1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.264 2005/11/15 11:23:43 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.
143 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
144 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
147 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
148 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
150 PH/03 Typo: missing ".o" in src/pcre/Makefile.
152 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
153 header line, restrict the check to what is listed in RFCs 2369 and 2929.
154 Also, for "Auto-Submitted", treat anything other than "no" as
155 non-personal, in accordance with RFC 3834. (Previously it treated
156 anything starting "auto-" as non-personal.)
158 TF/01 The control=submission/name=... option had a problem with syntax
159 errors if the name included a slash character. The /name= option
160 now slurps the rest of the string, so it can include any characters
161 but it must come last in the list of options (after /sender_retain
164 PH/05 Some modifications to the interface to the fake nameserver for the new
167 PH/06 Add ${if def:sender_address {(envelope-from <$sender_address>)\n\t}}
168 into the default Received: header string.
175 TK/01 Added the "success_on_redirect" address verification option. See
176 NewStuff for rationale and an example.
178 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
180 PH/02 Patch to exigrep to allow it to work on syslog lines.
182 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
183 fread() to skip over the body file's header line, because in Cygwin the
184 header line is locked and is inaccessible.
186 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
187 co-exist for some time) to make it clear that it is the Exim ID that is
188 referenced, not the Message-ID: header line.
190 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
191 string_format() function, because snprintf() does not exist on all
194 PH/06 The use of forbid_filter_existstest now also locks out the use of the
195 ${stat: expansion item.
197 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
198 protocol synchronization error", to keep the pedants happy.
200 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
201 well as for IRIX systems, when gcc is being used. See the host.c source
204 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
206 PH/10 Named domain lists were not working if used in a queue_smtp_domains
209 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
210 transport and to the smtp transport in LMTP mode.
212 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
214 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
215 run a filter in a subprocess. This could lead to confusion in subsequent
216 lookups in the parent process. There should also be a search_tidyup() at
217 the end of the subprocess.
219 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
220 only if the host matched helo_try_verify_hosts, which caused the
221 verification to occur when the EHLO/HELO command was issued. The ACL just
222 tested the remembered result. Now, if a previous verification attempt has
223 not happened, "verify = helo" does it there and then.
225 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
227 TK/03 Fix log output including CR from clamd.
229 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
230 exist provoked a memory error which could cause a segfault.
232 PH/15 Installed PCRE 6.2
234 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
236 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
237 of the problem. Specifically, suggested +O2 rather than +O1 for the
240 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
242 PH/20 If a delivery was routed to a non-standard port by means of an SRV
243 record, the port was not correctly logged when the outgoing_port log
244 selector was set (it logged the transort's default port).
246 PH/21 Added support for host-specific ports to manualroute, queryprogram,
247 fallback_hosts, and "hosts" in the smtp transport.
249 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
250 host errors such as "Connection refused".
252 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
253 authentication with radiusclient 0.4.9:
255 - Error returned from rc_read_config was caught wrongly
256 - Username/password not passed on to radius server due to wrong length.
258 The presumption is that some radiusclient API changes for 4.51/PH/17
259 were not taken care of correctly. The code is still untested by me (my
260 Linux distribution still has 0.3.2 of radiusclient), but it was
261 contributed by a Radius user.
263 PH/24 When doing a callout, the value of $domain wasn't set correctly when
264 expanding the "port" option of the smtp transport.
266 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
267 while reading a MIME header. Thanks to Tom Hughes for a patch.
269 PH/24 Include config.h inside local_scan.h so that configuration settings are
272 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
273 that in an ACL for RCPT (for example), you can examine exactly what was
276 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
277 commands, but it was not correctly comparing the address with the actual
278 client host address. Thus, it would show the EHLO address in Received:
279 header lines when this was not necessary.
281 PH/27 Added the % operator to ${eval:}.
283 PH/28 Exim tries to create and chdir to its spool directory when it starts;
284 it should be ignoring failures (because with -C, for example, it has lost
285 privilege). It wasn't ignoring creation failures other than "already
288 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
291 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
292 because an input line was too long, either on its own, or by virtue of
293 too many continuations, the temporary file was not being removed, and the
294 return code was incorrect.
296 PH/31 Missing "BOOL" in function definition in filtertest.c.
298 PH/32 Applied Sieve patches from the maintainer.
300 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
302 PH/33 Added "verify = not_blind".
304 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
305 Local/Makefile (with some defaults set). These are used in built scripts
306 such as exicyclog, but they have never been used in the exim_install
307 script (though there are many overriding facilities there). I have
308 arranged that the exim_install script now takes note of these two
311 PH/35 Installed configuration files for Dragonfly.
313 PH/36 When a locally submitted message by a trusted user did not contain a
314 From: header, and the sender address was obtained from -f or from an SMTP
315 MAIL command, and the trusted user did not use -F to supply a sender
316 name, $originator_name was incorrectly used when constructing a From:
317 header. Furthermore, $originator_name was used for submission mode
318 messages from external hosts without From: headers in a similar way,
319 which is clearly wrong.
321 PH/37 Added control=suppress_local_fixups.
323 PH/38 When log_selector = +received_sender was set, and the addition of the
324 sender made the log line's construction buffer exactly full, or one byte
325 less than full, an overflow happened when the terminating "\n" was
328 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
329 when the result of a list match is failure because a DNS lookup failed.
331 PH/40 RM_COMMAND is now used in the building process.
333 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
334 the "build-* directories that it finds.
336 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
337 address in a domain literal was a prefix of an interface address.
339 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
340 when verifying a sender address, unless rewrite_headers is false.
342 PH/44 Wrote a long comment about why errors_to addresses are verified as
343 recipients, not senders.
345 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
346 the ratelimit ACL was added.
348 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
350 PH/46 Added extra information about PostgreSQL errors to the error string.
352 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
353 avoiding the need to install special zones in a real server. This is
354 backwards compatible; if it can't find the fake resolver, it drops back.
355 Thus, both old and new test suites can be run.
357 TF/02 Added util/ratelimit.pl
359 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
360 clock is set back in time.
362 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
363 Candler <B.Candler@pobox.com>.
365 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
366 OK for addresses that are the result of redirections.
368 PH/48 A number of further additions for the benefit of the new test suite,
369 including a fake gethostbyname() that interfaces to the fake DNS resolver
372 TF/06 The fix for widen_domains has also been applied to qualify_single and
373 search_parents which are the other dnslookup options that can cause
376 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
379 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
381 TF/07 Exim produced the error message "an SRV record indicated no SMTP
382 service" if it encountered an MX record with an empty target hostname.
383 The message is now "an MX or SRV record indicated no SMTP service".
385 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
386 if the DNS of the sending site is misconfigured. This is quite a
387 common situation. This change restores the behaviour of treating a
388 helo verification defer as a failure.
390 PH/51 If self=fail was set on a router, the bounce message did not include the
391 actual error message.
397 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
399 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
400 command itself ended in error, the underlying message about the transport
401 filter timeout was being overwritten with the pipe command error. Now the
402 underlying error message should be appended to the second error message.
404 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
406 PH/02 Reduce the amount of output that "make" produces by default. Full output
407 can still be requested.
409 PH/03 The warning log line about a condition test deferring for a "warn" verb
410 was being output only once per connection, rather than after each
411 occurrence (because it was using the same function as for successful
412 "warn" verbs). This seems wrong, so I have changed it.
414 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
415 it should not have, which might have caused a crash in the right
416 circumstances, but probably never did.
418 PH/04 Installed a modified version of Tony Finch's patch to make submission
419 mode fix the return path as well as the Sender: header line, and to
420 add a /name= option so that you can make the user's friendly name appear
423 TF/03 Added the control = fakedefer ACL modifier.
425 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
426 Mark Lowes for thorough testing.
428 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
430 TK/03 Merged latest SRS patch from Miles Wilton.
432 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
433 with the definition in sysexits.h (which is #included earlier).
434 Fortunately, Exim does not actually use EX_OK. The code used to try to
435 preserve the sysexits.h value, by assumimg that macro definitions were
436 scanned for macro replacements. I have been disabused of this notion,
437 so now the code just undefines EX_OK before #including unistd.h.
439 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
440 in the smtp transport. When a block could not be written in a single
441 write() function, the timeout was being re-applied to each part-write.
442 This seems wrong - if the receiver was accepting one byte at a time it
443 would take for ever. The timeout is now adjusted when this happens. It
444 doesn't have to be particularly precise.
446 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
447 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
449 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
450 without a domain if the check to <postmaster@domain> fails.
452 SC/01 Eximstats: added -xls and the ability to specify output files
453 (patch written by Frank Heydlauf).
455 SC/02 Eximstats: use FileHandles for outputing results.
457 SC/03 Eximstats: allow any combination of xls, txt, and html output.
459 SC/04 Eximstats: fixed display of large numbers with -nvr option
461 SC/05 Eximstats: fixed merging of reports with empty tables.
463 SC/06 Eximstats: added the -include_original_destination flag
465 SC/07 Eximstats: removed tabs and trailing whitespace.
467 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
469 TK/06 MBOX spool code: Add real "From " MBOX separator line
470 so the .eml file is really in mbox format (even though
471 most programs do not really care). Patch from Alex Miller.
473 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
474 The latter is generated from $received_to and is only set if the
475 message has one envelope recipient. SA can use these headers,
476 obviously out-of-the-box. Patch from Alex Miller.
478 PH/08 The ${def test on a variable was returning false if the variable's
479 value was "0", contrary to what the specification has always said!
480 The result should be true unless the variable is empty.
482 PH/09 The syntax error of a character other than { following "${if
483 def:variable_name" (after optional whitespace) was not being diagnosed.
484 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
485 accidental colon was present, for example, could give incorrect results.
487 PH/10 Tidied the code in a number of places where the st_size field of a stat()
488 result is used (not including appendfile, where other changes are about
491 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
492 This involved changing a lot of size variables from int to off_t. It
493 should work with maildirs and everything.
495 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
496 spamd dying while we are connected to it.
498 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
501 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
502 function for BDB 4.3.
504 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
506 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
507 the sources for PCRE have been re-arranged and more files are now
510 PH/15 The code I had for printing potentially long long variables in PH/11
511 above was not the best (it lost precision). The length of off_t variables
512 is now inspected at build time, and an appropriate printing format (%ld
513 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
514 to be "long long int" or "long int". This is needed for the internal
515 formatting function string_vformat().
517 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
518 the configuration file to be ":syslog", then the script "guesses" where
519 the logs files are, rather than using the compiled in default. In our
520 case the guess is not the same as the compiled default, so the script
521 suddenly stopped working when I started to use syslog. The patch checks
522 to see if log_file_path is "". If so, it attempts to read it from exim
523 with no configuration file to get the compiled in version, before it
524 falls back to the previous guessing code."
526 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
527 implementing BATV in an Exim configuration. See NewStuff for the gory
530 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
531 Makefile that are specific to HP-UX.
533 PH/18 If the "use_postmaster" option was set for a recipient callout together
534 with the "random" option, the postmaster address was used as the MAIL
535 FROM address for the random test, but not for the subsequent recipient
536 test. It is now used for both.
538 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
539 patch removes a few documentation additions to RFC 3028, because the
540 latest draft now contains them. It adds the new en;ascii-case comparator
541 and a new error check for 8bit text in MIME parts. Comparator and
542 require names are now matched exactly. I enabled the subaddress
543 extension, but it is not well tested yet (read: it works for me)."
545 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
546 rework some of the code of TK/09 above to avoid the hardwired use of
547 "%lld" and "long long". Replaced the call to snprintf() with a call to
550 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
551 records point to non-existent hosts", "retry timeout exceeded", and
552 "retry time not reached for any host after a long failure period".
554 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
555 experimental DomainKeys support:
557 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
558 (2) On an error such as an illegally used "control", the wrong name for
559 the control was given.
561 These problems did NOT occur unless DomainKeys support was compiled.
563 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
565 PH/24 Added ${if match_ip condition.
567 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
568 databases so that it will be absolutely obvious if a crash occurs in the
569 DB library. This is a regular occurrence (often caused by mis-matched
572 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
573 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
574 picked up on a user's system that detects such things. There doesn't seem
575 to be a gcc warning option for this - only an attribute that has to be
576 put on the function's prototype. It seems that in Fedora Core 4 they have
577 set this on a number of new functions. No doubt there will be more in due
580 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
581 specify a transport. However, if an address that was verified by such a
582 router was the subject of a callout, Exim crashed because it tried to
583 read the rcpt_include_affixes from the non-existent transport. Now it
584 just assumes that the setting of that option is false. This bug was
585 introduced by 4.51/PH/31.
587 PH/28 Changed -d+all to exclude +memory, because that information is very
588 rarely of interest, but it makes the output a lot bigger. People tend to
589 do -d+all out of habit.
591 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
592 code in os-type was giving problems when libc.so lives in lib64, like on
595 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
596 aren't the modern standard, and it seems that some systems' include files
597 don't always have them. Exim was already checking for some of the newer
598 ones like T_AAAA, and defining it itself. I've added checks for all the
599 record types that Exim uses.
601 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
602 not automatically generating a new one, as it is supposed to. This
603 prevented TLS from working. If the file did exist, but contained invalid
604 data, a new version was generated, as expected. It was only the case of a
605 non-existent file that was broken.
607 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
608 with a change in libdomainkeys > 0.64.
610 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
611 from DNS. If the selector record carries the flag, it now has
612 precedence over the domain-wide flag.
614 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
616 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
617 the use of a port name that isn't defined in /etc/services) occurred, the
618 message was deferred as in a normal delivery, and thus remained on the
619 spool, instead of being failed because of the mua_wrapper setting. This
620 is now fixed, and I tidied up some of the mua_wrapper messages at the
623 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
624 the messages in a hash of arrays rather than using individual hashes.
625 This is a bit cleaner and results in dramatic memory savings, albeit
626 at a slight CPU cost.
628 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
629 as requested by Marc Sherman.
631 SC/10 Eximstats: added histograms for user specified patterns as requested
634 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
636 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
637 fopen() in the content-scanning modules that did not already have it.
643 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
644 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
646 TK/02 Fix ACL "control" statement not being available in MIME ACL.
648 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
650 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
651 to test Sieve filters that use "vacation".
653 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
654 that changes the way the GnuTLS parameters are stored in the cache file.
655 The new format can be generated externally. For backward compatibility,
656 if the data in the cache doesn't make sense, Exim assumes it has read an
657 old-format file, and it generates new data and writes a new file. This
658 means that you can't go back to an older release without removing the
661 PH/03 A redirect router that has both "unseen" and "one_time" set does not
662 work if there are any delivery delays because "one_time" forces the
663 parent to be marked "delivered", so its unseen clone is never tried
664 again. For this reason, Exim now forbids the simultaneous setting of
667 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
668 redirected to themselves ("homonym" addresses). Read the long ChangeLog
669 entry if you want to know the details. The fix, however, neglected to
670 consider the case when local delivery batching is involved. The test for
671 "previously delivered" was not happening when checking to see if an
672 address could be batched with a previous (undelivered) one; under
673 certain circumstances this could lead to multiple deliveries to the same
676 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
677 in its include files, and this causes problems building Exim.
679 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
680 header_syntax) but Exim was just ignoring anything given after a slash.
681 In particular, this caused confusion with an attempt to use "verify =
682 reverse_host_lookup/defer_ok". An error is now given when options are
683 supplied for verify items that do not have them. (Maybe reverse_host_
684 lookup should have a defer_ok option, but that's a different point.)
686 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
687 defined by RFC 821) to 2048, because there were problems with some AUTH
688 commands, and RFC 1869 says the size should be increased for extended
689 SMTP commands that take arguments.
691 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
694 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
695 "unknown" error; now it says that the functionality isn't in the binary.
697 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
698 an address' error message when a string expansion fails (syntax or
699 whatever). Otherwise the password may appear in the log. Following change
700 PH/42 below, there is no longer a chance of it appearing in a bounce
703 PH/11 Installed exipick version 20050225.0 from John Jetmore.
705 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
706 of its addresses was ever tried. (Bugzilla bug #2.)
708 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
709 the result incorrectly in the debug output. (It correctly added a newline
710 to what was transported.)
712 TF/01 Added $received_time.
714 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
715 commented out examples of how to interface to a virus scanner and to
716 SpamAssassin. Also added commented examples of av_scanner and
717 spamd_address settings.
719 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
720 and controls are allowed in which ACLs. There were a couple of minor
721 errors. Some of the entries in the conditions table (which is a table of
722 where they are NOT allowed) were getting very unwieldy; rewrote them as a
723 negation of where the condition IS allowed.
725 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
727 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
728 header file does not have a version number, so I've had to invent a new
729 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
730 API. The code is untested by me (my Linux distribution still has 0.3.2 of
731 radiusclient), but it was contributed by a Radius user.
733 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
734 files or directories, for OpenSSL.
736 PH/19 When an Exim process that is running as root has to create an Exim log
737 file, it does so in a subprocess that runs as exim:exim so as to get the
738 ownership right at creation (otherwise, other Exim processes might see
739 the file with the wrong ownership). There was no test for failure of this
740 fork() call, which would lead to the process getting stuck as it waited
741 for a non-existent subprocess. Forks do occasionally fail when resources
742 run out. I reviewed all the other calls to fork(); they all seem to check
745 PH/20 When checking for unexpected SMTP input at connect time (before writing
746 the banner), Exim was not dealing correctly with a non-positive return
747 from the read() function. If the client had disconnected by this time,
748 the result was a log entry for a synchronization error with an empty
749 string after "input=" when read() returned zero. If read() returned -1
750 (an event I could not check), uninitialized data bytes were printed.
751 There were reports of junk text (parts of files, etc) appearing after
754 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
756 PH/22 Added support for macro redefinition, and (re)definition in between
757 driver and ACL definitions.
759 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
760 forgetting to use the resulting value; it was using the unexpanded value.
762 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
763 hadn't been configured. The fix is from Juergen Kreileder, who
764 understands it better than I do:
766 "Here's what I see happening with three configured cyrus_sasl
767 authenticators configured (plain, login, cram-md5):
769 On startup auth_cyrus_sasl_init() gets called for each of these.
770 This means three calls to sasl_listmech() without a specified mech_list.
771 => SASL tests which mechs of all available mechs actually work
772 => three warnings about OTP not working
773 => the returned list contains: plain, login, cram-md5, digest-md5, ...
775 With the patch, sasl_listmech() also gets called three times. But now
776 SASL's mech_list option is set to the server_mech specified in the the
777 authenticator. Or in other words, the answer from sasl_listmech()
778 gets limited to just the mech you're testing for (which is different
780 => the return list contains just 'plain' or 'login', 'cram-md5' or
781 nothing depending on the value of ob->server_mech.
783 I've just tested the patch: Authentication still works fine,
784 unavailable mechs specified in the exim configuration are still
785 caught, and the auth.log warnings about OTP are gone."
787 PH/25 When debugging is enabled, the contents of the command line are added
788 to the debugging output, even when log_selector=+arguments is not
791 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
792 answer is "GNU", and only if the return is "GNU/something" is the answer
795 PH/27 $acl_verify_message is now set immediately after the failure of a
796 verification in an ACL, and so is available in subsequent modifiers. In
797 particular, the message can be preserved by coding like this:
799 warn !verify = sender
800 set acl_m0 = $acl_verify_message
802 Previously, $acl_verify_message was set only while expanding "message"
803 and "log_message" when a very denied access.
805 PH/28 Modified OS/os.c-Linux with
807 -#ifndef OS_LOAD_AVERAGE
808 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
810 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
811 nomenclature these days.)
813 PH/29 Installed patch from the Sieve maintainer that adds the options
814 sieve_useraddress and sieve_subaddress to the redirect router.
816 PH/30 In these circumstances:
817 . Two addresses routed to the same list of hosts;
818 . First host does not offer TLS;
819 . First host accepts first address;
820 . First host gives temporary error to second address;
821 . Second host offers TLS and a TLS session is established;
822 . Second host accepts second address.
823 Exim incorrectly logged both deliveries with the TLS parameters (cipher
824 and peerdn, if requested) that were in fact used only for the second
827 PH/31 When doing a callout as part of verifying an address, Exim was not paying
828 attention to any local part prefix or suffix that was matched by the
829 router that accepted the address. It now behaves in the same way as it
830 does for delivery: the affixes are removed from the local part unless
831 rcpt_include_affixes is set on the transport.
833 PH/32 Add the sender address, as F=<...>, to the log line when logging a
834 timeout during the DATA phase of an incoming message.
836 PH/33 Sieve envelope tests were broken for match types other than :is. I have
837 applied a patch sanctioned by the Sieve maintainer.
839 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
840 the uid or gid is negative. A case of a negative gid caused this to be
841 noticed. The fix allows for either to be negative.
843 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
844 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
847 PH/36 The change PH/12 above was broken. Fixed it.
849 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
850 the grounds that routing the same address twice would always produce the
851 same answer. This might have been true once, but it is certainly no
852 longer true now. Routing a child address may depend on the previous
853 routing that produced that child. Some complicated redirection strategies
854 went wrong when messages had multiple recipients, and made Exim's
855 behaviour dependent on the order in which the addresses were given.
857 I have moved the duplicate checking until after the routing is complete.
858 Exim scans the addresses that are assigned to local and remote
859 transports, and removes any duplicates. This means that more work will be
860 done, as duplicates will always all be routed, but duplicates are
861 presumably rare, so I don't expect this is of any significance.
863 For deliveries to pipes, files, and autoreplies, the duplicate checking
864 still happens during the routing process, since they are not going to be
867 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
868 It corrects a timeout issue with spamd. This is Ian's comment: "The
869 background is that sometimes spamd either never reads data from a
870 connection it has accepted, or it never writes response data. The exiscan
871 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
872 blindly assumes that writes won't block so it may never time out."
874 PH/39 Allow G after quota size as well as K and M.
876 PH/40 The value set for $authenticated_id in an authenticator may not contain
877 binary zeroes or newlines because the value is written to log lines and
878 to spool files. There was no check on this. Now the value is run through
879 the string_printing() function so that such characters are converted to
880 printable escape sequences.
882 PH/41 $message_linecount is a new variable that contains the total number of
883 lines in the message. Compare $body_linecount, which is the count for the
886 PH/42 Exim no longer gives details of delivery errors for specific addresses in
887 bounce and delay warning messages, except in certain special cases, which
890 (a) An SMTP error message from a remote host;
891 (b) A message specified in a :fail: redirection;
892 (c) A message specified in a "fail" command in a system filter;
893 (d) A message specified in a FAIL return from the queryprogram router;
894 (e) A message specified by the cannot_route_message router option.
896 In these cases only, Exim does include the error details in bounce and
897 warning messages. There are also a few cases where bland messages such
898 as "unrouteable address" or "local delivery error" are given.
900 PH/43 $value is now also set for the "else" part of a ${run expansion.
902 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
903 being worked on, but at least Exim now implements the latest version to
906 PH/45 In a pipe transport, although a timeout while waiting for the pipe
907 process to complete was treated as a delivery failure, a timeout while
908 writing the message to the pipe was logged, but erroneously treated as a
909 successful delivery. Such timeouts include transport filter timeouts. For
910 consistency with the overall process timeout, these timeouts are now
911 treated as errors, giving rise to delivery failures by default. However,
912 there is now a new Boolean option for the pipe transport called
913 timeout_defer, which, if set TRUE, converts the failures into defers for
914 both kinds of timeout. A transport filter timeout is now identified in
917 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
918 systems where "make" and "gmake" are different, calling "gmake" at top
919 level broke things. I've arranged for the value of $(MAKE) to be passed
920 from the Makefile to this script so that it can call the same version of
924 A note about Exim versions 4.44 and 4.50
925 ----------------------------------------
927 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
928 changes of various kinds. As a consequence, a big documentation update was
929 needed. This delayed the release for rather longer than seemed good, especially
930 in the light of a couple of (minor) security issues. Therefore, the changes
931 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
932 release. So 4.44 and 4.50 are in effect two different branches that both start
935 I have left the 4.50 change log unchanged; it contains all the changes since
936 4.43. The change log for 4.44 is below; many of its items are identical to
937 those for 4.50. This seems to be the most sensible way to preserve the
938 historical information.
944 1. Minor wording change to the doc/README.SIEVE file.
946 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
947 computation of the current number of files was incorrect.
949 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
950 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
953 4. Give more explanation in the error message when the command for a transport
954 filter fails to execute.
956 5. There are several places where Exim runs a non-Exim command in a
957 subprocess. The SIGUSR1 signal should be disabled for these processes. This
958 was being done only for the command run by the queryprogram router. It is
959 now done for all such subprocesses. The other cases are: ${run, transport
960 filters, and the commands run by the lmtp and pipe transports.
962 6. Added CONFIGURE_GROUP build-time option.
964 7. Some older OS have a limit of 256 on the maximum number of file
965 descriptors. Exim was using setrlimit() to set 1000 as a large value
966 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
967 systems. I've change it so that if it can't get 1000, it tries for 256.
969 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
970 was an oversight, and furthermore, ever since the addition of extra
971 controls (e.g. 4.43/32), the checks on when to allow different forms of
972 "control" were broken. There should now be diagnostics for all cases when a
973 control that does not make sense is encountered.
975 9. Added the /retain_sender option to "control=submission".
977 10. $recipients is now available in the predata ACL (oversight).
979 11. Tidy the search cache before the fork to do a delivery from a message
980 received from the command line. Otherwise the child will trigger a lookup
981 failure and thereby defer the delivery if it tries to use (for example) a
982 cached ldap connection that the parent has called unbind on.
984 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
985 of $address_data from the recipient verification was clobbered by the
988 13. The value of address_data from a sender verification is now available in
989 $sender_address_data in subsequent conditions in the ACL statement.
991 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
993 15. Added a new option "connect=<time>" to callout options, to set a different
996 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
997 was its contents. (It was OK if the option was not defined at all.)
999 17. A "Completed" log line is now written for messages that are removed from
1000 the spool by the -Mrm option.
1002 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1003 information about exactly what failed.
1005 19. Added -dd to debug only the daemon process.
1007 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1008 handles timeouts, both on the server side and network timeouts. Renamed the
1009 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1011 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1012 It is now set to "smtps".
1014 22. $host_address is now set to the target address during the checking of
1015 ignore_target_hosts.
1017 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1018 being passed; this would have caused $sender_host_name to have been used if
1019 matching the list had actually called for a host name (not very likely,
1020 since this list is usually IP addresses). A host name is now passed as
1023 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1024 code that specifies a non-restarting handler (typically sigaction() in
1025 modern systems) in an attempt to fix a rare and obscure crash bug.
1027 25. Narrowed the window for a race in the daemon that could cause it to ignore
1028 SIGCHLD signals. This is not a major problem, because they are used only to
1029 wake it up if nothing else does.
1031 26. A malformed maildirsize file could cause Exim to calculate negative values
1032 for the mailbox size or file count. Odd effects could occur as a result.
1033 The maildirsize information is now recalculated if the size or filecount
1036 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1037 support for a long time. Removed HAVE_SYS_VFS_H.
1039 28. Installed the latest version of exipick from John Jetmore.
1041 29. In an address list, if the pattern was not a regular expression, an empty
1042 subject address (from a bounce message) matched only if the pattern was an
1043 empty string. Non-empty patterns were not even tested. This was the wrong
1044 because it is perfectly reasonable to use an empty address as part of a
1045 database query. An empty address is now tested by patterns that are
1046 lookups. However, all the other forms of pattern expect the subject to
1047 contain a local part and a domain, and therefore, for them, an empty
1048 address still always fails if the pattern is not itself empty.
1050 30. Exim went into a mad DNS loop when attempting to do a callout where the
1051 host was specified on an smtp transport, and looking it up yielded more
1052 than one IP address.
1054 31. Re-factored the code for checking spool and log partition space into a
1055 function that finds that data and another that does the check. The former
1056 is then used to implement four new variables: $spool_space, $log_space,
1057 $spool_inodes, and $log_inodes.
1059 32. The RFC2047 encoding function was originally intended for short strings
1060 such as real names; it was not keeping to the 75-character limit for
1061 encoded words that the RFC imposes. It now respects the limit, and
1062 generates multiple encoded words if necessary. To be on the safe side, I
1063 have increased the buffer size for the ${rfc2047: expansion operator from
1066 33. It is now permitted to omit both strings after an "if" condition; if the
1067 condition is true, the result is "true". As before, when the second string
1068 is omitted, a false condition yields an empty string. This makes it less
1069 cumbersome to write custom ACL and router conditions.
1071 34. Failure to deliver a bounce message always caused it to be frozen, even if
1072 there was an errors_to setting on the router. The errors_to setting is now
1075 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1076 canonical form (fully expanded) before being placed in
1077 $sender_host_address.
1079 36. The table in the code that translates DNS record types into text (T_A to
1080 "A" for instance) was missing entries for NS and CNAME. It is just possible
1081 that this could have caused confusion if both these types were looked up
1082 for the same domain, because the text type is used as part of Exim's
1083 per-process caching. But the chance of anyone hitting this buglet seems
1086 37. The dnsdb lookup has been extended in a number of ways.
1088 (1) There is a new type, "zns", which walks up the domain tree until it
1089 finds some nameserver records. It should be used with care.
1091 (2) There is a new type, "mxh", which is like "mx" except that it returns
1092 just the host names, not the priorities.
1094 (3) It is now possible to give a list of domains (or IP addresses) to be
1095 looked up. The behaviour when one of the lookups defers can be
1096 controlled by a keyword.
1098 (4) It is now possible to specify the separator character for use when
1099 multiple records are returned.
1101 38. The dnslists ACL condition has been extended: it is now possible to supply
1102 a list of IP addresses and/or domains to be looked up in a particular DNS
1105 39. Added log_selector=+queue_time_overall.
1107 40. When running the queue in the test harness, wait just a tad after forking a
1108 delivery process, to get repeatability of debugging output.
1110 41. Include certificate and key file names in error message when GnuTLS fails
1111 to set them up, because the GnuTLS error message doesn't include the name
1112 of the failing file when there is a problem reading it.
1114 42. Allow both -bf and -bF in the same test run.
1116 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1118 44. The "Exiscan patch" is now merged into the mainline Exim source.
1120 45. Sometimes the final signoff response after QUIT could fail to get
1121 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1122 < 0 before doing a fflush(). This bug looks as though it goes back to the
1123 introduction of TLS in release 3.20, but "sometimes" must have been rare
1124 because the tests only now provoked it.
1126 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1127 (this can affect the format of dates).
1129 47. exim_tidydb, when checking for the continued existence of a message for
1130 which it has found a message-specific retry record, was not finding
1131 messages that were in split spool directories. Consequently, it was
1132 deleting retry records that should have stayed in existence.
1134 48. Steve fixed some bugs in eximstats.
1136 49. The SPA authentication driver was not abandoning authentication and moving
1137 on to the next authenticator when an expansion was forced to fail,
1138 contradicting the general specification for all authenticators. Instead it
1139 was generating a temporary error. It now behaves as specified.
1141 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1142 (the order specifies the preference for clients). The order is now AES256,
1143 AES128, 3DES, ARCFOUR128.
1145 51. Small patch to Sieve code - explicitly set From: when generating an
1148 52. Exim crashed if a remote delivery caused a very long error message to be
1149 recorded - for instance if somebody sent an entire SpamAssassin report back
1150 as a large number of 550 error lines. This bug was coincidentally fixed by
1151 increasing the size of one of Exim's internal buffers (big_buffer) that
1152 happened as part of the Exiscan merge. However, to be on the safe side, I
1153 have made the code more robust (and fixed the comments that describe what
1156 53. Now that there can be additional text after "Completed" in log lines (if
1157 the queue_time_overall log selector is set), a one-byte patch to exigrep
1158 was needed to allow it to recognize "Completed" as not the last thing in
1161 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1162 patch that reportedly fixes this has been added. I am not expert enough to
1163 create a test for it. This is what the patch creator wrote:
1165 "I found a little strange behaviour of ldap code when working with
1166 Windows 2003 AD Domain, where users was placed in more than one
1167 Organization Units. When I tried to give exim partial DN, the exit code
1168 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1169 But simultaneously result of request was absolutely normal ldap result,
1170 so I produce this patch..."
1172 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1173 REFERENCE, so I have modified the code to exclude the patch when that macro
1176 55. Some experimental protocols are using DNS PTR records for new purposes. The
1177 keys for these records are domain names, not reversed IP addresses. The
1178 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1179 leaves it alone. Component reversal etc. now happens only for IP addresses.
1182 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1184 57. Double the size of the debug message buffer (to 2048) so that more of very
1185 long debug lines gets shown.
1187 58. The exicyclog utility now does better if the number of log files to keep
1188 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1190 59. Two changes related to the smtp_active_hostname option:
1192 (1) $smtp_active_hostname is now available as a variable.
1193 (2) The default for smtp_banner uses $smtp_active_hostname instead
1194 of $primary_hostname.
1196 60. The host_aton() function is supposed to be passed a string that is known
1197 to be a valid IP address. However, in the case of IPv6 addresses, it was
1198 not checking this. This is a hostage to fortune. Exim now panics and dies
1199 if the condition is not met. A case was found where this could be provoked
1200 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1201 components; fortuitously, this particular loophole had already been fixed
1202 by change 4.50/55 above.
1204 If there are any other similar loopholes, the new check in host_aton()
1205 itself should stop them being exploited. The report I received stated that
1206 data on the command line could provoke the exploit when Exim was running as
1207 exim, but did not say which command line option was involved. All I could
1208 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1209 running as the user.
1212 61. There was a buffer overflow vulnerability in the SPA authentication code
1213 (which came originally from the Samba project). I have added a test to the
1214 spa_base64_to_bits() function which I hope fixes it.
1217 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1218 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1220 63. The daemon start-up calls getloadavg() while still root for those OS that
1221 need the first call to be done as root, but it missed one case: when
1222 deliver_queue_load_max is set with deliver_drop_privilege. This is
1223 necessary for the benefit of the queue runner, because there is no re-exec
1224 when deliver_drop_privilege is set.
1226 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1227 This has been fixed.
1229 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1230 was in use, was not putting the data itself into the right store pool;
1231 consequently, it could be overwritten for a subsequent message in the same
1232 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1235 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1237 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1238 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1239 the function were treating the return as a boolean value, which happened to
1240 work because 0=false and not-0=true, but is not correct code.
1242 68. The host_aton() function was not handling scoped IPv6 addresses (those
1243 with, for example, "%eth0" on the end) correctly.
1245 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1246 negated items (that is, ~something) in unsigned ints. Some compilers
1247 apparently mutter when there is no cast.
1249 70. If an address verification called from an ACL failed, and did not produce a
1250 user-specific message (i.e. there was only a "system" message), nothing was
1251 put in $acl_verify_message. In this situation, it now puts the system
1254 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1255 session; change 4.31/43 added the unwanted input to the log line - except
1256 that it did not do this in the start of session case. It now does.
1258 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1259 This is wrong and can cause the other end to generate a synchronization
1260 error if it is another Exim or anything else that does the synchronization
1261 check. A QUIT command is no longer sent after a timeout.
1263 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1264 during host lookups.
1266 74. The defer_ok option of callout verification was not working if it was used
1267 when verifying addresses in header lines, that is, for this case:
1269 verify = header_sender/callout=defer_ok
1271 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1272 those file descriptors could be used for SMTP connections. If anything
1273 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1274 could be sent to the SMTP client, causing chaos. The daemon now opens
1275 stdin, stdout, and stderr to /dev/null when it puts itself into the
1278 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1279 log by default. The user can override this with suitable Perl magic.
1281 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1282 the log message when discard triggers, was not working for the DATA ACL or
1283 for the non-SMTP ACL.
1285 78. Error message wording change in sieve.c.
1287 79. If smtp_accept_max_per_host was set, the number of connections could be
1288 restricted to fewer than expected, because the daemon was trying to set up
1289 a new connection before checking whether the processes handling previous
1290 connections had finished. The check for completed processes is now done
1291 earlier. On busy systems, this bug wouldn't be noticed because something
1292 else would have woken the daemon, and it would have reaped the completed
1295 80. If a message was submitted locally by a user whose login name contained one
1296 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1297 It caused a spool format error. I have fixed the spool reading code. A
1298 related problem was that the "from" clause in the Received: line became
1299 illegal because of the space(s). It is now covered by ${quote_local_part.
1301 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1304 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1306 83. Patch from Sieve maintainer to fix unterminated string problem in
1307 "vacation" handling.
1309 84. Some minor changes to the Linux configuration files to help with other
1310 OS variants using glibc.
1312 85. One more patch for Sieve to update vacation handling to latest spec.
1315 ----------------------------------------------------
1316 See the note above about the 4.44 and 4.50 releases.
1317 ----------------------------------------------------
1323 1. Change 4.43/35 introduced a bug that caused file counts to be
1324 incorrectly computed when quota_filecount was set in an appendfile
1327 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1328 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1331 3. Give more explanation in the error message when the command for a transport
1332 filter fails to execute.
1334 4. There are several places where Exim runs a non-Exim command in a
1335 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1336 was being done only for the command run by the queryprogram router. It is
1337 now done for all such subprocesses. The other cases are: ${run, transport
1338 filters, and the commands run by the lmtp and pipe transports.
1340 5. Some older OS have a limit of 256 on the maximum number of file
1341 descriptors. Exim was using setrlimit() to set 1000 as a large value
1342 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1343 systems. I've change it so that if it can't get 1000, it tries for 256.
1345 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1346 was an oversight, and furthermore, ever since the addition of extra
1347 controls (e.g. 4.43/32), the checks on when to allow different forms of
1348 "control" were broken. There should now be diagnostics for all cases when a
1349 control that does not make sense is encountered.
1351 7. $recipients is now available in the predata ACL (oversight).
1353 8. Tidy the search cache before the fork to do a delivery from a message
1354 received from the command line. Otherwise the child will trigger a lookup
1355 failure and thereby defer the delivery if it tries to use (for example) a
1356 cached ldap connection that the parent has called unbind on.
1358 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1359 of $address_data from the recipient verification was clobbered by the
1360 sender verification.
1362 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1363 was its contents. (It was OK if the option was not defined at all.)
1365 11. A "Completed" log line is now written for messages that are removed from
1366 the spool by the -Mrm option.
1368 12. $host_address is now set to the target address during the checking of
1369 ignore_target_hosts.
1371 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1372 being passed; this would have caused $sender_host_name to have been used if
1373 matching the list had actually called for a host name (not very likely,
1374 since this list is usually IP addresses). A host name is now passed as
1377 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1378 code that specifies a non-restarting handler (typically sigaction() in
1379 modern systems) in an attempt to fix a rare and obscure crash bug.
1381 15. Narrowed the window for a race in the daemon that could cause it to ignore
1382 SIGCHLD signals. This is not a major problem, because they are used only to
1383 wake it up if nothing else does.
1385 16. A malformed maildirsize file could cause Exim to calculate negative values
1386 for the mailbox size or file count. Odd effects could occur as a result.
1387 The maildirsize information is now recalculated if the size or filecount
1390 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1391 support for a long time. Removed HAVE_SYS_VFS_H.
1393 18. Updated exipick to current release from John Jetmore.
1395 19. Allow an empty sender to be matched against a lookup in an address list.
1396 Previously the only cases considered were a regular expression, or an
1399 20. Exim went into a mad DNS lookup loop when doing a callout where the
1400 host was specified on the transport, if the DNS lookup yielded more than
1403 21. The RFC2047 encoding function was originally intended for short strings
1404 such as real names; it was not keeping to the 75-character limit for
1405 encoded words that the RFC imposes. It now respects the limit, and
1406 generates multiple encoded words if necessary. To be on the safe side, I
1407 have increased the buffer size for the ${rfc2047: expansion operator from
1410 22. Failure to deliver a bounce message always caused it to be frozen, even if
1411 there was an errors_to setting on the router. The errors_to setting is now
1414 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1415 canonical form (fully expanded) before being placed in
1416 $sender_host_address.
1418 24. Updated eximstats to version 1.33
1420 25. Include certificate and key file names in error message when GnuTLS fails
1421 to set them up, because the GnuTLS error message doesn't include the name
1422 of the failing file when there is a problem reading it.
1424 26. Expand error message when OpenSSL has problems setting up cert/key files.
1427 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1428 (this can affect the format of dates).
1430 28. exim_tidydb, when checking for the continued existence of a message for
1431 which it has found a message-specific retry record, was not finding
1432 messages that were in split spool directories. Consequently, it was
1433 deleting retry records that should have stayed in existence.
1435 29. eximstats updated to version 1.35
1436 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1437 1.35 - bugfix such that pie charts by volume are generated correctly
1439 30. The SPA authentication driver was not abandoning authentication and moving
1440 on to the next authenticator when an expansion was forced to fail,
1441 contradicting the general specification for all authenticators. Instead it
1442 was generating a temporary error. It now behaves as specified.
1444 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1445 (the order specifies the preference for clients). The order is now AES256,
1446 AES128, 3DES, ARCFOUR128.
1448 31. Small patch to Sieve code - explicitly set From: when generating an
1451 32. Exim crashed if a remote delivery caused a very long error message to be
1452 recorded - for instance if somebody sent an entire SpamAssassin report back
1453 as a large number of 550 error lines. This bug was coincidentally fixed by
1454 increasing the size of one of Exim's internal buffers (big_buffer) that
1455 happened as part of the Exiscan merge. However, to be on the safe side, I
1456 have made the code more robust (and fixed the comments that describe what
1459 33. Some experimental protocols are using DNS PTR records for new purposes. The
1460 keys for these records are domain names, not reversed IP addresses. The
1461 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1462 leaves it alone. Component reversal etc. now happens only for IP addresses.
1465 34. The host_aton() function is supposed to be passed a string that is known
1466 to be a valid IP address. However, in the case of IPv6 addresses, it was
1467 not checking this. This is a hostage to fortune. Exim now panics and dies
1468 if the condition is not met. A case was found where this could be provoked
1469 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1470 components; fortuitously, this particular loophole had already been fixed
1471 by change 4.50/55 or 4.44/33 above.
1473 If there are any other similar loopholes, the new check in host_aton()
1474 itself should stop them being exploited. The report I received stated that
1475 data on the command line could provoke the exploit when Exim was running as
1476 exim, but did not say which command line option was involved. All I could
1477 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1478 running as the user.
1481 35. There was a buffer overflow vulnerability in the SPA authentication code
1482 (which came originally from the Samba project). I have added a test to the
1483 spa_base64_to_bits() function which I hope fixes it.
1486 36. The daemon start-up calls getloadavg() while still root for those OS that
1487 need the first call to be done as root, but it missed one case: when
1488 deliver_queue_load_max is set with deliver_drop_privilege. This is
1489 necessary for the benefit of the queue runner, because there is no re-exec
1490 when deliver_drop_privilege is set.
1492 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1493 was in use, was not putting the data itself into the right store pool;
1494 consequently, it could be overwritten for a subsequent message in the same
1495 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1498 38. Sometimes the final signoff response after QUIT could fail to get
1499 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1500 < 0 before doing a fflush(). This bug looks as though it goes back to the
1501 introduction of TLS in release 3.20, but "sometimes" must have been rare
1502 because the tests only now provoked it.
1508 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1509 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1510 Now it returns an integer. A number of calls were still expecting a T/F
1511 return. Fortuitously, in all cases, the tests worked in OK situations,
1512 which is the norm. However, things would have gone wrong on any write
1513 failures on the smtp file descriptor. This function is used when sending
1514 messages over SMTP and also when doing verify callouts.
1516 2. When Exim is called to do synchronous delivery of a locally submitted
1517 message (the -odf or -odi options), it no longer closes stderr before doing
1520 3. Implemented the mua_wrapper option.
1522 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1524 5. Implemented the functions header_remove(), header_testname(),
1525 header_add_at_position(), and receive_remove_recipient(), and exported them
1528 6. If an ACL "warn" statement specified the addition of headers, Exim already
1529 inserted X-ACL-Warn: at the start if there was no header name. However, it
1530 was not making this test for the second and subsequent header lines if
1531 there were newlines in the string. This meant that an invalid header could
1532 be inserted if Exim was badly configured.
1534 7. Allow an ACL "warn" statement to add header lines at the start or after all
1535 the Received: headers, as well as at the end.
1537 8. Added the rcpt_4xx retry error code.
1539 9. Added postmaster_mailfrom=xxx to callout verification option.
1541 10. Added mailfrom=xxxx to the callout verification option, for verify=
1544 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1545 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1547 12. Inserted some casts to stop certain compilers warning when using pointer
1548 differences as field lengths or precisions in printf-type calls (mostly
1549 affecting debugging statements).
1551 13. Added optional readline() support for -be (dynamically loaded).
1553 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1554 same clock tick as a message's arrival, so that its received time was the
1555 same as the "first fail" time on the retry record, and that message
1556 remained on the queue past the ultimate address timeout, every queue runner
1557 would try a delivery (because it was past the ultimate address timeout) but
1558 after another failure, the ultimate address timeout, which should have then
1559 bounced the address, did not kick in. This was a "< instead of <=" error;
1560 in most cases the first failure would have been in the next clock tick
1561 after the received time, and all would be well.
1563 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1564 being recognized when the domain list was tested by the match_domain
1565 condition in an expansion string.
1567 16. Added the ${str2b64: operator.
1569 17. Exim was always calling setrlimit() to set a large limit for the number of
1570 processes, without checking whether the existing limit was already
1571 adequate. (It did check for the limit on file descriptors.) Furthermore,
1572 errors from getrlimit() and setrlimit() were being ignored. Now they are
1573 logged to the main and panic logs, but Exim does carry on, to try to do its
1574 job under whatever limits there are.
1576 18. Imported PCRE 5.0.
1578 19. Trivial typo in log message " temporarily refused connection" (the leading
1581 20. If the log selector return_path_on_delivery was set and an address was
1582 redirected to /dev/null, the delivery process crashed because it assumed
1583 that a return path would always be set for a "successful" delivery. In this
1584 case, the whole delivery is bypassed as an optimization, and therefore no
1587 21. Internal re-arrangement: the function for sending a challenge and reading
1588 a response while authentication was assuming a zero-terminated challenge
1589 string. It's now changed to take a pointer and a length, to allow for
1590 binary data in such strings.
1592 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1594 23. Exim was not respecting finduser_retries when seeking the login of the
1595 uid under which it was called; it was always trying 10 times. (The default
1596 setting of finduser_retries is zero.) Also, it was sleeping after the final
1597 failure, which is pointless.
1599 24. Implemented tls_on_connect_ports.
1601 25. Implemented acl_smtp_predata.
1603 26. If the domain in control=submission is set empty, Exim assumes that the
1604 authenticated id is a complete email address when it generates From: or
1605 Sender: header lines.
1607 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1608 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1609 chown and chgrp in /bin and hostname in /usr/bin.
1611 28. Exim was keeping the "process log" file open after each use, just as it
1612 does for the main log. This opens the possibility of it remaining open for
1613 long periods when the USR1 signal hits a daemon. Occasional processlog
1614 errors were reported, that could have been caused by this. Anyway, it seems
1615 much more sensible not to leave this file open at all, so that is what now
1618 29. The long-running daemon process does not normally write to the log once it
1619 has entered its main loop, and it closes the log before doing so. This is
1620 so that log files can straightforwardly be renamed and moved. However,
1621 there are a couple of unusual error situations where the daemon does write
1622 log entries, and I had neglected to close the log afterwards.
1624 30. The text of an SMTP error response that was received during a remote
1625 delivery was being truncated at 512 bytes. This is too short for some of
1626 the long messages that one sometimes sees. I've increased the limit to
1629 31. It is now possible to make retry rules that apply only when a message has a
1630 specific sender, in particular, an empty sender.
1632 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1633 it possible to be selective about when SMTP synchronization is enforced.
1635 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1637 32. Implemented hosts_connection_nolog.
1639 33. Added an ACL for QUIT.
1641 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1644 35. Added mailbox_size and mailbox_filecount to appendfile.
1646 36. Added control = no_multiline_responses to ACLs.
1648 37. There was a bug in the logic of the code that waits for the clock to tick
1649 in the case where the clock went backwards by a substantial amount such
1650 that the microsecond fraction of "now" was more than the microsecond
1651 fraction of "then" (but the whole seconds number was less).
1653 38. Added support for the libradius Radius client library this is found on
1654 FreeBSD (previously only the radiusclient library was supported).
1660 1. When certain lookups returned multiple values in the form name=value, the
1661 quoting of the values was not always being done properly. Specifically:
1662 (a) If the value started with a double quote, but contained no whitespace,
1664 (b) If the value contained whitespace other than a space character (i.e.
1665 tabs or newlines or carriage returns) it was not quoted.
1666 This fix has been applied to the mysql and pgsql lookups by writing a
1667 separate quoting function and calling it from the lookup code. The fix
1668 should probably also be applied to nisplus, ibase and oracle lookups, but
1669 since I cannot test any of those, I have not disturbed their existing code.
1671 2. A hit in the callout cache for a specific address caused a log line with no
1672 reason for rejecting RCPT. Now it says "Previous (cached) callout
1673 verification failure".
1675 3. There was an off-by-one bug in the queryprogram router. An over-long
1676 return line was truncated at 256 instead of 255 characters, thereby
1677 overflowing its buffer with the terminating zero. As well as fixing this, I
1678 have increased the buffer size to 1024 (and made a note to document this).
1680 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1681 when Exim is waiting for an SMTP response from a remote server, Exim
1682 restarts its select() call on the socket, thereby resetting its timeout.
1683 This is not a problem when such interrupts are rare. Somebody set up a cron
1684 job to run exiwhat every 2 minutes, which is less than the normal select()
1685 timeout (5 or 10 minutes). This meant that the select() timeout never
1686 kicked in because it was always reset. I have fixed this by comparing the
1687 time when an interrupt arrives with the time at the start of the first call
1688 to select(). If more time than the timeout has elapsed, the interrupt is
1689 treated as a timeout.
1691 5. Some internal re-factoring in preparation for the addition of Sieve
1692 extensions (by MH). In particular, the "personal" test is moved to a
1693 separate function, and given an option for scanning Cc: and Bcc: (which is
1694 not set for Exim filters).
1696 6. When Exim created an email address using the login of the caller as the
1697 local part (e.g. when creating a From: or Sender: header line), it was not
1698 quoting the local part when it contained special characters such as @.
1700 7. Installed new OpenBSD configuration files.
1702 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1703 try to make them clearer.
1705 9. Callout options, other than the timeout value, were being ignored when
1706 verifying sender addresses in header lines. For example, when using
1708 verify = header_sender/callout=no_cache
1710 the cache was (incorrectly) being used.
1712 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1713 only the Cygwin environment.
1715 11. When return_path_on_delivery was set as a log selector, if different remote
1716 addresses in the same message used different return paths and parallel
1717 remote delivery occurred, the wrong values would sometimes be logged.
1718 (Whenever a remote delivery process finished, the return path value from
1719 the most recently started remote delivery process was logged.)
1721 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1722 lines when AUTH and/or TLS are in use. This is the "received protocol"
1723 field. Exim used to use "asmtp" for authenticated SMTP, without any
1724 indication (in the protocol name) for TLS use. Now it follows the RFC and
1725 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1726 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1727 names appear in log lines as well as in Received: header lines.
1729 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1730 extensions, and comparison tests, and to fix some bugs.
1732 14. Changes to the "personal" filter test:
1734 (1) The test was buggy in that it was just doing the equivalent of
1735 "contains" tests on header lines. For example, if a user's address was
1736 anne@some.where, the "personal" test would incorrectly be true for
1738 To: susanne@some.where
1740 This test is now done by extracting each address from the header in turn,
1741 and checking the entire address. Other tests that are part of "personal"
1742 are now done using regular expressions (for example, to check local parts
1743 of addresses in From: header lines).
1745 (2) The list of non-personal local parts in From: addresses has been
1746 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1747 taken from the Sieve specification recommendations.
1749 (3) If the message contains any header line starting with "List-" it is
1750 treated as non-personal.
1752 (4) The test for "circular" in the Subject: header line has been removed
1753 because it now seems ill-conceived.
1755 15. Minor typos in src/EDITME comments corrected.
1757 16. Installed latest exipick from John Jetmore.
1759 17. If headers_add on a router specified a text string that was too long for
1760 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1761 of string_sprintf() is now avoided.
1763 18. $message_body_size was not set (it was always zero) when running the DATA
1764 ACL and the local_scan() function.
1766 19. For the "mail" command in an Exim filter, no default was being set for
1767 the once_repeat time, causing a random time value to be used if "once" was
1768 specified. (If the value happened to be <= 0, no repeat happened.) The
1769 default is now 0s, meaning "never repeat". The "vacation" command was OK
1770 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1771 (I found it when inspecting the code).
1773 20. There is now an overall timeout for performing a callout verification. It
1774 defaults to 4 times the callout timeout, which applies to individual SMTP
1775 commands during the callout. The overall timeout applies when there is more
1776 than one host that can be tried. The timeout is checked before trying the
1777 next host. This prevents very long delays if there are a large number of
1778 hosts and all are timing out (e.g. when the network connections are timing
1779 out). The value of the overall timeout can be changed by specifying an
1780 additional sub-option for "callout", called "maxwait". For example:
1782 verify = sender/callout=5s,maxwait=20s
1784 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1785 to the end before writing, but this should make it even safer).
1787 22. Exim was forgetting that it had advertised PIPELINING for the second and
1788 subsequent messages on an SMTP connection. It was also not resetting its
1789 memory on STARTTLS and an internal HELO.
1791 23. When Exim logs an SMTP synchronization error within a session, it now
1792 records whether PIPELINING has been advertised or not.
1794 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1795 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1796 rather than long int.
1798 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1800 26. Added the never_mail option to autoreply.
1806 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1807 crash if the getsockname() call failed; this can happen if a connection is
1808 closed very soon after it is established. The problem was simply in the
1809 order in which certain operations were done, causing Exim to try to write
1810 to the SMTP stream before it had set up the file descriptor. The bug has
1811 been fixed by making things happen in the correct order.
1817 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1818 before the connection was closed, thus losing the rejection response.
1820 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1821 some early Solaris releases, but causes trouble in current releases where
1822 socklen_t is defined.
1824 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1827 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1830 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1831 that releasing the top of it at the end releases what was used for sub-
1832 expansions (unless the block got too big). However, discard this block if
1833 the first thing is a variable or header, so that we can use its block when
1834 it is dynamic (useful for very large $message_headers, for example).
1836 6. Lookups now cache *every* query, not just the most recent. A new, separate
1837 store pool is used for this. It can be recovered when all lookup caches are
1838 flushed. Lookups now release memory at the end of their result strings.
1839 This has involved some general refactoring of the lookup sources.
1841 7. Some code has been added to the store_xxx() functions to reduce the amount
1842 of flapping under certain conditions.
1844 8. log_incoming_interface used to affect only the <= reception log lines. Now
1845 it causes the local interface and port to be added to several more SMTP log
1846 lines, for example "SMTP connection from", and rejection lines.
1848 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1850 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1852 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1853 could be overwritten at the end of the current message (or the start of a
1854 new message if it was set in a HELO ACL). The value is now preserved for
1855 the duration of the SMTP connection.
1857 12. If a transport had a headers_rewrite setting, and a matching header line
1858 contained an unqualified address, that address was qualified, even if it
1859 did not match any rewriting rules. The underlying bug was that the values
1860 of the flags that permit the existence of unqualified sender and recipient
1861 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1862 non-local messages, and by -bnq for local messages) were not being
1863 preserved with the message after it was received.
1865 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1866 "next input=" as part of the text comprising the host identity instead of
1867 the correct text. The code was using the same buffer for two different
1868 strings. However, depending on which order the printing function evaluated
1869 its arguments, the bug did not always show up. Under Linux, for example, my
1870 test suite worked just fine.
1872 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1873 doesn't work with some older versions of Perl. It has been changed to "my",
1874 which in any case is probably the better facility to use.
1876 15. A really picky compiler found some instances of statements for creating
1877 error messages that either had too many or two few arguments for the format
1880 16. The size of the buffer for calls to the DNS resolver has been increased
1881 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1882 for addresses that have a lot of PTR records. This alleviates a problem; it
1883 does not fully solve it.
1885 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1886 buffer now truncates the list and logs the incident, which is the same
1887 action as happens when Exim is looking up a host name and its aliases.
1888 Previously in this situation something unpredictable would happen;
1889 sometimes it was "internal error: store_reset failed".
1891 18. If a server dropped the connection unexpectedly when an Exim client was
1892 using GnuTLS and trying to read a response, the client delivery process
1893 crashed while trying to generate an error log message.
1895 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1896 string, for example:
1898 warn message = H1: something\nH2: something
1900 the text was added as a single header line from Exim's point of view
1901 though it ended up OK in the delivered message. However, searching for the
1902 second and subsequent header lines using $h_h2: did not work. This has been
1903 fixed. Similarly, if a system filter added multiple headers in this way,
1904 the routers could not see them.
1906 20. Expanded the error message when iplsearch is called with an invalid key to
1907 suggest using net-iplsearch in a host list.
1909 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1910 ACLs are no longer actually imposed (and a message to that effect is
1913 22. If a "gecos" field in a passwd entry contained escaped characters, in
1914 particular, if it contained a \" sequence, Exim got it wrong when building
1915 a From: or a Sender: header from that name. A second bug also caused
1916 incorrect handling when an unquoted " was present following a character
1917 that needed quoting.
1919 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1920 was not being matched caselessly.
1922 24. Arranged for all hyphens in the exim.8 source to be escaped with
1925 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1926 properly. Recipient callout cache records were still being keyed to include
1927 the sender, even when use_sender was set false. This led to far more
1928 callouts that were necessary. The sender is no longer included in the key
1929 when use_sender is false.
1931 26. Added "control = submission" modifier to ACLs.
1933 27. Added the ${base62d: operator to decode base 62 numbers.
1935 28. dnsdb lookups can now access SRV records.
1937 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1938 the configuration file.
1940 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1941 (-v) mode. This makes the output for a verbose queue run more intelligible.
1943 31. Added a use_postmaster feature to recipient callouts.
1945 32. Added the $body_zerocount variable, containing the number of binary zero
1946 bytes in the message body.
1948 33. The time of last modification of the "new" subdirectory is now used as the
1949 "mailbox time last read" when there is a quota error for a maildir
1952 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1954 35. Added +ignore_unknown as a special item in host lists.
1956 36. Code for decoding IPv6 addresses in host lists is now included, even if
1957 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1958 address was recognized as an IP address, but was then not correctly decoded
1959 into binary, causing unexpected and incorrect effects when compared with
1966 1. Very minor rewording of debugging text in manualroute to say "list of
1967 hosts" instead of "hostlist".
1969 2. If verify=header_syntax was set, and a header line with an unqualified
1970 address (no domain) and a large number of spaces between the end of the
1971 name and the colon was received, the reception process suffered a buffer
1972 overflow, and (when I tested it) crashed. This was caused by some obsolete
1973 code that should have been removed. The fix is to remove it!
1975 3. When running in the test harness, delay a bit after writing a bounce
1976 message to get a bit more predictability in the log output.
1978 4. Added a call to search_tidyup() just before forking a reception process. In
1979 theory, someone could use a lookup in the expansion of smtp_accept_max_
1980 per_host which, without the tidyup, could leave open a database connection.
1982 5. Added the variables $recipient_data and $sender_data which get set from a
1983 lookup success in an ACL "recipients" or "senders" condition, or a router
1984 "senders" option, similar to $domain_data and $local_part_data.
1986 6. Moved the writing of debug_print from before to after the "senders" test
1989 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1990 problems for message scanning, either using a data ACL, or using
1991 local_scan() because the Received: header was not generated till after they
1992 were called (in order to set the time as the time of reception completion).
1993 I have revised the way this works. The header is now generated after the
1994 body is received, but before the ACL or local_scan() are called. After they
1995 are run, the timestamp in the header is updated.
2001 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2002 before starting a queue runner without re-exec. This happened only when
2003 deliver_drop_privilege was set or when the Exim user was set to root. The
2004 effect of the bug was that timeouts during subsequent deliveries caused
2005 crashes instead of being properly handled. The handler is now left at its
2006 default (and expected) setting.
2008 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2009 message, again when deliver_drop_privilege is set or Exim is run as root.
2010 The bug described in (1) was not present in this case, but the tidying up
2011 of the other signals was missing. I have made the two cases consistent.
2013 3. The ignore_target_hosts setting on a manualroute router was being ignored
2014 for hosts that were looked up using the /MX notation.
2016 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2019 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2020 operated on the sender address. After changing the $sender_address to <>
2021 for the sender address verify, Exim was re-instated it as the original
2022 (before rewriting) address, but remembering that it had rewritten it, so it
2023 wasn't rewriting it again. This bug also had the effect of breaking the
2024 sender address verification caching when the sender address was rewritten.
2026 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2027 This has been changed so that if the ip literal address matches
2028 ignore_target_hosts, the router declines.
2030 7. Added expansion conditions match_domain, match_address, and match_local_
2031 part (NOT match_host).
2033 8. The placeholder for the Received: header didn't have a length field set.
2035 9. Added code to Exim itself and to exim_lock to test for a specific race
2036 condition that could lead to file corruption when using MBX delivery. The
2037 issue is with the lockfile that is created in /tmp. If this file is removed
2038 after a process has opened it but before that process has acquired a lock,
2039 there is the potential for a second process to recreate the file and also
2040 acquire a lock. This could lead to two Exim processes writing to the file
2041 at the same time. The added code performs the same test as UW imapd; it
2042 checks after acquiring the lock that its file descriptor still refers to
2043 the same named file.
2045 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2046 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2049 11. Added the smtp_active_hostname option. If used, this will typically be made
2050 to depend on the incoming interface address. Because $interface_address is
2051 not set up until the daemon has forked a reception process, error responses
2052 that can happen earlier (such as "too many connections") no longer contain
2055 12. If an expansion in a condition on a "warn" statement fails because a lookup
2056 defers, the "warn" statement is abandoned, and the next ACL statement is
2057 processed. Previously this caused the whole ACL to be aborted.
2059 13. Added the iplsearch lookup type.
2061 14. Added ident_timeout as a log selector.
2063 15. Added tls_certificate_verified as a log selector.
2065 16. Added a global option tls_require_ciphers (compare the smtp transport
2066 option of the same name). This controls incoming TLS connections.
2068 17. I finally figured out how to make tls_require_ciphers do a similar thing
2069 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2070 before starting the TLS session.
2072 18. Tabs are now shown as \t in -bP output.
2074 19. If the log selector return_path_on_delivery was set, Exim crashed when
2075 bouncing a message because it had too many Received: header lines.
2077 20. If two routers both had headers_remove settings, and the first one included
2078 a superfluous trailing colon, the final name in the first list and the
2079 first name in the second list were incorrectly joined into one item (with a
2080 colon in the middle).
2086 1. Added -C and -D options to the exinext utility, mainly to make it easier
2087 to include in the automated testing, but these could be helpful when
2088 multiple configurations are in use.
2090 2. The exinext utility was not formatting the output nicely when there was
2091 an alternate port involved in the retry record key, nor when there was a
2092 message id as well (for retries that were specific to a specific message
2093 and a specific host). It was also confused by IPv6 addresses, because of
2094 the additional colons they contain. I have fixed the IPv4 problem, and
2095 patched it up to do a reasonable job for IPv6.
2097 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2098 delivery, the log line now contains "pipelined" if PIPELINING was used.
2100 4. An SMTP transport process used to panic and die if the bind() call to set
2101 an explicit outgoing interface failed. This has been changed; it is now
2102 treated in the same way as a connect() failure.
2104 5. A reference to $sender_host_name in the part of a conditional expansion
2105 that was being skipped was still causing a DNS lookup. This no longer
2108 6. The def: expansion condition was not recognizing references to header lines
2109 that used bh_ and bheader_.
2111 7. Added the _cache feature to named lists.
2113 8. The code for checking quota_filecount in the appendfile transport was
2114 allowing one more file than it should have been.
2116 9. For compatibility with Sendmail, the command line option
2124 and sets the incoming protocol and host name (for trusted callers). The
2125 host name and its colon can be omitted when only the protocol is to be set.
2126 Note the Exim already has two private options, -pd and -ps, that refer to
2127 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2128 "s", but I don't think that's a major issue.
2130 10. A number of refactoring changes to the code, none of which should affect
2133 (a) The number of logging options was getting close to filling up the
2134 32-bit word that was used as a bit map. I have split them into two classes:
2135 those that are passed in the argument to log_write(), and those that are
2136 only ever tested independently outside of that function. These are now in
2137 separate 32-bit words, so there is plenty of room for expansion again.
2138 There is no change in the user interface or the logging behaviour.
2140 (b) When building, for example, log lines, the code previously used a
2141 macro that called string_cat() twice, in order to add two strings. This is
2142 not really sufficiently general. Furthermore, there was one instance where
2143 it was actually wrong because one of the argument was used twice, and in
2144 one call a function was used. (As it happened, calling the function twice
2145 did not affect the overall behaviour.) The macro has been replaced by a
2146 function that can join an arbitrary number of extra strings onto a growing
2149 (c) The code for expansion conditions now uses a table and a binary chop
2150 instead of a serial search (which was left over from when there were very
2151 few conditions). Also, it now recognizes conditions like "pam" even when
2152 the relevant support is not compiled in: a suitably worded error message is
2153 given if an attempt is made to use such a condition.
2155 11. Added ${time_interval:xxxxx}.
2157 12. A bug was causing one of the ddress fields not to be passed back correctly
2158 from remote delivery subprocesses. The field in question was not being
2159 subsequently used, so this caused to problems in practice.
2161 13. Added new log selectors queue_time and deliver_time.
2163 14. Might have fixed a bug in maildirsizefile handling that threw up
2164 "unexpected character" debug warnings, and recalculated the data
2165 unnecessarily. In any case, I expanded the warning message to give more
2168 15. Added the message "Restricted characters in address" to the statements in
2169 the default ACL that block characters like @ and % in local parts.
2171 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2172 Three changes have been made:
2174 (a) There was a serious bug; a negative response to MAIL caused the whole
2175 recipient domain to be cached as invalid, thereby blocking all messages
2176 to all local parts at the same domain, from all senders. This bug has
2177 been fixed. The domain is no longer cached after a negative response to
2178 MAIL if the sender used is not empty.
2180 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2183 (c) A new callout option, "use_sender" has been added for people who want
2184 the modified behaviour.
2190 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2193 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2194 indeed breaks things for older releases.
2196 3. Added additional logging to the case where there is a problem reading data
2197 from a filter that is running in a subprocess using a pipe, in order to
2198 try to track down a specific problem.
2200 4. Testing facility fudge: when running in the test harness and attempting
2201 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2202 getting "No route to host". Convert this to a timeout.
2204 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2207 6. Some OS don't have socklen_t but use size_t instead. This affects the
2208 fifth argument of getsockopt() amongst other things. This is now
2209 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2210 can be set for individual OS. I have set it for SunOS5, OSF1, and
2211 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2212 some earlier ones do not.
2214 7. Change 4.30/15 was not doing the test caselessly.
2216 8. The standard form for an IPv6 address literal was being rejected by address
2217 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2218 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2219 this, as well as the form without the "IPv6" on the front (but only when
2220 address literals are enabled, of course).
2222 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2224 10. Exim crashed if a message with an empty sender address specified by -f
2225 encountered a router with an errors_to setting. This could be provoked only
2226 by a command such as
2230 where an empty string was supplied; "<>" did not hit this bug.
2232 11. Installed PCRE release 4.5.
2234 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2235 remained set. It is now erased.
2237 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2238 times from message ids (which are base 36 rather than the normal 62).
2240 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2241 were being counted as actual protocol errors, and logged if the log
2242 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2243 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2244 and DATA following a set of rejected RCPTs do not count as protocol errors.
2245 In other words, Exim assumes they were pipelined, though this may not
2246 actually be the case. Of course, in all cases the client gets an
2247 appropriate error code.
2249 15. If a lookup fails in an ACL condition, a message about the failure may
2250 be available; it is used if testing the ACL cannot continue, because most
2251 such messages specify what the cause of the deferral is. However, some
2252 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2253 that caused an old message to be retained and used if a later statement
2254 caused a defer, replacing the real cause of the deferral.
2256 16. If an IP address had so many PTR records that the DNS lookup buffer
2257 was not large enough to hold them, Exim could crash while trying to process
2258 the truncated data. It now detects and logs this case.
2260 17. Further to 4.21/58, another change has been made: if (and only if) the
2261 first line of a message (the first header line) ends with CRLF, a bare LF
2262 in a subsequent header line has a space inserted after it, so as not to
2263 terminate the header.
2265 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2266 unnecessarily, used atoi() instead of strtol(), and didn't check the
2267 termination when getting file sizes from file names by regex.
2269 19. Completely re-implemented the support for maildirsize files, in the light
2270 of a number of problems with the previous contributed implementation
2271 (4.30/29). In particular:
2273 . If the quota is zero, the maildirsize file is maintained, but no quota is
2276 . If the maildir directory does not exist, it is created before any attempt
2277 to write a maildirsize file.
2279 . The quota value in the file is just a cache; if the quota is changed in
2280 the transport, the new value overrides.
2282 . A regular expression is available for excluding directories from the
2285 20. The autoreply transport checks the characters in options that define the
2286 message's headers; it allows continued headers, but it was checking with
2287 isspace() after an embedded newline instead of explicitly looking for a
2290 21. If all the "regular" hosts to which an address was routed had passed their
2291 expiry times, and had not reached their retry times, the address was
2292 bounced, even if fallback hosts were defined. Now Exim should go on to try
2295 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2296 equivalent code in the SMTP transport. Some hosts send humungous responses
2297 to HELO/EHLO, more than 1024 it seems.
2299 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2300 gives compiler warnings in some environments. I've now done it "properly",
2303 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2304 (because of problems with the built-in one) was declared to return uschar *
2305 instead of char *, causing compiler failure.
2307 25. Fixed a file descriptor leak when processing alias/forward files.
2309 26. Fixed a minor format string issue in dbfn.c.
2311 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2313 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2314 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2317 29. When a sender address is verified, it is cached, to save repeating the test
2318 when there is more than one recipient in a message. However, when the
2319 verification involves a callout, it is possible for different callout
2320 options to be set for different recipients. It is too complicated to keep
2321 track of this in the cache, so now Exim always runs a verification when a
2322 callout is required, relying on the callout cache for the optimization.
2323 The overhead is duplication of the address routing, but this should not be
2326 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2327 to be verified with callout=postmaster, and the main callout worked but the
2328 postmaster check failed, the verification correctly failed. However, if a
2329 subsequent RCPT command asked for sender verification *without* the
2330 postmaster check, incorrect caching caused this verification also to fail,
2333 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2334 it was not caching the DNS options (qualify_single, search_parents) that
2335 were used when the lookup failed. A subsequent lookup with different
2336 options therefore always gave the same answer, though there were cases
2337 where it should not have. (Example: a "domains = !$mx_any" option on a
2338 dnslookup router: the "domains" option is always processed without any
2339 widening, but the router might have qualify_single set.) Now Exim uses the
2340 cached value only when the same options are set.
2342 32. Added John Jetmore's "exipick" utility to the distribution.
2344 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2345 than a timeout (e.g. a certificate is required, and is not provided), an
2346 Exim server now closes the connection immediately. Previously it waited for
2347 the client to close - but if the client is SSL, it seems that they each
2348 wait for each other, leading to a delay before one of them times out.
2350 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2351 maintained 0.8.x compatibility because I don't think many are using it, and
2352 it is clearly obsolete.
2354 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2357 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2358 client certificate was expired. A simple patch fixes this, though I don't
2359 understand the full logic of why the verify callback is called multiple
2362 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2363 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2364 which causes problems with some clients (such as the Certicom SSL Plus
2365 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2366 disables the coutermeasure allowing Eudora to connect."
2368 38. Exim was not checking that a write() to a log file succeeded. This could
2369 lead to Bad Things if a log got too big, in particular if it hit a file
2370 size limit. Exim now panics and dies if it cannot write to a log file, just
2371 as it does if it cannot open a log file.
2373 39. Modified OS/Makefile-Linux so that it now contains
2375 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2377 The two -D definitions ensure that Exim is compiled with large file
2378 support, which makes it possible to handle log files that are bigger than
2381 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2382 instance) a domain was checked against a named list that involved a lookup,
2383 causing $domain_data to be set, then another domain was checked against the
2384 same list, then the first domain was re-checked, the value of $domain_data
2385 after the final check could be wrong. In particular, if the second check
2386 failed, it could be set empty. This bug probably also applied to
2389 41. The strip_trailing_dot option was not being applied to the address given
2390 with the -f command-line option.
2392 42. The code for reading a message's header from the spool was incrementing
2393 $received_count, but never initializing it. This meant that the value was
2394 incorrect (doubled) while delivering a message in the same process in which
2395 it was received. In the most common configuration of Exim, this never
2396 happens - a fresh exec is done - but it can happen when
2397 deliver_drop_privilege is set.
2399 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2400 it now includes up to 150 characters of the unexpected data in the log
2403 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2404 and building data strings. The size of both of these buffers was 10 000
2405 bytes - far larger than anybody would *ever* want, thought I. Needless to
2406 say, somebody hit the limit. I have increased the maximum line length to
2407 20 000 and the maximum data length of concatenated lines to 100 000. I have
2408 also fixed two bugs, because there was no checking on these buffers. Tsk,
2409 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2410 buffer is too small.
2412 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2413 lsearch lookups. Now it does.
2415 46. When parsing a route_list item in a manualroute router, a fixed-length
2416 buffer was used for the list of hosts. I made this 1024 bytes long,
2417 thinking that nobody would ever have a list of hosts that long. Wrong.
2418 Somebody had a whole pile of complicated expansion conditions, and the
2419 string was silently truncated, leading to an expansion error. It turns out
2420 that it is easier to change to an unlimited length (owing to other changes
2421 that have happened since this code was originally written) than to build
2422 structure for giving a limitation error. The length of the item that
2423 expands into the list of hosts is now unlimited.
2425 47. The lsearch lookup could not handle data where the length of text line was
2426 more than 4095 characters. Such lines were truncated, leading to shortened
2427 data being returned. It should now handle lines of any length.
2429 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2430 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2433 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2434 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2435 no longer happens in comment lines. A list of replacements is now placed
2436 at the head of all of the source files, except those whose only change is
2437 to replace PERL_COMMAND in the very first #! line.
2439 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2440 messages on the queue, with a bottom-up merge sort, using code contributed
2441 by Michael Haardt. This should make operations like -bp somewhat faster on
2442 large queues. It won't affect queue runners, except when queue_run_in_order
2445 51. Installed eximstats 1.31 in the distribution.
2447 52. Added support for SRV lookups to the dnslookup router.
2449 53. If an ACL referred to $message_body or $message_body_end, the value was not
2450 reset for any messages that followed in the same SMTP session.
2452 54. The store-handling optimization for building very long strings was not
2453 differentiating between the different store pools. I don't think this
2454 actually made any difference in practice, but I've tidied it.
2456 55. While running the routers to verify a sender address, $sender_address
2457 was still set to the sender address. This is wrong, because when routing to
2458 send a bounce to the sender, it would be empty. Therefore, I have changed
2459 it so that, while verifying a sender address, $sender_address is set to <>.
2460 (There is no change to what happens when verifying a recipient address.)
2462 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2463 target A or AAAA records (if not already returned) without resetting the
2464 qualify_single or search_parents options of the DNS resolver. These are
2465 inappropriate in this case because the targets of MX and SRV records must
2466 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2467 target that, when qualified, matched something in the local domain. These
2468 two options are now turned off when doing these lookups.
2470 57. It seems that at least some releases of Reiserfs (which does not have the
2471 concept of a fixed number of inodes) returns zero and not -1 for the
2472 number of available inodes. This interacted badly with check_spool_inodes,
2473 which assumed that -1 was the "no such thing" setting. What I have done is
2474 to check that the total number of inodes is greater than zero before doing
2475 the test of how many are available.
2477 58. When a "warn" ACL statement has a log_message modifier, the message is
2478 remembered, and not repeated. This is to avoid a lot of repetition when a
2479 message has many recipients that cause the same warning to be written.
2480 Howewer, Exim was preserving the list of already written lines for an
2481 entire SMTP session, which doesn't seem right. The memory is now reset if a
2482 new message is started.
2484 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2485 debugging output unless log_rewrite was also set.
2487 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2488 of (int)(handle) when we know that handle contains (void *)(-1).
2490 61. The Exim daemon panic-logs an error return when it closes the incoming
2491 connection. However "connection reset by peer" seems to be common, and
2492 isn't really an error worthy of noting specially, so that particular error
2495 62. When Exim is trying to find all the local interfaces, it used to panic and
2496 die if the ioctl to get the interface flags failed. However, it seems that
2497 on at least one OS (Solaris 9) it is possible to have an interface that is
2498 included in the list of interfaces, but for which you get a failure error
2499 for this call. This happens when the interface is not "plumbed" into a
2500 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2501 failure of the "get flags" call assumes that the interface is down.
2503 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2504 makes life easier for people who are doing arithmetic on fields extracted
2505 from dates, where you often get leading zeros that should not be
2506 interpreted as octal.
2508 64. Added qualify_domain to the redirect router, to override the global
2511 65. If a pathologically long header line contained very many addresses (the
2512 report of this problem mentioned 10 000) and each of them was rewritten,
2513 Exim could use up a very large amount of memory. (It kept on making new
2514 copies of the header line as it rewrote, and never released the old ones.)
2515 At the expense of a bit more processing, the header rewriting function has
2516 been changed so that it no longer eats memory in this way.
2518 66. The generation of the Received: header has been moved from the time that a
2519 message starts to be received, to the time that it finishes. The timestamp
2520 in the Received: header should now be very close to that of the <= log
2521 line. There are two side-effects of this change:
2523 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2524 logged header lines no longer include the local Received: line, because
2525 it has not yet been created. The same applies to a copy of the message
2526 that is returned to a non-SMTP sender when a message is rejected.
2528 (b) When a filter file is tested using -bf, no additional Received: header
2529 is added to the test message. After some thought, I decided that this
2532 This change does not affect the value of $received_for. It is still set
2533 after address rewriting, but before local_scan() is called.
2535 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2537 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2538 gave an unhelpful panic error message, and a defer error. I have managed to
2539 change this behaviour so that it now rejects any supplied certificate,
2540 which seems right, as the list of acceptable certificates is empty.
2542 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2543 gave an unhelpful defer error. I have not managed to make this reject any
2544 supplied certificates, but the error message it gives is "no certificate
2545 supplied", which is not helpful.
2547 70. exigrep's output now also includes lines that are not associated with any
2548 message, but which match the given pattern. Implemented by a patch from
2549 Martin Sluka, which also tidied up the Perl a bit.
2551 71. Recipient callout verification, like sender verification, was using <> in
2552 the MAIL FROM command. This isn't really the right thing, since the actual
2553 sender may affect whether the remote host accepts the recipient or not. I
2554 have changed it to use the actual sender in the callout; this means that
2555 the cache record is now keyed on a recipient/sender pair, not just the
2556 recipient address. There doesn't seem to be a real danger of callout loops,
2557 since a callout by the remote host to check the sender would use <>.
2558 [SEE ABOVE: changed after hitting problems.]
2560 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2561 temporary errors. However, in the case of such a code being given after
2562 the end of a data transmission (i.e. after ".") Exim was failing to write
2563 a retry record for the message. (Yes, there was some broken host that was
2564 actually sending 8xx at this point.)
2566 73. An unknown lookup type in a host list could cause Exim to panic-die when
2567 the list was checked. (An example that provoked this was putting <; in the
2568 middle of a list instead of at the start.) If this happened during a DATA
2569 ACL check, a -D file could be left lying around. This kind of configuration
2570 error no longer causes Exim to die; instead it causes a defer errror. The
2571 incident is still logged to the main and panic logs.
2573 74. Buglet left over from Exim 3 conversion. The message "too many messages
2574 in one connection" was written to the rejectlog but not the mainlog, except
2575 when address rewriting (yes!) was being logged.
2577 75. Added write_rejectlog option.
2579 76. When a system filter was run not as root (that is, when system_filter_user
2580 was set), the values of the $n variables were not being returned to the
2581 main process; thus, they were not subsequently available in the $sn
2584 77. Added +return_path_on_delivery log selector.
2586 78. A connection timeout was being treated differently from recipients deferred
2587 when testing hosts_max_try with a message that was older than the host's
2588 retry timeout. (The host should not be counted, thus allowing all hosts to
2589 be tried at least once before bouncing.) This may have been the cause of an
2590 occasionally reported bug whereby a message would remain on the queue
2591 longer than the retry timeout, but would be bounced if a delivery was
2592 forced. I say "may" because I never totally pinned down the problem;
2593 setting up timeout/retry tests is difficult. See also the next item.
2595 79. The ultimate address timeout was not being applied to errors that involved
2596 a combination of host plus message (for example, a timeout on a MAIL
2597 command). When an address resolved to a number of possible hosts, and they
2598 were not all tried for each delivery (e.g. because of hosts_max_try), a
2599 message could remain on the queue longer than the retry timeout.
2601 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2604 81. Fixed an obscure SMTP outgoing bug which required at least the following
2605 conditions: (a) there was another message waiting for the same server;
2606 (b) the server returned 5xx to all RCPT commands in the first message so
2607 that the message was not completed; (c) the server dropped the connection
2608 or gave a negative response to the RSET that Exim sends to abort the
2609 transaction. The observed case was a dropped connection after DATA that had
2610 been sent in pipelining mode. That is, the server had advertised PIPELINING
2611 but was not implementing it correctly. The effect of the bug was incorrect
2612 behaviour, such as trying another host, and this could lead to a crash.
2618 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2619 and daemon.c were passed as pointers to ints; they should have been
2620 pointers to socklen_t variables (which are typically unsigned ints).
2622 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2625 3. Fixed a really odd bug that affected only the testing scheme; patching a
2626 certain fixed string in the binary changed the value of another string that
2627 happened to be identical to the end of the original first string.
2629 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2630 name", it returns that address as the IP address. On some operating
2631 systems (e.g. Solaris), it also passes back the IP address string as the
2632 "host name". However, on others (e.g. Linux), it passes back an empty
2633 string. Exim wasn't checking for this, and was changing the host name to an
2634 empty string, assuming it had been canonicized.
2636 5. Although rare, it is permitted to have more than one PTR record for a given
2637 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2638 all the names associated with an address, because they do in Solaris.
2639 However, it seems that they do not in Linux for data that comes from the
2640 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2641 I found this out when I moved to a new Linux workstation and tried to run
2642 the Exim test suite.
2644 To get round this problem I have changed the code so that it now does its
2645 own call to the DNS to look up PTR records when searching for a host name.
2646 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2647 addresses that are only in /etc/hosts are still found.
2649 This behaviour is, however, controlled by an option called host_lookup_
2650 order, which defaults to "bydns:byaddr". If people want to use the other
2651 order, or indeed, just use one or the other means of lookup, they can
2652 specify it in this variable.
2654 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2655 some operating systems, this comes back from gethostbyaddr() as an empty
2656 string, and this is what Exim used to test for. However, it seems that in
2657 other systems, "." is yielded. Exim now tests for this case too.
2659 7. The values of check_spool_space and check_log_space are now held internally
2660 as a number of kilobytes instead of an absolute number of bytes. If a
2661 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2662 kilobyte. This means that much larger values can be stored.
2664 8. Exim monitor: an attempt to get the action menu when not actually pointing
2665 at a message produces an empty menu entitled "No message selected". This
2666 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2667 no entries in it ("Shell widget menu has zero width and/or height"). So I
2668 have added a single, blank menu entry in this case.
2670 9. Added ${quote_local_part.
2672 10. MIME decoding is now applied to the contents of Subject: header lines when
2675 11. Now that a reference to $sender_host_address automatically causes a reverse
2676 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2677 host lookup before query-style lookups in lists that might use this
2678 variable. This has therefore been abolished, and the "net-" prefix is no
2679 longer necessary for query-style lookups.
2681 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2682 have been a typo for LFLAGS, so it has been changed.
2684 13. The install script calls Exim with "-C /dev/null" in order to find the
2685 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2686 to be output. Howeve, since Exim outputs its version number before the
2687 error, it didn't break the script. It just looked ugly. I fixed this by
2688 always allowing "-C /dev/null" if the caller is root.
2690 14. Ignore overlarge ACL variable number when reading spool file - insurance
2691 against a later release with more variables having written the file.
2693 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2694 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2695 this, as well as the form without the "IPv6" on the front.
2697 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2698 OS/Makefile-Darwin file.
2700 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2701 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2703 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2704 variable. However, some OS use ints for inodes. I've added cast to long int
2705 to get rid of the compiler warning.
2707 19. I had forgotten to lock out "/../" in configuration file names when
2708 ALT_CONFIG_PREFIX was set.
2710 20. Routers used for verification do not need to specify transports. However,
2711 if such a router generated a host list, and callout was configured, Exim
2712 crashed, because it could not find a port number from the (non-existent)
2713 transport. It now assumes port 25 in this circumstance.
2715 21. Added the -t option to exigrep.
2717 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2718 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2719 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2720 them is set, LOOKUP_LSEARCH is forced.
2722 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2724 24. Added sender and host information to the "rejected by local_scan()" log
2725 line; previously there was no indication of these.
2727 25. Added .include_if_exists.
2729 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2730 Linux. It turns out that not all file systems support this. Apparently some
2731 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2732 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2733 not supported on the file descriptor, is now ignored when Exim is trying to
2734 sync a directory. This applies only to Linux.
2736 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2738 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2739 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2742 29. Added support for maildirsize files from supplied patch (modified a bit).
2744 30. The use of :fail: followed by an empty string could lead Exim to respond to
2745 sender verification failures with (e.g.):
2747 550 Verification failed for <xxx>
2748 550 Sender verify failed
2750 where the first response line was missing the '-' that indicates it is not
2751 the final line of the response.
2753 31. The loop for finding the name of the user that called Exim had a hardwired
2754 limit of 10; it now uses the value of finduser_retries, which is used for
2755 all other user lookups.
2757 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2760 33. Exim was neglecting to zero errno before one call of strtol() when
2761 expanding a string and expecting an integer value. On some systems this
2762 resulted in spurious "integer overflow" errors. Also, it was casting the
2763 result into an int without checking.
2765 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2766 did not work. The code looks as if it has *never* worked, though it appears
2767 to have been documented since at least releast 1.62. I have made it work.
2769 35. The "timeout_DNS" error in retry rules, also documented since at least
2770 1.62, also never worked. As it isn't clear exactly what this means, and
2771 clearly it isn't a major issue, I have abolished the feature by treating it
2772 as "timeout", and writing a warning to the main and panic logs.
2774 36. The display of retry rules for -brt wasn't always showing the error code
2777 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2778 timeout_connect_A, timeout_connect_MX.
2780 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2781 to the empty sender.
2783 39. The daemon was not analysing the content of -oX till after it had closed
2784 stderr and disconnected from the controlling terminal. This meant that any
2785 syntax errors were only noted on the panic log, and the return code from
2786 the command was 0. By re-arranging the code a little, I've made the
2787 decoding happen first, so such errors now appear on stderr, and the return
2788 code is 1. However, the actual setting up of the sockets still happens in
2789 the disconnected process, so errors there are still only recorded on the
2792 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2793 connections (as happens on some IP stacks) was logged at start up time as
2794 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2795 it from "IPv6 and IPv4", which means that two separate sockets are being
2798 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2799 says whether AF_INET or AF_INET6 was passed as an argument.
2801 42. Exiwhat output was messed up when time zones were included in log
2804 43. Exiwhat now gives more information about the daemon's listening ports,
2805 and whether -tls-on-connect was used.
2807 44. The "port" option of the smtp transport is now expanded.
2809 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2810 silently ignored. Now an error message is written to the main and panic
2813 46. There's a new ACL modifier called "logwrite" which writes to a log file
2814 as soon as it is encountered.
2816 47. Added $local_user_uid and $local_user_gid at routing time.
2818 48. Exim crashed when trying to verify a sender address that was being
2821 49. Exim was recognizing only a space character after ".include". It now also
2822 recognizes a tab character.
2824 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2825 extracting the relevant information from the specification. The man page no
2826 longer contains scrambled data for the -d option, and I've added a section
2827 at the front about calling Exim under different names.
2829 51. Added "extra_headers" argument to the "mail" command in filter files.
2831 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2834 53. Installed eximstats 1.29.
2836 54. Added transport_filter_timeout as a generic transport option.
2838 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2839 Cc: header lines. This was required by RFC 822, but it not required by RFC
2842 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2843 incoming messages that did not have them. Now it does so only if the
2844 message originates locally, that is, if there is no associated remote host
2845 address. When Resent- header lines are present, this applies to the Resent-
2846 lines rather than the non-Resent- lines.
2848 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2849 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2851 58. Messages for configuration errors now include the name of the main
2852 configuration file - useful now that there may be more than one file in a
2853 list (.included file names were always shown).
2855 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2856 for those rare installations that do not start the daemon as root or run it
2857 setuid root. I've cut out the call to initgroups() if the daemon is not
2860 60. The Exim user and group can now be bound into the binary as text strings
2861 that are looked up at the start of Exim's processing.
2863 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2865 62. Added $mailstore_basename variable.
2867 63. Installed patch to sieve.c from Michael Haardt.
2869 64. When Exim failed to open the panic log after failing to open the main log,
2870 the original message it was trying to log was written to stderr and debug
2871 output, but if they were not available (the usual case in production), it
2872 was lost. Now it is written to syslog before the two lines that record the
2873 failures to open the logs.
2875 65. Users' Exim filters run in subprocesses under the user's uid. It is
2876 possible for a "deliver" command or an alias in a "personal" command to
2877 provoke an address rewrite. If logging of address rewriting is configured,
2878 this fails because the process is not running as root or exim. There may be
2879 a better way of dealing with this, but for the moment (because 4.30 needs
2880 to be released), I have disabled address rewrite logging when running a
2881 filter in a non-root, non-exim process.
2887 1. The buildconfig auxiliary program wasn't quoting the value set for
2888 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2889 not defined. This bug was masked in 4.22 by the effect that was fixed in
2892 2. Some messages that were rejected after a message id was allocated were
2893 shown as "incomplete" by exigrep. It no longer does this for messages that
2894 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2896 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2897 have allow_domain_literals set, the ID did not get logged in the <= line.
2898 Domain literals are now always recognized in Message-ID: header lines.
2900 4. The first argument for a ${extract expansion item is the key name or field
2901 number. Leading and trailing spaces in this item were not being ignored,
2902 causing some misleading effects.
2904 5. When deliver_drop_privilege was set, single queue runner processes started
2905 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2906 same command in the process it spins off) were not dropping privilege.
2908 6. When the daemon running as "exim" started a queue runner, it always
2909 re-executed Exim in the spun-off process. This is a waste of effort when
2910 deliver_drop_privilege is set. The new process now just calls the
2911 queue-runner function directly.
2917 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2920 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2921 ignored. Though the use of -oP was forcing the writing of a pid file, it
2922 was always written to the default place.
2924 3. If the message "no IP address found for host xxxx" is generated during
2925 incoming verification, it is now followed by identification of the incoming
2926 connection (so you can more easily find what provoked it).
2928 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2930 5. Added some features to "harden" Exim a bit more against certain attacks:
2932 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2933 be put in Local/Makefile. This is like the never_users runtime option,
2934 but it cannot be overridden. The default setting is "root".
2936 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2937 prefix string with which any file named in a -C command line option
2940 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2941 is retained for -C and -D only if the caller of Exim is root. Without
2942 it, the exim user may also use -C and -D and retain privilege.
2944 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2945 command line option is disabled.
2947 6. Macro names set by the -D option must start with an upper case letter, just
2948 like macro names defined in the configuration file.
2950 7. Added "dereference=" facility to LDAP.
2952 8. Two instances of the typo "uknown" in the source files are fixed.
2954 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2955 the Configure-Makefile script screwed up while processing it.
2957 10. Incorporated PCRE 4.4.
2959 11. The SMTP synchronization check was not operating right at the start of an
2960 SMTP session. For example, it could not catch a HELO sent before the client
2961 waited for the greeting. There is now a check for outstanding input at the
2962 point when the greeting is written. Because of the duplex, asynchronous
2963 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2964 way, but not yet received, when the check is performed.
2966 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2967 on TCP/IP sockets, because this apparently causes some broken clients to
2970 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2971 unchanged) from the Cygwin maintainer.
2973 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2974 was supported instead of testing for mbx. Effectively a typo.
2976 15. The spa authenticator server code was not checking that the input it
2977 received was valid base64.
2979 16. The debug output line for the "set" modifier in ACLs was not showing the
2980 name of the variable that was being set.
2982 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2984 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2985 lookup on its own. Something else was needed to trigger the lookup. For
2986 example, a match in host_lookup or the need for a host name in a host list.
2987 Now, if $sender_host_name is referenced and the host name has not yet been
2988 looked up, a lookup is performed. If the lookup fails, the variable remains
2989 empty, and $host_lookup_failed is set to "1".
2991 19. Added "eqi" as a case-independent comparison operator.
2993 20. The saslauthd authentication condition could segfault if neither service
2994 nor realm was specified.
2996 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2997 error message that was logged was misleading, and incoming SMTP
2998 connections were dropped. The message is now more accurate, and temporary
2999 errors are given to SMTP connections.
3001 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3002 (except RSET) with a 421 error, until QUIT is received. However, it was
3003 failing to send a response to QUIT.
3005 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3006 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3007 ACL is now run after the helo_try_verify_hosts code.
3009 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3010 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3011 case-independent, so other case variants are also recognized). Apparently
3012 some systems use these upper case variants.
3014 25. If more than two messages were waiting for the same host, and a transport
3015 filter was specified for the transport, Exim sent two messages over the
3016 same TCP/IP connection, and then failed with "socket operation on non-
3017 socket" when it tried to send the third.
3019 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3021 27. The extern definition of crypt16() in expand.c was not being excluded when
3022 the OS had its own crypt16() function.
3024 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3025 as a preferred synonym for return_size_limit, both as an option and as an
3028 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3030 30. Changed the default configuration ACL to relax the local part checking rule
3031 for addresses that are not in any local domains. For these addresses,
3032 slashes and pipe symbols are allowed within local parts, but the sequence
3033 /../ is explicitly forbidden.
3035 31. SPA server authentication was not clearing the challenge buffer before
3038 32. log_message in a "warn" ACL statement was writing to the reject log as
3039 well as to the main log, which contradicts the documentation and doesn't
3040 seem right (because no rejection is happening). So I have stopped it.
3042 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3043 However, I am unable to do any testing of this.
3045 34. Fixed an infelicity in the appendfile transport. When checking directories
3046 for a mailbox, to see if any needed to be created, it was accidentally
3047 using path names with one or more superfluous leading slashes; tracing
3048 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3050 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3051 contents are added to the log line that is written for every discarded
3052 recipient. (Previously a log_message setting was ignored.)
3054 36. The ${quote: operator now quotes the string if it is empty.
3056 37. The install script runs exim in order to find its version number. If for
3057 some reason other than non-existence or emptiness, which it checks, it
3058 could not run './exim', it was installing it with an empty version number,
3059 i.e. as "exim-". This error state is now caught, and the installation is
3062 38. An argument was missing from the function that creates an error message
3063 when Exim fails to connect to the socket for saslauthd authentication.
3064 This could cause Exim to crash, or give a corrupted message.
3066 39. Added isip, isip4, and isip6 to ${if conditions.
3068 40. The ACL variables $acl_xx are now saved with the message, and can be
3069 accessed later in routers, transports, and filters.
3071 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3072 strings in the file are not string-expanded.
3074 42. If a MAIL command specified a SIZE value that was too large to fit into an
3075 int variable, the check against message_size_limit failed. Such values are
3076 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3077 day this will have to be increased, but I don't think I want to be around
3078 when emails are that large.
3085 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3086 iconv() is not standard in FreeBSD.
3088 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3089 IPv6 enabled. The observed symptom was a segmentation fault on return from
3090 the function os_common_find_running_interfaces() in src/os.c.
3092 3. In the check_special_case() function in daemon.c I had used "errno" as an
3093 argument name, which causes warnings on some systems. This was basically a
3094 typo, since it was named "eno" in the comments!
3096 4. The code that waits for the clock to tick (at a resolution of some fraction
3097 of a second) so as to ensure message-id uniqueness was always waiting for
3098 at least one whole tick, when it could have waited for less. [This is
3099 almost certainly not relevant at current processor speeds, where it is
3100 unlikely to ever wait at all. But we try to future-proof.]
3102 5. The function that sleeps for a time interval that includes fractions of a
3103 second contained a race. It did not block SIGALRM between setting the
3104 timer, and suspending (a couple of lines later). If the interval was short
3105 and the sigsuspend() was delayed until after it had expired, the suspension
3106 never ended. On busy systems this could lead to processes getting stuck for
3109 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3110 process, before it forks any delivery processes. The open lookup caching
3111 mechanism meant that the open file or database connection was passed into
3112 the delivery process. The problem was that delivery processes always tidy
3113 up cached lookup data. This could cause a problem for the next delivery
3114 process started by the queue runner, because the external queue runner
3115 process does not know about the closure. So the next delivery process
3116 still has data in the lookup cache. In the case of a file lookup, there was
3117 no problem because closing a file descriptor in a subprocess doesn't affect
3118 the parent. However, if the lookup was caching a connection to a database,
3119 the connection was closed, and the second delivery process was likely to
3120 see errors such as "PGSQL: query failed: server closed the connection
3121 unexpectedly". The problem has been fixed by closing all cached lookups
3122 in a queue runner before running a delivery process.
3124 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3125 seem to have the "const" qualifier which it has on other OS. I've
3128 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3129 *of the same type* (client or server) with the same public name that an
3130 error should be diagnosed.
3132 9. When Exim looked up a host name for an IP address, but failed to find the
3133 original IP address when looking up the host name (a safety check), it
3134 output the message "<ip address> does not match any IP for NULL", which was
3135 confusing, to say the least. The bug was that the host name should have
3136 appeared instead of "NULL".
3138 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3139 user that is built into the binary, and the -C or -D options is used, root
3140 privilege is dropped before the configuration file is read. In addition,
3141 logging is switched to stderr instead of the normal log files. If the
3142 configuration then re-defines the Exim user, the unprivileged environment
3143 is probably not what is expected, so Exim logs a panic warning message (but
3146 However, if deliver_drop_privilege is set, the unprivileged state may well
3147 be exactly what is intended, so the warning has been cut out in that case,
3148 and Exim is allowed to try to write to its normal log files.
3154 1. smtp_return_error_details was not giving details for temporary sender
3155 or receiver verification errors.
3157 2. Diagnose a configuration error if two authenticators have the same public
3160 3. Exim used not to create the message log file for a message until the first
3161 delivery attempt. This could be confusing when incoming messages were held
3162 for policy or load reasons. The message log file is now created at the time
3163 the message is received, and an initial "Received" line is written to it.
3165 4. The automatically generated man page for command line options had a minor
3166 bug that caused no ill effects; however, a more serious problem was that
3167 the procedure for building the man page automatically didn't always
3168 operate. Consequently, release 4.20 contains an out-of-date version. This
3169 shouldn't happen again.
3171 5. When building Exim with embedded Perl support, the script that builds the
3172 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3173 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3175 6. The freeze_tell option was not being used for messages that were frozen on
3176 arrival, either by an ACL or by local_scan().
3178 7. Added the smtp_incomplete_transaction log selector.
3180 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3181 was accepting AUTH without a new EHLO.
3183 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3184 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3187 10. Logging of TCP/IP connections (when configured) now happens in the main
3188 daemon process instead of the child process, so that the TCP/IP connection
3189 count is more accurate (but it can never be perfect).
3191 11. The use of "drop" in a nested ACL was not being handled correctly in the
3192 outer ACL. Now, if condition failure induced by the nested "drop" causes
3193 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3194 or "require"), the connection is dropped.
3196 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3197 that yield "discard" can now be used with an "accept" or a "discard" verb,
3198 but an error is generated for any others (because I can't see a useful way
3199 to define what should happen).
3201 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3202 are now processed in the same way as lines in the Exim configuration file.
3203 In particular, continuation lines are supported.
3205 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3207 15. Added -ti meaning -t -i.
3209 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3210 domains, and warn by logging if others are found.
3212 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3213 returned from GIFCONF when Exim is trying to find the list of interfaces on
3214 a host. The code in os.c has been modified to copy each ifreq to an aligned
3215 structure in all cases.
3217 Also, in some cases, the returned ifreq's were being copied to a 'struct
3218 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3219 means the last couple of bytes of an IPv6 address could be chopped if the
3220 ifreq contained only a normal sockaddr (14 bytes storage).
3222 18. Named domain lists were not supported in the hosts_treat_as_local option.
3223 An entry such as +xxxx was not recognized, and was treated as a literal
3226 19. Ensure that header lines added by a DATA ACL are included in the reject log
3227 if the ACL subsequently rejects the message.
3229 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3230 MD5 (which is deprecated).
3232 21. When testing a filter file using -bf, Exim was writing a message when it
3233 took the sender from a "From " line in the message, but it was not doing so
3234 when it took $return_path from a Return-Path: header line. It now does.
3236 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3237 with a valid header line field name (a series of printing characters
3238 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3240 23. Changed "disc" in the source to "disk" to conform to the documentation and
3241 the book and for uniformity.
3243 24. Ignore Sendmail's -Ooption=value command line item.
3245 25. When execve() failed while trying to run a command in a pipe transport,
3246 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3247 could be confused with a return value of 69 from the command itself. This
3248 has been changed to 127, the value the shell returns if it is asked to run
3249 a non-existent command. The wording for the related log line suggests a
3250 non-existent command as the problem.
3252 26. If received_header_text expands to an empty string, do not add a Received:
3253 header line to the message. (Well, it adds a token one on the spool, but
3254 marks it "old" so that it doesn't get used or transmitted.)
3256 27. Installed eximstats 1.28 (addition of -nt option).
3258 28. There was no check for failure on the call to getsockname() in the daemon
3259 code. This can fail if there is a shortage of resources on the system, with
3260 ENOMEM, for example. A temporary error is now given on failure.
3262 29. Contrary to the C standard, it seems that in some environments, the
3263 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3264 program. Exim now does this explicitly; it affects the formatting of
3265 timestamps using strftime().
3267 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3268 complaints. I've now initialized all the variables, to avoid this.
3270 32. Header lines added by a system filter were not being "seen" during
3271 transport-time rewrites.
3273 33. The info_callback() function passed to OpenSSL is set up with type void
3274 (*)(SSL *, int, int), as described somewhere. However, when calling the
3275 function (actually a macro) that sets it up, the type void(*)() is
3276 expected. I've put in a cast to prevent warnings from picky compilers.
3278 34. If a DNS black list lookup found a CNAME record, but there were no A
3279 records associated with the domain it pointed at, Exim crashed.
3281 35. If a DNS black list lookup returned more than one A record, Exim ignored
3282 all but the first. It now scans all returned addresses if a particular IP
3283 value is being sought. In this situation, the contents of the
3284 $dnslist_value variable are a list of all the addresses, separated by a
3287 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3288 to accept a host name and all its aliases if the forward lookup for any of
3289 them yielded the IP address of the incoming connection. Now it accepts only
3290 those names whose forward lookup yields the correct IP address. Any other
3291 names are discarded. This closes a loophole whereby a rogue DNS
3292 administrator could create reverse DNS records to break through a
3293 wildcarded host restriction in an ACL.
3295 37. If a user filter or a system filter that ran in a subprocess used any of
3296 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3297 the wrong values were passed to the pipe command ($thisaddress had the
3298 value of $0, $0 had the value of $1, etc). This bug was introduced by
3299 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3301 38. Improved the line breaking for long SMTP error messages from ACLs.
3302 Previously, if there was no break point between 40 and 75 characters, Exim
3303 left the rest of the message alone. Two changes have been made: (a) I've
3304 reduced the minimum length to 35 characters; (b) if it can't find a break
3305 point between 35 and 75 characters, it looks ahead and uses the first one
3306 that it finds. This may give the occasional overlong line, but at least the
3307 remaining text gets split now.
3309 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3310 file descriptors might be low, and that setting 1000 would always raise it.
3311 It turns out that in some environments, the limit is already over 1000 and
3312 that lowering it causes trouble. So now Exim takes care not to decrease it.
3314 40. When delivering a message, the value of $return_path is set to $sender_
3315 address at the start of routing (routers may change the value). By an
3316 oversight, this default was not being set up when an address was tested by
3317 -bt or -bv, which affected the outcome if any router or filter referred to
3320 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3321 log only when "message" or "log_message" are set. However, if one of the
3322 conditions was an address verification, or a call to a nested ACL, the
3323 messages generated by the underlying test were being passed through. This
3324 no longer happens. The underlying message is available in $acl_verify_
3325 message for both "message" and "log_message" expansions, so it can be
3326 passed through if needed.
3328 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3329 new expansion $bh_ to give the encoded byte string without charset
3330 translation. Translation happens only if iconv() is available; HAVE_ICONV
3331 indicates this at build time. HEADERS_CHARSET gives the charset to
3332 translate to; headers_charset can change it in the configuration, and
3333 "headers charset" can change it in an individual filter file.
3335 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3336 that creates RFC 2047 encoded "words" labels them as that charset instead
3337 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3338 expansion operator; (ii) when Exim creates a From: line for a local
3339 message; (iii) when a header line is rewritten to include a "phrase" part.
3341 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3342 buggy, causing it to skip the first lines of messages whose message ID
3343 ended in 'D'. This would not have bitten before Exim release 4.14, because
3344 message IDs were unlikely to end in 'D' before then. The effect was to have
3345 incorrect size information for certain domains.
3347 45. #include "config.h" was missing at the start of the crypt16.c module. This
3348 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3351 46. If there was a timeout during a "random" callout check, Exim treated it as
3352 a failure of the random address, and carried on sending RSET and the real
3353 address. If the delay was just some slowness somewhere, the response to the
3354 original RCPT would be taken as a response to RSET and so on, causing
3355 mayhem of various kinds.
3357 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3358 when I implemented it. It didn't allow for the fact that some option values
3359 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3360 the right test for positive values.
3362 48. Domain names in DNS records are case-independent. Exim always looks them up
3363 in lower case. Some resolvers return domain names in exactly the case they
3364 appear in the zone file, that is, they may contain uppercase letters. Not
3365 all resolvers do this - some return always lower case. Exim was treating a
3366 change of case by a resolver as a change of domain, similar to a widening
3367 of a domain abbreviation. This triggered its re-routing code and so it was
3368 trying to route what was effectively the same domain again. This normally
3369 caused routing to fail (because the router wouldn't handle the domain
3370 twice). Now Exim checks for this case specially, and just changes the
3371 casing of the domain that it ultimately uses when it transmits the message
3374 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3377 50. If a filter generated a file delivery with a non-absolute name (possible if
3378 no home directory exists for the router), the forbid_file option was not
3381 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3382 the existing equality matching.
3384 52. Exim was using ints instead of ino_t variables in some places where it was
3385 dealing with inode numbers.
3387 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3388 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3389 called TMPDIR, and if it finds it is different, it changes its value.
3391 54. The smtp_printf() function is now made available to local_scan() so
3392 additional output lines can be written before returning. There is also an
3393 smtp_fflush() function to enable the detection of a dropped connection.
3394 The variables smtp_input and smtp_batched_input are exported to
3397 55. Changed the default runtime configuration: the message "Unknown user"
3398 has been removed from the ACL, and instead placed on the localuser router,
3399 using the cannot_route_message feature. This means that any verification
3400 failures that generate their own messages won't get overridden. Similarly,
3401 the "Unrouteable address" message that was in the ACL for unverifiable
3402 relay addresses has also been removed.
3404 56. Added hosts_avoid_esmtp to the smtp transport.
3406 57. The exicyclog script was not checking for the esoteric option
3407 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3408 will work only if exicyclog is run under the appropriate euid.
3410 58. Following a discussion on the list, the rules by which Exim recognises line
3411 endings on incoming messages have been changed. The -dropcr and drop_cr
3412 options are now no-ops, retained only for backwards compatibility. The
3413 following line terminators are recognized: LF CRLF CR. However, special
3414 processing applies to CR:
3416 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3417 nor a local message in the state where . is a terminator.
3419 (ii) If a bare CR is encountered in a header line, an extra space is added
3420 after the line terminator so as not to end the header. The reasoning
3421 behind this is that bare CRs in header lines are most likely either
3422 to be mistakes, or people trying to play silly games.
3424 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3425 was being incorrectly given as 18 bytes larger than it should have been.
3426 This is a VOB (very old bug).
3428 60. This may never have affected anything current, but just in case it has:
3429 When the local host is found other than at the start of a list of hosts,
3430 the local host, those with the same MX, and any that follow, are discarded.
3431 When the list in question was part of a longer list of hosts, the following
3432 hosts (not currently being processed) were also being discarded. This no
3433 longer happens. I'm not sure if this situation could ever has previously
3436 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3439 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3440 header. This is something that is recommended in a new Internet Draft, and
3441 is something that is documented as being done by Sendmail. There are two
3442 possible values. For messages generated by the autoreply transport, Exim
3445 Auto-Submitted: auto-replied
3447 whereas for all other generated messages (e.g. bounces) it adds
3449 Auto-Submitted: auto-generated
3451 63. The "personal" condition in filters now includes a test for the
3452 Auto-Submitted: header. If it contains the string "auto-" the message it
3453 not considered personal.
3455 64. Added rcpt_include_affixes as a generic transport option.
3457 65. Added queue_only_override (default true).
3459 66. Added the syslog_duplication option.
3461 67. If what should have been the first header line of a message consisted of
3462 a space followed by a colon, Exim was mis-interpreting it as a header line.
3463 It isn't of course - it is syntactically invalid and should therefore be
3464 treated as the start of the message body. The misbehaviour could have
3465 caused a number of strange effects, including loss of data in subsequent
3466 header lines, and spool format errors.
3468 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3469 client host had authenticated. This control can now be exercised by an ACL
3470 for more flexibility.
3472 69. By default, callouts do not happen when testing with -bh. There is now a
3473 variant, -bhc, which does actually run the callout code, including
3474 consulting and updating the callout cache.
3476 70. Added support for saslauthd authentication, courtesy of Alexander
3479 71. If statvfs() failed on the spool or log directories while checking their
3480 size for availability, Exim confusingly gave the error "space shortage".
3481 Furthermore, in debugging mode it crashed with a floating point exception.
3482 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3483 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3484 serious problem, Exim now writes to the main and panic logs when this
3485 happens, with details of the failure. It then refuses to accept the
3486 incoming message, giving the message "spool directory problem" or "log
3487 directory problem" with a 421 code for SMTP messages.
3489 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3490 0, 1, and 2 exist, because some OS complain for execs without them (see
3491 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3492 descriptors. However, the code omitted to check that the open succeeded,
3493 causing mysterious errors if for some reason the permissions on /dev/null
3494 got screwed. Now Exim writes a message to the main and panic logs, and
3495 bombs out if it can't open /dev/null.
3497 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3498 interact so that it is all more flexible. It is supposed to remain
3499 backwards compatible. Also added extra_local_interfaces.
3501 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3502 to bomb out with an assertion failure - to the client this appears as a
3503 connection drop. This problem occurs in the part of the code that was taken
3504 from the Samba project. Fortunately, the assertion is in a very simple
3505 function, so I have fixed this by reproducing the function inline in the
3506 one place where it is called, and arranging for authentication to fail
3507 instead of killing the process with assert().
3509 75. The SPA client code was not working when the server requested OEM rather
3510 than Unicode encoding.
3512 76. Added code to make require_files with a specific uid setting more usable in
3513 the case where statting the file as root fails - usually a non-root-mounted
3514 NFS file system. When this happens and the failure is EACCES, Exim now
3515 forks a subprocess and does the per-uid checking as the relevant uid.
3517 77. Added process_log_path.
3519 78. If log_file_path was not explicitly set, a setting of check_log_space or
3520 check_log_inodes was ignored.
3522 79. If a space check for the spool or log partitions fails, the incident is now
3523 logged. Of course, in the latter case the data may get lost...
3525 80. Added the %p formatting code to string_format() so that it can be used to
3526 print addresses in debug_print(). Adjusted all the address printing in the
3527 debugging in store.c to use %p rather than %d.
3529 81. There was a concern that a line of code in smtp_in.c could overflow a
3530 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3531 initially expressed, the concern was not well-founded, because trailing
3532 spaces are removed early. However, if the trailing spaces were followed by
3533 a NULL, they did not get removed, so the overflow was possible. Two fixes
3536 (a) I re-wrote the offending code in a cleaner fashion.
3537 (b) If an incoming SMTP command contains a NULL character, it is rejected
3540 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3541 runs initgroups(), so that if the Exim user is in any additional groups,
3542 they will be used during message reception.
3548 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob