements is now placed
2448 at the head of all of the source files, except those whose only change is
2449 to replace PERL_COMMAND in the very first #! line.
2451 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2452 messages on the queue, with a bottom-up merge sort, using code contributed
2453 by Michael Haardt. This should make operations like -bp somewhat faster on
2454 large queues. It won't affect queue runners, except when queue_run_in_order
2457 51. Installed eximstats 1.31 in the distribution.
2459 52. Added support for SRV lookups to the dnslookup router.
2461 53. If an ACL referred to $message_body or $message_body_end, the value was not
2462 reset for any messages that followed in the same SMTP session.
2464 54. The store-handling optimization for building very long strings was not
2465 differentiating between the different store pools. I don't think this
2466 actually made any difference in practice, but I've tidied it.
2468 55. While running the routers to verify a sender address, $sender_address
2469 was still set to the sender address. This is wrong, because when routing to
2470 send a bounce to the sender, it would be empty. Therefore, I have changed
2471 it so that, while verifying a sender address, $sender_address is set to <>.
2472 (There is no change to what happens when verifying a recipient address.)
2474 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2475 target A or AAAA records (if not already returned) without resetting the
2476 qualify_single or search_parents options of the DNS resolver. These are
2477 inappropriate in this case because the targets of MX and SRV records must
2478 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2479 target that, when qualified, matched something in the local domain. These
2480 two options are now turned off when doing these lookups.
2482 57. It seems that at least some releases of Reiserfs (which does not have the
2483 concept of a fixed number of inodes) returns zero and not -1 for the
2484 number of available inodes. This interacted badly with check_spool_inodes,
2485 which assumed that -1 was the "no such thing" setting. What I have done is
2486 to check that the total number of inodes is greater than zero before doing
2487 the test of how many are available.
2489 58. When a "warn" ACL statement has a log_message modifier, the message is
2490 remembered, and not repeated. This is to avoid a lot of repetition when a
2491 message has many recipients that cause the same warning to be written.
2492 Howewer, Exim was preserving the list of already written lines for an
2493 entire SMTP session, which doesn't seem right. The memory is now reset if a
2494 new message is started.
2496 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2497 debugging output unless log_rewrite was also set.
2499 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2500 of (int)(handle) when we know that handle contains (void *)(-1).
2502 61. The Exim daemon panic-logs an error return when it closes the incoming
2503 connection. However "connection reset by peer" seems to be common, and
2504 isn't really an error worthy of noting specially, so that particular error
2507 62. When Exim is trying to find all the local interfaces, it used to panic and
2508 die if the ioctl to get the interface flags failed. However, it seems that
2509 on at least one OS (Solaris 9) it is possible to have an interface that is
2510 included in the list of interfaces, but for which you get a failure error
2511 for this call. This happens when the interface is not "plumbed" into a
2512 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2513 failure of the "get flags" call assumes that the interface is down.
2515 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2516 makes life easier for people who are doing arithmetic on fields extracted
2517 from dates, where you often get leading zeros that should not be
2518 interpreted as octal.
2520 64. Added qualify_domain to the redirect router, to override the global
2523 65. If a pathologically long header line contained very many addresses (the
2524 report of this problem mentioned 10 000) and each of them was rewritten,
2525 Exim could use up a very large amount of memory. (It kept on making new
2526 copies of the header line as it rewrote, and never released the old ones.)
2527 At the expense of a bit more processing, the header rewriting function has
2528 been changed so that it no longer eats memory in this way.
2530 66. The generation of the Received: header has been moved from the time that a
2531 message starts to be received, to the time that it finishes. The timestamp
2532 in the Received: header should now be very close to that of the <= log
2533 line. There are two side-effects of this change:
2535 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2536 logged header lines no longer include the local Received: line, because
2537 it has not yet been created. The same applies to a copy of the message
2538 that is returned to a non-SMTP sender when a message is rejected.
2540 (b) When a filter file is tested using -bf, no additional Received: header
2541 is added to the test message. After some thought, I decided that this
2544 This change does not affect the value of $received_for. It is still set
2545 after address rewriting, but before local_scan() is called.
2547 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2549 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2550 gave an unhelpful panic error message, and a defer error. I have managed to
2551 change this behaviour so that it now rejects any supplied certificate,
2552 which seems right, as the list of acceptable certificates is empty.
2554 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2555 gave an unhelpful defer error. I have not managed to make this reject any
2556 supplied certificates, but the error message it gives is "no certificate
2557 supplied", which is not helpful.
2559 70. exigrep's output now also includes lines that are not associated with any
2560 message, but which match the given pattern. Implemented by a patch from
2561 Martin Sluka, which also tidied up the Perl a bit.
2563 71. Recipient callout verification, like sender verification, was using <> in
2564 the MAIL FROM command. This isn't really the right thing, since the actual
2565 sender may affect whether the remote host accepts the recipient or not. I
2566 have changed it to use the actual sender in the callout; this means that
2567 the cache record is now keyed on a recipient/sender pair, not just the
2568 recipient address. There doesn't seem to be a real danger of callout loops,
2569 since a callout by the remote host to check the sender would use <>.
2570 [SEE ABOVE: changed after hitting problems.]
2572 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2573 temporary errors. However, in the case of such a code being given after
2574 the end of a data transmission (i.e. after ".") Exim was failing to write
2575 a retry record for the message. (Yes, there was some broken host that was
2576 actually sending 8xx at this point.)
2578 73. An unknown lookup type in a host list could cause Exim to panic-die when
2579 the list was checked. (An example that provoked this was putting <; in the
2580 middle of a list instead of at the start.) If this happened during a DATA
2581 ACL check, a -D file could be left lying around. This kind of configuration
2582 error no longer causes Exim to die; instead it causes a defer errror. The
2583 incident is still logged to the main and panic logs.
2585 74. Buglet left over from Exim 3 conversion. The message "too many messages
2586 in one connection" was written to the rejectlog but not the mainlog, except
2587 when address rewriting (yes!) was being logged.
2589 75. Added write_rejectlog option.
2591 76. When a system filter was run not as root (that is, when system_filter_user
2592 was set), the values of the $n variables were not being returned to the
2593 main process; thus, they were not subsequently available in the $sn
2596 77. Added +return_path_on_delivery log selector.
2598 78. A connection timeout was being treated differently from recipients deferred
2599 when testing hosts_max_try with a message that was older than the host's
2600 retry timeout. (The host should not be counted, thus allowing all hosts to
2601 be tried at least once before bouncing.) This may have been the cause of an
2602 occasionally reported bug whereby a message would remain on the queue
2603 longer than the retry timeout, but would be bounced if a delivery was
2604 forced. I say "may" because I never totally pinned down the problem;
2605 setting up timeout/retry tests is difficult. See also the next item.
2607 79. The ultimate address timeout was not being applied to errors that involved
2608 a combination of host plus message (for example, a timeout on a MAIL
2609 command). When an address resolved to a number of possible hosts, and they
2610 were not all tried for each delivery (e.g. because of hosts_max_try), a
2611 message could remain on the queue longer than the retry timeout.
2613 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2616 81. Fixed an obscure SMTP outgoing bug which required at least the following
2617 conditions: (a) there was another message waiting for the same server;
2618 (b) the server returned 5xx to all RCPT commands in the first message so
2619 that the message was not completed; (c) the server dropped the connection
2620 or gave a negative response to the RSET that Exim sends to abort the
2621 transaction. The observed case was a dropped connection after DATA that had
2622 been sent in pipelining mode. That is, the server had advertised PIPELINING
2623 but was not implementing it correctly. The effect of the bug was incorrect
2624 behaviour, such as trying another host, and this could lead to a crash.
2630 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2631 and daemon.c were passed as pointers to ints; they should have been
2632 pointers to socklen_t variables (which are typically unsigned ints).
2634 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2637 3. Fixed a really odd bug that affected only the testing scheme; patching a
2638 certain fixed string in the binary changed the value of another string that
2639 happened to be identical to the end of the original first string.
2641 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2642 name", it returns that address as the IP address. On some operating
2643 systems (e.g. Solaris), it also passes back the IP address string as the
2644 "host name". However, on others (e.g. Linux), it passes back an empty
2645 string. Exim wasn't checking for this, and was changing the host name to an
2646 empty string, assuming it had been canonicized.
2648 5. Although rare, it is permitted to have more than one PTR record for a given
2649 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2650 all the names associated with an address, because they do in Solaris.
2651 However, it seems that they do not in Linux for data that comes from the
2652 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2653 I found this out when I moved to a new Linux workstation and tried to run
2654 the Exim test suite.
2656 To get round this problem I have changed the code so that it now does its
2657 own call to the DNS to look up PTR records when searching for a host name.
2658 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2659 addresses that are only in /etc/hosts are still found.
2661 This behaviour is, however, controlled by an option called host_lookup_
2662 order, which defaults to "bydns:byaddr". If people want to use the other
2663 order, or indeed, just use one or the other means of lookup, they can
2664 specify it in this variable.
2666 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2667 some operating systems, this comes back from gethostbyaddr() as an empty
2668 string, and this is what Exim used to test for. However, it seems that in
2669 other systems, "." is yielded. Exim now tests for this case too.
2671 7. The values of check_spool_space and check_log_space are now held internally
2672 as a number of kilobytes instead of an absolute number of bytes. If a
2673 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2674 kilobyte. This means that much larger values can be stored.
2676 8. Exim monitor: an attempt to get the action menu when not actually pointing
2677 at a message produces an empty menu entitled "No message selected". This
2678 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2679 no entries in it ("Shell widget menu has zero width and/or height"). So I
2680 have added a single, blank menu entry in this case.
2682 9. Added ${quote_local_part.
2684 10. MIME decoding is now applied to the contents of Subject: header lines when
2687 11. Now that a reference to $sender_host_address automatically causes a reverse
2688 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2689 host lookup before query-style lookups in lists that might use this
2690 variable. This has therefore been abolished, and the "net-" prefix is no
2691 longer necessary for query-style lookups.
2693 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2694 have been a typo for LFLAGS, so it has been changed.
2696 13. The install script calls Exim with "-C /dev/null" in order to find the
2697 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2698 to be output. Howeve, since Exim outputs its version number before the
2699 error, it didn't break the script. It just looked ugly. I fixed this by
2700 always allowing "-C /dev/null" if the caller is root.
2702 14. Ignore overlarge ACL variable number when reading spool file - insurance
2703 against a later release with more variables having written the file.
2705 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2706 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2707 this, as well as the form without the "IPv6" on the front.
2709 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2710 OS/Makefile-Darwin file.
2712 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2713 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2715 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2716 variable. However, some OS use ints for inodes. I've added cast to long int
2717 to get rid of the compiler warning.
2719 19. I had forgotten to lock out "/../" in configuration file names when
2720 ALT_CONFIG_PREFIX was set.
2722 20. Routers used for verification do not need to specify transports. However,
2723 if such a router generated a host list, and callout was configured, Exim
2724 crashed, because it could not find a port number from the (non-existent)
2725 transport. It now assumes port 25 in this circumstance.
2727 21. Added the -t option to exigrep.
2729 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2730 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2731 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2732 them is set, LOOKUP_LSEARCH is forced.
2734 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2736 24. Added sender and host information to the "rejected by local_scan()" log
2737 line; previously there was no indication of these.
2739 25. Added .include_if_exists.
2741 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2742 Linux. It turns out that not all file systems support this. Apparently some
2743 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2744 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2745 not supported on the file descriptor, is now ignored when Exim is trying to
2746 sync a directory. This applies only to Linux.
2748 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2750 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2751 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2754 29. Added support for maildirsize files from supplied patch (modified a bit).
2756 30. The use of :fail: followed by an empty string could lead Exim to respond to
2757 sender verification failures with (e.g.):
2759 550 Verification failed for <xxx>
2760 550 Sender verify failed
2762 where the first response line was missing the '-' that indicates it is not
2763 the final line of the response.
2765 31. The loop for finding the name of the user that called Exim had a hardwired
2766 limit of 10; it now uses the value of finduser_retries, which is used for
2767 all other user lookups.
2769 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2772 33. Exim was neglecting to zero errno before one call of strtol() when
2773 expanding a string and expecting an integer value. On some systems this
2774 resulted in spurious "integer overflow" errors. Also, it was casting the
2775 result into an int without checking.
2777 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2778 did not work. The code looks as if it has *never* worked, though it appears
2779 to have been documented since at least releast 1.62. I have made it work.
2781 35. The "timeout_DNS" error in retry rules, also documented since at least
2782 1.62, also never worked. As it isn't clear exactly what this means, and
2783 clearly it isn't a major issue, I have abolished the feature by treating it
2784 as "timeout", and writing a warning to the main and panic logs.
2786 36. The display of retry rules for -brt wasn't always showing the error code
2789 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2790 timeout_connect_A, timeout_connect_MX.
2792 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2793 to the empty sender.
2795 39. The daemon was not analysing the content of -oX till after it had closed
2796 stderr and disconnected from the controlling terminal. This meant that any
2797 syntax errors were only noted on the panic log, and the return code from
2798 the command was 0. By re-arranging the code a little, I've made the
2799 decoding happen first, so such errors now appear on stderr, and the return
2800 code is 1. However, the actual setting up of the sockets still happens in
2801 the disconnected process, so errors there are still only recorded on the
2804 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2805 connections (as happens on some IP stacks) was logged at start up time as
2806 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2807 it from "IPv6 and IPv4", which means that two separate sockets are being
2810 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2811 says whether AF_INET or AF_INET6 was passed as an argument.
2813 42. Exiwhat output was messed up when time zones were included in log
2816 43. Exiwhat now gives more information about the daemon's listening ports,
2817 and whether -tls-on-connect was used.
2819 44. The "port" option of the smtp transport is now expanded.
2821 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2822 silently ignored. Now an error message is written to the main and panic
2825 46. There's a new ACL modifier called "logwrite" which writes to a log file
2826 as soon as it is encountered.
2828 47. Added $local_user_uid and $local_user_gid at routing time.
2830 48. Exim crashed when trying to verify a sender address that was being
2833 49. Exim was recognizing only a space character after ".include". It now also
2834 recognizes a tab character.
2836 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2837 extracting the relevant information from the specification. The man page no
2838 longer contains scrambled data for the -d option, and I've added a section
2839 at the front about calling Exim under different names.
2841 51. Added "extra_headers" argument to the "mail" command in filter files.
2843 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2846 53. Installed eximstats 1.29.
2848 54. Added transport_filter_timeout as a generic transport option.
2850 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2851 Cc: header lines. This was required by RFC 822, but it not required by RFC
2854 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2855 incoming messages that did not have them. Now it does so only if the
2856 message originates locally, that is, if there is no associated remote host
2857 address. When Resent- header lines are present, this applies to the Resent-
2858 lines rather than the non-Resent- lines.
2860 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2861 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2863 58. Messages for configuration errors now include the name of the main
2864 configuration file - useful now that there may be more than one file in a
2865 list (.included file names were always shown).
2867 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2868 for those rare installations that do not start the daemon as root or run it
2869 setuid root. I've cut out the call to initgroups() if the daemon is not
2872 60. The Exim user and group can now be bound into the binary as text strings
2873 that are looked up at the start of Exim's processing.
2875 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2877 62. Added $mailstore_basename variable.
2879 63. Installed patch to sieve.c from Michael Haardt.
2881 64. When Exim failed to open the panic log after failing to open the main log,
2882 the original message it was trying to log was written to stderr and debug
2883 output, but if they were not available (the usual case in production), it
2884 was lost. Now it is written to syslog before the two lines that record the
2885 failures to open the logs.
2887 65. Users' Exim filters run in subprocesses under the user's uid. It is
2888 possible for a "deliver" command or an alias in a "personal" command to
2889 provoke an address rewrite. If logging of address rewriting is configured,
2890 this fails because the process is not running as root or exim. There may be
2891 a better way of dealing with this, but for the moment (because 4.30 needs
2892 to be released), I have disabled address rewrite logging when running a
2893 filter in a non-root, non-exim process.
2899 1. The buildconfig auxiliary program wasn't quoting the value set for
2900 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2901 not defined. This bug was masked in 4.22 by the effect that was fixed in
2904 2. Some messages that were rejected after a message id was allocated were
2905 shown as "incomplete" by exigrep. It no longer does this for messages that
2906 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2908 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2909 have allow_domain_literals set, the ID did not get logged in the <= line.
2910 Domain literals are now always recognized in Message-ID: header lines.
2912 4. The first argument for a ${extract expansion item is the key name or field
2913 number. Leading and trailing spaces in this item were not being ignored,
2914 causing some misleading effects.
2916 5. When deliver_drop_privilege was set, single queue runner processes started
2917 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2918 same command in the process it spins off) were not dropping privilege.
2920 6. When the daemon running as "exim" started a queue runner, it always
2921 re-executed Exim in the spun-off process. This is a waste of effort when
2922 deliver_drop_privilege is set. The new process now just calls the
2923 queue-runner function directly.
2929 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2932 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2933 ignored. Though the use of -oP was forcing the writing of a pid file, it
2934 was always written to the default place.
2936 3. If the message "no IP address found for host xxxx" is generated during
2937 incoming verification, it is now followed by identification of the incoming
2938 connection (so you can more easily find what provoked it).
2940 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2942 5. Added some features to "harden" Exim a bit more against certain attacks:
2944 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2945 be put in Local/Makefile. This is like the never_users runtime option,
2946 but it cannot be overridden. The default setting is "root".
2948 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2949 prefix string with which any file named in a -C command line option
2952 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2953 is retained for -C and -D only if the caller of Exim is root. Without
2954 it, the exim user may also use -C and -D and retain privilege.
2956 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2957 command line option is disabled.
2959 6. Macro names set by the -D option must start with an upper case letter, just
2960 like macro names defined in the configuration file.
2962 7. Added "dereference=" facility to LDAP.
2964 8. Two instances of the typo "uknown" in the source files are fixed.
2966 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2967 the Configure-Makefile script screwed up while processing it.
2969 10. Incorporated PCRE 4.4.
2971 11. The SMTP synchronization check was not operating right at the start of an
2972 SMTP session. For example, it could not catch a HELO sent before the client
2973 waited for the greeting. There is now a check for outstanding input at the
2974 point when the greeting is written. Because of the duplex, asynchronous
2975 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2976 way, but not yet received, when the check is performed.
2978 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2979 on TCP/IP sockets, because this apparently causes some broken clients to
2982 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2983 unchanged) from the Cygwin maintainer.
2985 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2986 was supported instead of testing for mbx. Effectively a typo.
2988 15. The spa authenticator server code was not checking that the input it
2989 received was valid base64.
2991 16. The debug output line for the "set" modifier in ACLs was not showing the
2992 name of the variable that was being set.
2994 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2996 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2997 lookup on its own. Something else was needed to trigger the lookup. For
2998 example, a match in host_lookup or the need for a host name in a host list.
2999 Now, if $sender_host_name is referenced and the host name has not yet been
3000 looked up, a lookup is performed. If the lookup fails, the variable remains
3001 empty, and $host_lookup_failed is set to "1".
3003 19. Added "eqi" as a case-independent comparison operator.
3005 20. The saslauthd authentication condition could segfault if neither service
3006 nor realm was specified.
3008 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3009 error message that was logged was misleading, and incoming SMTP
3010 connections were dropped. The message is now more accurate, and temporary
3011 errors are given to SMTP connections.
3013 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3014 (except RSET) with a 421 error, until QUIT is received. However, it was
3015 failing to send a response to QUIT.
3017 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3018 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3019 ACL is now run after the helo_try_verify_hosts code.
3021 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3022 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3023 case-independent, so other case variants are also recognized). Apparently
3024 some systems use these upper case variants.
3026 25. If more than two messages were waiting for the same host, and a transport
3027 filter was specified for the transport, Exim sent two messages over the
3028 same TCP/IP connection, and then failed with "socket operation on non-
3029 socket" when it tried to send the third.
3031 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3033 27. The extern definition of crypt16() in expand.c was not being excluded when
3034 the OS had its own crypt16() function.
3036 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3037 as a preferred synonym for return_size_limit, both as an option and as an
3040 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3042 30. Changed the default configuration ACL to relax the local part checking rule
3043 for addresses that are not in any local domains. For these addresses,
3044 slashes and pipe symbols are allowed within local parts, but the sequence
3045 /../ is explicitly forbidden.
3047 31. SPA server authentication was not clearing the challenge buffer before
3050 32. log_message in a "warn" ACL statement was writing to the reject log as
3051 well as to the main log, which contradicts the documentation and doesn't
3052 seem right (because no rejection is happening). So I have stopped it.
3054 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3055 However, I am unable to do any testing of this.
3057 34. Fixed an infelicity in the appendfile transport. When checking directories
3058 for a mailbox, to see if any needed to be created, it was accidentally
3059 using path names with one or more superfluous leading slashes; tracing
3060 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3062 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3063 contents are added to the log line that is written for every discarded
3064 recipient. (Previously a log_message setting was ignored.)
3066 36. The ${quote: operator now quotes the string if it is empty.
3068 37. The install script runs exim in order to find its version number. If for
3069 some reason other than non-existence or emptiness, which it checks, it
3070 could not run './exim', it was installing it with an empty version number,
3071 i.e. as "exim-". This error state is now caught, and the installation is
3074 38. An argument was missing from the function that creates an error message
3075 when Exim fails to connect to the socket for saslauthd authentication.
3076 This could cause Exim to crash, or give a corrupted message.
3078 39. Added isip, isip4, and isip6 to ${if conditions.
3080 40. The ACL variables $acl_xx are now saved with the message, and can be
3081 accessed later in routers, transports, and filters.
3083 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3084 strings in the file are not string-expanded.
3086 42. If a MAIL command specified a SIZE value that was too large to fit into an
3087 int variable, the check against message_size_limit failed. Such values are
3088 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3089 day this will have to be increased, but I don't think I want to be around
3090 when emails are that large.
3097 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3098 iconv() is not standard in FreeBSD.
3100 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3101 IPv6 enabled. The observed symptom was a segmentation fault on return from
3102 the function os_common_find_running_interfaces() in src/os.c.
3104 3. In the check_special_case() function in daemon.c I had used "errno" as an
3105 argument name, which causes warnings on some systems. This was basically a
3106 typo, since it was named "eno" in the comments!
3108 4. The code that waits for the clock to tick (at a resolution of some fraction
3109 of a second) so as to ensure message-id uniqueness was always waiting for
3110 at least one whole tick, when it could have waited for less. [This is
3111 almost certainly not relevant at current processor speeds, where it is
3112 unlikely to ever wait at all. But we try to future-proof.]
3114 5. The function that sleeps for a time interval that includes fractions of a
3115 second contained a race. It did not block SIGALRM between setting the
3116 timer, and suspending (a couple of lines later). If the interval was short
3117 and the sigsuspend() was delayed until after it had expired, the suspension
3118 never ended. On busy systems this could lead to processes getting stuck for
3121 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3122 process, before it forks any delivery processes. The open lookup caching
3123 mechanism meant that the open file or database connection was passed into
3124 the delivery process. The problem was that delivery processes always tidy
3125 up cached lookup data. This could cause a problem for the next delivery
3126 process started by the queue runner, because the external queue runner
3127 process does not know about the closure. So the next delivery process
3128 still has data in the lookup cache. In the case of a file lookup, there was
3129 no problem because closing a file descriptor in a subprocess doesn't affect
3130 the parent. However, if the lookup was caching a connection to a database,
3131 the connection was closed, and the second delivery process was likely to
3132 see errors such as "PGSQL: query failed: server closed the connection
3133 unexpectedly". The problem has been fixed by closing all cached lookups
3134 in a queue runner before running a delivery process.
3136 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3137 seem to have the "const" qualifier which it has on other OS. I've
3140 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3141 *of the same type* (client or server) with the same public name that an
3142 error should be diagnosed.
3144 9. When Exim looked up a host name for an IP address, but failed to find the
3145 original IP address when looking up the host name (a safety check), it
3146 output the message "<ip address> does not match any IP for NULL", which was
3147 confusing, to say the least. The bug was that the host name should have
3148 appeared instead of "NULL".
3150 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3151 user that is built into the binary, and the -C or -D options is used, root
3152 privilege is dropped before the configuration file is read. In addition,
3153 logging is switched to stderr instead of the normal log files. If the
3154 configuration then re-defines the Exim user, the unprivileged environment
3155 is probably not what is expected, so Exim logs a panic warning message (but
3158 However, if deliver_drop_privilege is set, the unprivileged state may well
3159 be exactly what is intended, so the warning has been cut out in that case,
3160 and Exim is allowed to try to write to its normal log files.
3166 1. smtp_return_error_details was not giving details for temporary sender
3167 or receiver verification errors.
3169 2. Diagnose a configuration error if two authenticators have the same public
3172 3. Exim used not to create the message log file for a message until the first
3173 delivery attempt. This could be confusing when incoming messages were held
3174 for policy or load reasons. The message log file is now created at the time
3175 the message is received, and an initial "Received" line is written to it.
3177 4. The automatically generated man page for command line options had a minor
3178 bug that caused no ill effects; however, a more serious problem was that
3179 the procedure for building the man page automatically didn't always
3180 operate. Consequently, release 4.20 contains an out-of-date version. This
3181 shouldn't happen again.
3183 5. When building Exim with embedded Perl support, the script that builds the
3184 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3185 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3187 6. The freeze_tell option was not being used for messages that were frozen on
3188 arrival, either by an ACL or by local_scan().
3190 7. Added the smtp_incomplete_transaction log selector.
3192 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3193 was accepting AUTH without a new EHLO.
3195 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3196 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3199 10. Logging of TCP/IP connections (when configured) now happens in the main
3200 daemon process instead of the child process, so that the TCP/IP connection
3201 count is more accurate (but it can never be perfect).
3203 11. The use of "drop" in a nested ACL was not being handled correctly in the
3204 outer ACL. Now, if condition failure induced by the nested "drop" causes
3205 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3206 or "require"), the connection is dropped.
3208 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3209 that yield "discard" can now be used with an "accept" or a "discard" verb,
3210 but an error is generated for any others (because I can't see a useful way
3211 to define what should happen).
3213 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3214 are now processed in the same way as lines in the Exim configuration file.
3215 In particular, continuation lines are supported.
3217 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3219 15. Added -ti meaning -t -i.
3221 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3222 domains, and warn by logging if others are found.
3224 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3225 returned from GIFCONF when Exim is trying to find the list of interfaces on
3226 a host. The code in os.c has been modified to copy each ifreq to an aligned
3227 structure in all cases.
3229 Also, in some cases, the returned ifreq's were being copied to a 'struct
3230 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3231 means the last couple of bytes of an IPv6 address could be chopped if the
3232 ifreq contained only a normal sockaddr (14 bytes storage).
3234 18. Named domain lists were not supported in the hosts_treat_as_local option.
3235 An entry such as +xxxx was not recognized, and was treated as a literal
3238 19. Ensure that header lines added by a DATA ACL are included in the reject log
3239 if the ACL subsequently rejects the message.
3241 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3242 MD5 (which is deprecated).
3244 21. When testing a filter file using -bf, Exim was writing a message when it
3245 took the sender from a "From " line in the message, but it was not doing so
3246 when it took $return_path from a Return-Path: header line. It now does.
3248 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3249 with a valid header line field name (a series of printing characters
3250 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3252 23. Changed "disc" in the source to "disk" to conform to the documentation and
3253 the book and for uniformity.
3255 24. Ignore Sendmail's -Ooption=value command line item.
3257 25. When execve() failed while trying to run a command in a pipe transport,
3258 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3259 could be confused with a return value of 69 from the command itself. This
3260 has been changed to 127, the value the shell returns if it is asked to run
3261 a non-existent command. The wording for the related log line suggests a
3262 non-existent command as the problem.
3264 26. If received_header_text expands to an empty string, do not add a Received:
3265 header line to the message. (Well, it adds a token one on the spool, but
3266 marks it "old" so that it doesn't get used or transmitted.)
3268 27. Installed eximstats 1.28 (addition of -nt option).
3270 28. There was no check for failure on the call to getsockname() in the daemon
3271 code. This can fail if there is a shortage of resources on the system, with
3272 ENOMEM, for example. A temporary error is now given on failure.
3274 29. Contrary to the C standard, it seems that in some environments, the
3275 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3276 program. Exim now does this explicitly; it affects the formatting of
3277 timestamps using strftime().
3279 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3280 complaints. I've now initialized all the variables, to avoid this.
3282 32. Header lines added by a system filter were not being "seen" during
3283 transport-time rewrites.
3285 33. The info_callback() function passed to OpenSSL is set up with type void
3286 (*)(SSL *, int, int), as described somewhere. However, when calling the
3287 function (actually a macro) that sets it up, the type void(*)() is
3288 expected. I've put in a cast to prevent warnings from picky compilers.
3290 34. If a DNS black list lookup found a CNAME record, but there were no A
3291 records associated with the domain it pointed at, Exim crashed.
3293 35. If a DNS black list lookup returned more than one A record, Exim ignored
3294 all but the first. It now scans all returned addresses if a particular IP
3295 value is being sought. In this situation, the contents of the
3296 $dnslist_value variable are a list of all the addresses, separated by a
3299 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3300 to accept a host name and all its aliases if the forward lookup for any of
3301 them yielded the IP address of the incoming connection. Now it accepts only
3302 those names whose forward lookup yields the correct IP address. Any other
3303 names are discarded. This closes a loophole whereby a rogue DNS
3304 administrator could create reverse DNS records to break through a
3305 wildcarded host restriction in an ACL.
3307 37. If a user filter or a system filter that ran in a subprocess used any of
3308 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3309 the wrong values were passed to the pipe command ($thisaddress had the
3310 value of $0, $0 had the value of $1, etc). This bug was introduced by
3311 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3313 38. Improved the line breaking for long SMTP error messages from ACLs.
3314 Previously, if there was no break point between 40 and 75 characters, Exim
3315 left the rest of the message alone. Two changes have been made: (a) I've
3316 reduced the minimum length to 35 characters; (b) if it can't find a break
3317 point between 35 and 75 characters, it looks ahead and uses the first one
3318 that it finds. This may give the occasional overlong line, but at least the
3319 remaining text gets split now.
3321 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3322 file descriptors might be low, and that setting 1000 would always raise it.
3323 It turns out that in some environments, the limit is already over 1000 and
3324 that lowering it causes trouble. So now Exim takes care not to decrease it.
3326 40. When delivering a message, the value of $return_path is set to $sender_
3327 address at the start of routing (routers may change the value). By an
3328 oversight, this default was not being set up when an address was tested by
3329 -bt or -bv, which affected the outcome if any router or filter referred to
3332 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3333 log only when "message" or "log_message" are set. However, if one of the
3334 conditions was an address verification, or a call to a nested ACL, the
3335 messages generated by the underlying test were being passed through. This
3336 no longer happens. The underlying message is available in $acl_verify_
3337 message for both "message" and "log_message" expansions, so it can be
3338 passed through if needed.
3340 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3341 new expansion $bh_ to give the encoded byte string without charset
3342 translation. Translation happens only if iconv() is available; HAVE_ICONV
3343 indicates this at build time. HEADERS_CHARSET gives the charset to
3344 translate to; headers_charset can change it in the configuration, and
3345 "headers charset" can change it in an individual filter file.
3347 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3348 that creates RFC 2047 encoded "words" labels them as that charset instead
3349 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3350 expansion operator; (ii) when Exim creates a From: line for a local
3351 message; (iii) when a header line is rewritten to include a "phrase" part.
3353 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3354 buggy, causing it to skip the first lines of messages whose message ID
3355 ended in 'D'. This would not have bitten before Exim release 4.14, because
3356 message IDs were unlikely to end in 'D' before then. The effect was to have
3357 incorrect size information for certain domains.
3359 45. #include "config.h" was missing at the start of the crypt16.c module. This
3360 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3363 46. If there was a timeout during a "random" callout check, Exim treated it as
3364 a failure of the random address, and carried on sending RSET and the real
3365 address. If the delay was just some slowness somewhere, the response to the
3366 original RCPT would be taken as a response to RSET and so on, causing
3367 mayhem of various kinds.
3369 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3370 when I implemented it. It didn't allow for the fact that some option values
3371 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3372 the right test for positive values.
3374 48. Domain names in DNS records are case-independent. Exim always looks them up
3375 in lower case. Some resolvers return domain names in exactly the case they
3376 appear in the zone file, that is, they may contain uppercase letters. Not
3377 all resolvers do this - some return always lower case. Exim was treating a
3378 change of case by a resolver as a change of domain, similar to a widening
3379 of a domain abbreviation. This triggered its re-routing code and so it was
3380 trying to route what was effectively the same domain again. This normally
3381 caused routing to fail (because the router wouldn't handle the domain
3382 twice). Now Exim checks for this case specially, and just changes the
3383 casing of the domain that it ultimately uses when it transmits the message
3386 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3389 50. If a filter generated a file delivery with a non-absolute name (possible if
3390 no home directory exists for the router), the forbid_file option was not
3393 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3394 the existing equality matching.
3396 52. Exim was using ints instead of ino_t variables in some places where it was
3397 dealing with inode numbers.
3399 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3400 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3401 called TMPDIR, and if it finds it is different, it changes its value.
3403 54. The smtp_printf() function is now made available to local_scan() so
3404 additional output lines can be written before returning. There is also an
3405 smtp_fflush() function to enable the detection of a dropped connection.
3406 The variables smtp_input and smtp_batched_input are exported to
3409 55. Changed the default runtime configuration: the message "Unknown user"
3410 has been removed from the ACL, and instead placed on the localuser router,
3411 using the cannot_route_message feature. This means that any verification
3412 failures that generate their own messages won't get overridden. Similarly,
3413 the "Unrouteable address" message that was in the ACL for unverifiable
3414 relay addresses has also been removed.
3416 56. Added hosts_avoid_esmtp to the smtp transport.
3418 57. The exicyclog script was not checking for the esoteric option
3419 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3420 will work only if exicyclog is run under the appropriate euid.
3422 58. Following a discussion on the list, the rules by which Exim recognises line
3423 endings on incoming messages have been changed. The -dropcr and drop_cr
3424 options are now no-ops, retained only for backwards compatibility. The
3425 following line terminators are recognized: LF CRLF CR. However, special
3426 processing applies to CR:
3428 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3429 nor a local message in the state where . is a terminator.
3431 (ii) If a bare CR is encountered in a header line, an extra space is added
3432 after the line terminator so as not to end the header. The reasoning
3433 behind this is that bare CRs in header lines are most likely either
3434 to be mistakes, or people trying to play silly games.
3436 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3437 was being incorrectly given as 18 bytes larger than it should have been.
3438 This is a VOB (very old bug).
3440 60. This may never have affected anything current, but just in case it has:
3441 When the local host is found other than at the start of a list of hosts,
3442 the local host, those with the same MX, and any that follow, are discarded.
3443 When the list in question was part of a longer list of hosts, the following
3444 hosts (not currently being processed) were also being discarded. This no
3445 longer happens. I'm not sure if this situation could ever has previously
3448 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3451 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3452 header. This is something that is recommended in a new Internet Draft, and
3453 is something that is documented as being done by Sendmail. There are two
3454 possible values. For messages generated by the autoreply transport, Exim
3457 Auto-Submitted: auto-replied
3459 whereas for all other generated messages (e.g. bounces) it adds
3461 Auto-Submitted: auto-generated
3463 63. The "personal" condition in filters now includes a test for the
3464 Auto-Submitted: header. If it contains the string "auto-" the message it
3465 not considered personal.
3467 64. Added rcpt_include_affixes as a generic transport option.
3469 65. Added queue_only_override (default true).
3471 66. Added the syslog_duplication option.
3473 67. If what should have been the first header line of a message consisted of
3474 a space followed by a colon, Exim was mis-interpreting it as a header line.
3475 It isn't of course - it is syntactically invalid and should therefore be
3476 treated as the start of the message body. The misbehaviour could have
3477 caused a number of strange effects, including loss of data in subsequent
3478 header lines, and spool format errors.
3480 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3481 client host had authenticated. This control can now be exercised by an ACL
3482 for more flexibility.
3484 69. By default, callouts do not happen when testing with -bh. There is now a
3485 variant, -bhc, which does actually run the callout code, including
3486 consulting and updating the callout cache.
3488 70. Added support for saslauthd authentication, courtesy of Alexander
3491 71. If statvfs() failed on the spool or log directories while checking their
3492 size for availability, Exim confusingly gave the error "space shortage".
3493 Furthermore, in debugging mode it crashed with a floating point exception.
3494 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3495 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3496 serious problem, Exim now writes to the main and panic logs when this
3497 happens, with details of the failure. It then refuses to accept the
3498 incoming message, giving the message "spool directory problem" or "log
3499 directory problem" with a 421 code for SMTP messages.
3501 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3502 0, 1, and 2 exist, because some OS complain for execs without them (see
3503 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3504 descriptors. However, the code omitted to check that the open succeeded,
3505 causing mysterious errors if for some reason the permissions on /dev/null
3506 got screwed. Now Exim writes a message to the main and panic logs, and
3507 bombs out if it can't open /dev/null.
3509 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3510 interact so that it is all more flexible. It is supposed to remain
3511 backwards compatible. Also added extra_local_interfaces.
3513 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3514 to bomb out with an assertion failure - to the client this appears as a
3515 connection drop. This problem occurs in the part of the code that was taken
3516 from the Samba project. Fortunately, the assertion is in a very simple
3517 function, so I have fixed this by reproducing the function inline in the
3518 one place where it is called, and arranging for authentication to fail
3519 instead of killing the process with assert().
3521 75. The SPA client code was not working when the server requested OEM rather
3522 than Unicode encoding.
3524 76. Added code to make require_files with a specific uid setting more usable in
3525 the case where statting the file as root fails - usually a non-root-mounted
3526 NFS file system. When this happens and the failure is EACCES, Exim now
3527 forks a subprocess and does the per-uid checking as the relevant uid.
3529 77. Added process_log_path.
3531 78. If log_file_path was not explicitly set, a setting of check_log_space or
3532 check_log_inodes was ignored.
3534 79. If a space check for the spool or log partitions fails, the incident is now
3535 logged. Of course, in the latter case the data may get lost...
3537 80. Added the %p formatting code to string_format() so that it can be used to
3538 print addresses in debug_print(). Adjusted all the address printing in the
3539 debugging in store.c to use %p rather than %d.
3541 81. There was a concern that a line of code in smtp_in.c could overflow a
3542 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3543 initially expressed, the concern was not well-founded, because trailing
3544 spaces are removed early. However, if the trailing spaces were followed by
3545 a NULL, they did not get removed, so the overflow was possible. Two fixes
3548 (a) I re-wrote the offending code in a cleaner fashion.
3549 (b) If an incoming SMTP command contains a NULL character, it is rejected
3552 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3553 runs initgroups(), so that if the Exim user is in any additional groups,
3554 they will be used during message reception.
3560 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob