1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.239 2005/09/22 12:02:21 fanf2 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added the "success_on_redirect" address verification option. See
10 NewStuff for rationale and an example.
12 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
14 PH/02 Patch to exigrep to allow it to work on syslog lines.
16 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
17 fread() to skip over the body file's header line, because in Cygwin the
18 header line is locked and is inaccessible.
20 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
21 co-exist for some time) to make it clear that it is the Exim ID that is
22 referenced, not the Message-ID: header line.
24 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
25 string_format() function, because snprintf() does not exist on all
28 PH/06 The use of forbid_filter_existstest now also locks out the use of the
29 ${stat: expansion item.
31 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
32 protocol synchronization error", to keep the pedants happy.
34 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
35 well as for IRIX systems, when gcc is being used. See the host.c source
38 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
40 PH/10 Named domain lists were not working if used in a queue_smtp_domains
43 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
44 transport and to the smtp transport in LMTP mode.
46 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
48 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
49 run a filter in a subprocess. This could lead to confusion in subsequent
50 lookups in the parent process. There should also be a search_tidyup() at
51 the end of the subprocess.
53 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
54 only if the host matched helo_try_verify_hosts, which caused the
55 verification to occur when the EHLO/HELO command was issued. The ACL just
56 tested the remembered result. Now, if a previous verification attempt has
57 not happened, "verify = helo" does it there and then.
59 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
61 TK/03 Fix log output including CR from clamd.
63 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
64 exist provoked a memory error which could cause a segfault.
66 PH/15 Installed PCRE 6.2
68 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
70 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
71 of the problem. Specifically, suggested +O2 rather than +O1 for the
74 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
76 PH/20 If a delivery was routed to a non-standard port by means of an SRV
77 record, the port was not correctly logged when the outgoing_port log
78 selector was set (it logged the transort's default port).
80 PH/21 Added support for host-specific ports to manualroute, queryprogram,
81 fallback_hosts, and "hosts" in the smtp transport.
83 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
84 host errors such as "Connection refused".
86 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
87 authentication with radiusclient 0.4.9:
89 - Error returned from rc_read_config was caught wrongly
90 - Username/password not passed on to radius server due to wrong length.
92 The presumption is that some radiusclient API changes for 4.51/PH/17
93 were not taken care of correctly. The code is still untested by me (my
94 Linux distribution still has 0.3.2 of radiusclient), but it was
95 contributed by a Radius user.
97 PH/24 When doing a callout, the value of $domain wasn't set correctly when
98 expanding the "port" option of the smtp transport.
100 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
101 while reading a MIME header. Thanks to Tom Hughes for a patch.
103 PH/24 Include config.h inside local_scan.h so that configuration settings are
106 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
107 that in an ACL for RCPT (for example), you can examine exactly what was
110 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
111 commands, but it was not correctly comparing the address with the actual
112 client host address. Thus, it would show the EHLO address in Received:
113 header lines when this was not necessary.
115 PH/27 Added the % operator to ${eval:}.
117 PH/28 Exim tries to create and chdir to its spool directory when it starts;
118 it should be ignoring failures (because with -C, for example, it has lost
119 privilege). It wasn't ignoring creation failures other than "already
122 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
125 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
126 because an input line was too long, either on its own, or by virtue of
127 too many continuations, the temporary file was not being removed, and the
128 return code was incorrect.
130 PH/31 Missing "BOOL" in function definition in filtertest.c.
132 PH/32 Applied Sieve patches from the maintainer.
134 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
136 PH/33 Added "verify = not_blind".
138 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
139 Local/Makefile (with some defaults set). These are used in built scripts
140 such as exicyclog, but they have never been used in the exim_install
141 script (though there are many overriding facilities there). I have
142 arranged that the exim_install script now takes note of these two
145 PH/35 Installed configuration files for Dragonfly.
147 PH/36 When a locally submitted message by a trusted user did not contain a
148 From: header, and the sender address was obtained from -f or from an SMTP
149 MAIL command, and the trusted user did not use -F to supply a sender
150 name, $originator_name was incorrectly used when constructing a From:
151 header. Furthermore, $originator_name was used for submission mode
152 messages from external hosts without From: headers in a similar way,
153 which is clearly wrong.
155 PH/37 Added control=suppress_local_fixups.
157 PH/38 When log_selector = +received_sender was set, and the addition of the
158 sender made the log line's construction buffer exactly full, or one byte
159 less than full, an overflow happened when the terminating "\n" was
162 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
163 when the result of a list match is failure because a DNS lookup failed.
165 PH/40 RM_COMMAND is now used in the building process.
167 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
168 the "build-* directories that it finds.
170 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
171 address in a domain literal was a prefix of an interface address.
173 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
174 when verifying a sender address, unless rewrite_headers is false.
176 PH/44 Wrote a long comment about why errors_to addresses are verified as
177 recipients, not senders.
179 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
180 the ratelimit ACL was added.
182 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
184 PH/46 Added extra information about PostgreSQL errors to the error string.
186 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
187 avoiding the need to install special zones in a real server. This is
188 backwards compatible; if it can't find the fake resolver, it drops back.
189 Thus, both old and new test suites can be run.
191 TF/02 Added util/ratelimit.pl
193 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
194 clock is set back in time.
196 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
197 Candler <B.Candler@pobox.com>.
199 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
200 OK for addresses that are the result of redirections.
202 PH/48 A number of further additions for the benefit of the new test suite,
203 including a fake gethostbyname() that interfaces to the fake DNS resolver
206 TF/06 The fix for widen_domains has also been applied to qualify_single and
207 search_parents which are the other dnslookup options that can cause
210 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
213 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
215 TF/07 Exim produced the error message "an SRV record indicated no SMTP
216 service" if it encountered an MX record with an empty target hostname.
217 The message is now "an MX or SRV record indicated no SMTP service".
219 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
220 if the DNS of the sending site is misconfigured. This is quite a
221 common situation. This change restores the behaviour of treating a
222 helo verification defer as a failure.
228 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
230 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
231 command itself ended in error, the underlying message about the transport
232 filter timeout was being overwritten with the pipe command error. Now the
233 underlying error message should be appended to the second error message.
235 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
237 PH/02 Reduce the amount of output that "make" produces by default. Full output
238 can still be requested.
240 PH/03 The warning log line about a condition test deferring for a "warn" verb
241 was being output only once per connection, rather than after each
242 occurrence (because it was using the same function as for successful
243 "warn" verbs). This seems wrong, so I have changed it.
245 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
246 it should not have, which might have caused a crash in the right
247 circumstances, but probably never did.
249 PH/04 Installed a modified version of Tony Finch's patch to make submission
250 mode fix the return path as well as the Sender: header line, and to
251 add a /name= option so that you can make the user's friendly name appear
254 TF/03 Added the control = fakedefer ACL modifier.
256 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
257 Mark Lowes for thorough testing.
259 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
261 TK/03 Merged latest SRS patch from Miles Wilton.
263 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
264 with the definition in sysexits.h (which is #included earlier).
265 Fortunately, Exim does not actually use EX_OK. The code used to try to
266 preserve the sysexits.h value, by assumimg that macro definitions were
267 scanned for macro replacements. I have been disabused of this notion,
268 so now the code just undefines EX_OK before #including unistd.h.
270 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
271 in the smtp transport. When a block could not be written in a single
272 write() function, the timeout was being re-applied to each part-write.
273 This seems wrong - if the receiver was accepting one byte at a time it
274 would take for ever. The timeout is now adjusted when this happens. It
275 doesn't have to be particularly precise.
277 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
278 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
280 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
281 without a domain if the check to <postmaster@domain> fails.
283 SC/01 Eximstats: added -xls and the ability to specify output files
284 (patch written by Frank Heydlauf).
286 SC/02 Eximstats: use FileHandles for outputing results.
288 SC/03 Eximstats: allow any combination of xls, txt, and html output.
290 SC/04 Eximstats: fixed display of large numbers with -nvr option
292 SC/05 Eximstats: fixed merging of reports with empty tables.
294 SC/06 Eximstats: added the -include_original_destination flag
296 SC/07 Eximstats: removed tabs and trailing whitespace.
298 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
300 TK/06 MBOX spool code: Add real "From " MBOX separator line
301 so the .eml file is really in mbox format (even though
302 most programs do not really care). Patch from Alex Miller.
304 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
305 The latter is generated from $received_to and is only set if the
306 message has one envelope recipient. SA can use these headers,
307 obviously out-of-the-box. Patch from Alex Miller.
309 PH/08 The ${def test on a variable was returning false if the variable's
310 value was "0", contrary to what the specification has always said!
311 The result should be true unless the variable is empty.
313 PH/09 The syntax error of a character other than { following "${if
314 def:variable_name" (after optional whitespace) was not being diagnosed.
315 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
316 accidental colon was present, for example, could give incorrect results.
318 PH/10 Tidied the code in a number of places where the st_size field of a stat()
319 result is used (not including appendfile, where other changes are about
322 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
323 This involved changing a lot of size variables from int to off_t. It
324 should work with maildirs and everything.
326 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
327 spamd dying while we are connected to it.
329 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
332 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
333 function for BDB 4.3.
335 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
337 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
338 the sources for PCRE have been re-arranged and more files are now
341 PH/15 The code I had for printing potentially long long variables in PH/11
342 above was not the best (it lost precision). The length of off_t variables
343 is now inspected at build time, and an appropriate printing format (%ld
344 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
345 to be "long long int" or "long int". This is needed for the internal
346 formatting function string_vformat().
348 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
349 the configuration file to be ":syslog", then the script "guesses" where
350 the logs files are, rather than using the compiled in default. In our
351 case the guess is not the same as the compiled default, so the script
352 suddenly stopped working when I started to use syslog. The patch checks
353 to see if log_file_path is "". If so, it attempts to read it from exim
354 with no configuration file to get the compiled in version, before it
355 falls back to the previous guessing code."
357 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
358 implementing BATV in an Exim configuration. See NewStuff for the gory
361 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
362 Makefile that are specific to HP-UX.
364 PH/18 If the "use_postmaster" option was set for a recipient callout together
365 with the "random" option, the postmaster address was used as the MAIL
366 FROM address for the random test, but not for the subsequent recipient
367 test. It is now used for both.
369 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
370 patch removes a few documentation additions to RFC 3028, because the
371 latest draft now contains them. It adds the new en;ascii-case comparator
372 and a new error check for 8bit text in MIME parts. Comparator and
373 require names are now matched exactly. I enabled the subaddress
374 extension, but it is not well tested yet (read: it works for me)."
376 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
377 rework some of the code of TK/09 above to avoid the hardwired use of
378 "%lld" and "long long". Replaced the call to snprintf() with a call to
381 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
382 records point to non-existent hosts", "retry timeout exceeded", and
383 "retry time not reached for any host after a long failure period".
385 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
386 experimental DomainKeys support:
388 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
389 (2) On an error such as an illegally used "control", the wrong name for
390 the control was given.
392 These problems did NOT occur unless DomainKeys support was compiled.
394 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
396 PH/24 Added ${if match_ip condition.
398 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
399 databases so that it will be absolutely obvious if a crash occurs in the
400 DB library. This is a regular occurrence (often caused by mis-matched
403 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
404 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
405 picked up on a user's system that detects such things. There doesn't seem
406 to be a gcc warning option for this - only an attribute that has to be
407 put on the function's prototype. It seems that in Fedora Core 4 they have
408 set this on a number of new functions. No doubt there will be more in due
411 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
412 specify a transport. However, if an address that was verified by such a
413 router was the subject of a callout, Exim crashed because it tried to
414 read the rcpt_include_affixes from the non-existent transport. Now it
415 just assumes that the setting of that option is false. This bug was
416 introduced by 4.51/PH/31.
418 PH/28 Changed -d+all to exclude +memory, because that information is very
419 rarely of interest, but it makes the output a lot bigger. People tend to
420 do -d+all out of habit.
422 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
423 code in os-type was giving problems when libc.so lives in lib64, like on
426 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
427 aren't the modern standard, and it seems that some systems' include files
428 don't always have them. Exim was already checking for some of the newer
429 ones like T_AAAA, and defining it itself. I've added checks for all the
430 record types that Exim uses.
432 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
433 not automatically generating a new one, as it is supposed to. This
434 prevented TLS from working. If the file did exist, but contained invalid
435 data, a new version was generated, as expected. It was only the case of a
436 non-existent file that was broken.
438 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
439 with a change in libdomainkeys > 0.64.
441 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
442 from DNS. If the selector record carries the flag, it now has
443 precedence over the domain-wide flag.
445 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
447 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
448 the use of a port name that isn't defined in /etc/services) occurred, the
449 message was deferred as in a normal delivery, and thus remained on the
450 spool, instead of being failed because of the mua_wrapper setting. This
451 is now fixed, and I tidied up some of the mua_wrapper messages at the
454 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
455 the messages in a hash of arrays rather than using individual hashes.
456 This is a bit cleaner and results in dramatic memory savings, albeit
457 at a slight CPU cost.
459 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
460 as requested by Marc Sherman.
462 SC/10 Eximstats: added histograms for user specified patterns as requested
465 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
467 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
468 fopen() in the content-scanning modules that did not already have it.
474 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
475 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
477 TK/02 Fix ACL "control" statement not being available in MIME ACL.
479 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
481 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
482 to test Sieve filters that use "vacation".
484 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
485 that changes the way the GnuTLS parameters are stored in the cache file.
486 The new format can be generated externally. For backward compatibility,
487 if the data in the cache doesn't make sense, Exim assumes it has read an
488 old-format file, and it generates new data and writes a new file. This
489 means that you can't go back to an older release without removing the
492 PH/03 A redirect router that has both "unseen" and "one_time" set does not
493 work if there are any delivery delays because "one_time" forces the
494 parent to be marked "delivered", so its unseen clone is never tried
495 again. For this reason, Exim now forbids the simultaneous setting of
498 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
499 redirected to themselves ("homonym" addresses). Read the long ChangeLog
500 entry if you want to know the details. The fix, however, neglected to
501 consider the case when local delivery batching is involved. The test for
502 "previously delivered" was not happening when checking to see if an
503 address could be batched with a previous (undelivered) one; under
504 certain circumstances this could lead to multiple deliveries to the same
507 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
508 in its include files, and this causes problems building Exim.
510 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
511 header_syntax) but Exim was just ignoring anything given after a slash.
512 In particular, this caused confusion with an attempt to use "verify =
513 reverse_host_lookup/defer_ok". An error is now given when options are
514 supplied for verify items that do not have them. (Maybe reverse_host_
515 lookup should have a defer_ok option, but that's a different point.)
517 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
518 defined by RFC 821) to 2048, because there were problems with some AUTH
519 commands, and RFC 1869 says the size should be increased for extended
520 SMTP commands that take arguments.
522 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
525 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
526 "unknown" error; now it says that the functionality isn't in the binary.
528 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
529 an address' error message when a string expansion fails (syntax or
530 whatever). Otherwise the password may appear in the log. Following change
531 PH/42 below, there is no longer a chance of it appearing in a bounce
534 PH/11 Installed exipick version 20050225.0 from John Jetmore.
536 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
537 of its addresses was ever tried. (Bugzilla bug #2.)
539 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
540 the result incorrectly in the debug output. (It correctly added a newline
541 to what was transported.)
543 TF/01 Added $received_time.
545 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
546 commented out examples of how to interface to a virus scanner and to
547 SpamAssassin. Also added commented examples of av_scanner and
548 spamd_address settings.
550 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
551 and controls are allowed in which ACLs. There were a couple of minor
552 errors. Some of the entries in the conditions table (which is a table of
553 where they are NOT allowed) were getting very unwieldy; rewrote them as a
554 negation of where the condition IS allowed.
556 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
558 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
559 header file does not have a version number, so I've had to invent a new
560 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
561 API. The code is untested by me (my Linux distribution still has 0.3.2 of
562 radiusclient), but it was contributed by a Radius user.
564 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
565 files or directories, for OpenSSL.
567 PH/19 When an Exim process that is running as root has to create an Exim log
568 file, it does so in a subprocess that runs as exim:exim so as to get the
569 ownership right at creation (otherwise, other Exim processes might see
570 the file with the wrong ownership). There was no test for failure of this
571 fork() call, which would lead to the process getting stuck as it waited
572 for a non-existent subprocess. Forks do occasionally fail when resources
573 run out. I reviewed all the other calls to fork(); they all seem to check
576 PH/20 When checking for unexpected SMTP input at connect time (before writing
577 the banner), Exim was not dealing correctly with a non-positive return
578 from the read() function. If the client had disconnected by this time,
579 the result was a log entry for a synchronization error with an empty
580 string after "input=" when read() returned zero. If read() returned -1
581 (an event I could not check), uninitialized data bytes were printed.
582 There were reports of junk text (parts of files, etc) appearing after
585 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
587 PH/22 Added support for macro redefinition, and (re)definition in between
588 driver and ACL definitions.
590 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
591 forgetting to use the resulting value; it was using the unexpanded value.
593 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
594 hadn't been configured. The fix is from Juergen Kreileder, who
595 understands it better than I do:
597 "Here's what I see happening with three configured cyrus_sasl
598 authenticators configured (plain, login, cram-md5):
600 On startup auth_cyrus_sasl_init() gets called for each of these.
601 This means three calls to sasl_listmech() without a specified mech_list.
602 => SASL tests which mechs of all available mechs actually work
603 => three warnings about OTP not working
604 => the returned list contains: plain, login, cram-md5, digest-md5, ...
606 With the patch, sasl_listmech() also gets called three times. But now
607 SASL's mech_list option is set to the server_mech specified in the the
608 authenticator. Or in other words, the answer from sasl_listmech()
609 gets limited to just the mech you're testing for (which is different
611 => the return list contains just 'plain' or 'login', 'cram-md5' or
612 nothing depending on the value of ob->server_mech.
614 I've just tested the patch: Authentication still works fine,
615 unavailable mechs specified in the exim configuration are still
616 caught, and the auth.log warnings about OTP are gone."
618 PH/25 When debugging is enabled, the contents of the command line are added
619 to the debugging output, even when log_selector=+arguments is not
622 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
623 answer is "GNU", and only if the return is "GNU/something" is the answer
626 PH/27 $acl_verify_message is now set immediately after the failure of a
627 verification in an ACL, and so is available in subsequent modifiers. In
628 particular, the message can be preserved by coding like this:
630 warn !verify = sender
631 set acl_m0 = $acl_verify_message
633 Previously, $acl_verify_message was set only while expanding "message"
634 and "log_message" when a very denied access.
636 PH/28 Modified OS/os.c-Linux with
638 -#ifndef OS_LOAD_AVERAGE
639 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
641 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
642 nomenclature these days.)
644 PH/29 Installed patch from the Sieve maintainer that adds the options
645 sieve_useraddress and sieve_subaddress to the redirect router.
647 PH/30 In these circumstances:
648 . Two addresses routed to the same list of hosts;
649 . First host does not offer TLS;
650 . First host accepts first address;
651 . First host gives temporary error to second address;
652 . Second host offers TLS and a TLS session is established;
653 . Second host accepts second address.
654 Exim incorrectly logged both deliveries with the TLS parameters (cipher
655 and peerdn, if requested) that were in fact used only for the second
658 PH/31 When doing a callout as part of verifying an address, Exim was not paying
659 attention to any local part prefix or suffix that was matched by the
660 router that accepted the address. It now behaves in the same way as it
661 does for delivery: the affixes are removed from the local part unless
662 rcpt_include_affixes is set on the transport.
664 PH/32 Add the sender address, as F=<...>, to the log line when logging a
665 timeout during the DATA phase of an incoming message.
667 PH/33 Sieve envelope tests were broken for match types other than :is. I have
668 applied a patch sanctioned by the Sieve maintainer.
670 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
671 the uid or gid is negative. A case of a negative gid caused this to be
672 noticed. The fix allows for either to be negative.
674 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
675 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
678 PH/36 The change PH/12 above was broken. Fixed it.
680 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
681 the grounds that routing the same address twice would always produce the
682 same answer. This might have been true once, but it is certainly no
683 longer true now. Routing a child address may depend on the previous
684 routing that produced that child. Some complicated redirection strategies
685 went wrong when messages had multiple recipients, and made Exim's
686 behaviour dependent on the order in which the addresses were given.
688 I have moved the duplicate checking until after the routing is complete.
689 Exim scans the addresses that are assigned to local and remote
690 transports, and removes any duplicates. This means that more work will be
691 done, as duplicates will always all be routed, but duplicates are
692 presumably rare, so I don't expect this is of any significance.
694 For deliveries to pipes, files, and autoreplies, the duplicate checking
695 still happens during the routing process, since they are not going to be
698 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
699 It corrects a timeout issue with spamd. This is Ian's comment: "The
700 background is that sometimes spamd either never reads data from a
701 connection it has accepted, or it never writes response data. The exiscan
702 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
703 blindly assumes that writes won't block so it may never time out."
705 PH/39 Allow G after quota size as well as K and M.
707 PH/40 The value set for $authenticated_id in an authenticator may not contain
708 binary zeroes or newlines because the value is written to log lines and
709 to spool files. There was no check on this. Now the value is run through
710 the string_printing() function so that such characters are converted to
711 printable escape sequences.
713 PH/41 $message_linecount is a new variable that contains the total number of
714 lines in the message. Compare $body_linecount, which is the count for the
717 PH/42 Exim no longer gives details of delivery errors for specific addresses in
718 bounce and delay warning messages, except in certain special cases, which
721 (a) An SMTP error message from a remote host;
722 (b) A message specified in a :fail: redirection;
723 (c) A message specified in a "fail" command in a system filter;
724 (d) A message specified in a FAIL return from the queryprogram router;
725 (e) A message specified by the cannot_route_message router option.
727 In these cases only, Exim does include the error details in bounce and
728 warning messages. There are also a few cases where bland messages such
729 as "unrouteable address" or "local delivery error" are given.
731 PH/43 $value is now also set for the "else" part of a ${run expansion.
733 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
734 being worked on, but at least Exim now implements the latest version to
737 PH/45 In a pipe transport, although a timeout while waiting for the pipe
738 process to complete was treated as a delivery failure, a timeout while
739 writing the message to the pipe was logged, but erroneously treated as a
740 successful delivery. Such timeouts include transport filter timeouts. For
741 consistency with the overall process timeout, these timeouts are now
742 treated as errors, giving rise to delivery failures by default. However,
743 there is now a new Boolean option for the pipe transport called
744 timeout_defer, which, if set TRUE, converts the failures into defers for
745 both kinds of timeout. A transport filter timeout is now identified in
748 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
749 systems where "make" and "gmake" are different, calling "gmake" at top
750 level broke things. I've arranged for the value of $(MAKE) to be passed
751 from the Makefile to this script so that it can call the same version of
755 A note about Exim versions 4.44 and 4.50
756 ----------------------------------------
758 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
759 changes of various kinds. As a consequence, a big documentation update was
760 needed. This delayed the release for rather longer than seemed good, especially
761 in the light of a couple of (minor) security issues. Therefore, the changes
762 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
763 release. So 4.44 and 4.50 are in effect two different branches that both start
766 I have left the 4.50 change log unchanged; it contains all the changes since
767 4.43. The change log for 4.44 is below; many of its items are identical to
768 those for 4.50. This seems to be the most sensible way to preserve the
769 historical information.
775 1. Minor wording change to the doc/README.SIEVE file.
777 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
778 computation of the current number of files was incorrect.
780 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
781 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
784 4. Give more explanation in the error message when the command for a transport
785 filter fails to execute.
787 5. There are several places where Exim runs a non-Exim command in a
788 subprocess. The SIGUSR1 signal should be disabled for these processes. This
789 was being done only for the command run by the queryprogram router. It is
790 now done for all such subprocesses. The other cases are: ${run, transport
791 filters, and the commands run by the lmtp and pipe transports.
793 6. Added CONFIGURE_GROUP build-time option.
795 7. Some older OS have a limit of 256 on the maximum number of file
796 descriptors. Exim was using setrlimit() to set 1000 as a large value
797 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
798 systems. I've change it so that if it can't get 1000, it tries for 256.
800 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
801 was an oversight, and furthermore, ever since the addition of extra
802 controls (e.g. 4.43/32), the checks on when to allow different forms of
803 "control" were broken. There should now be diagnostics for all cases when a
804 control that does not make sense is encountered.
806 9. Added the /retain_sender option to "control=submission".
808 10. $recipients is now available in the predata ACL (oversight).
810 11. Tidy the search cache before the fork to do a delivery from a message
811 received from the command line. Otherwise the child will trigger a lookup
812 failure and thereby defer the delivery if it tries to use (for example) a
813 cached ldap connection that the parent has called unbind on.
815 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
816 of $address_data from the recipient verification was clobbered by the
819 13. The value of address_data from a sender verification is now available in
820 $sender_address_data in subsequent conditions in the ACL statement.
822 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
824 15. Added a new option "connect=<time>" to callout options, to set a different
827 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
828 was its contents. (It was OK if the option was not defined at all.)
830 17. A "Completed" log line is now written for messages that are removed from
831 the spool by the -Mrm option.
833 18. New variables $sender_verify_failure and $recipient_verify_failure contain
834 information about exactly what failed.
836 19. Added -dd to debug only the daemon process.
838 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
839 handles timeouts, both on the server side and network timeouts. Renamed the
840 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
842 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
843 It is now set to "smtps".
845 22. $host_address is now set to the target address during the checking of
848 23. When checking ignore_target_hosts for an ipliteral router, no host name was
849 being passed; this would have caused $sender_host_name to have been used if
850 matching the list had actually called for a host name (not very likely,
851 since this list is usually IP addresses). A host name is now passed as
854 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
855 code that specifies a non-restarting handler (typically sigaction() in
856 modern systems) in an attempt to fix a rare and obscure crash bug.
858 25. Narrowed the window for a race in the daemon that could cause it to ignore
859 SIGCHLD signals. This is not a major problem, because they are used only to
860 wake it up if nothing else does.
862 26. A malformed maildirsize file could cause Exim to calculate negative values
863 for the mailbox size or file count. Odd effects could occur as a result.
864 The maildirsize information is now recalculated if the size or filecount
867 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
868 support for a long time. Removed HAVE_SYS_VFS_H.
870 28. Installed the latest version of exipick from John Jetmore.
872 29. In an address list, if the pattern was not a regular expression, an empty
873 subject address (from a bounce message) matched only if the pattern was an
874 empty string. Non-empty patterns were not even tested. This was the wrong
875 because it is perfectly reasonable to use an empty address as part of a
876 database query. An empty address is now tested by patterns that are
877 lookups. However, all the other forms of pattern expect the subject to
878 contain a local part and a domain, and therefore, for them, an empty
879 address still always fails if the pattern is not itself empty.
881 30. Exim went into a mad DNS loop when attempting to do a callout where the
882 host was specified on an smtp transport, and looking it up yielded more
885 31. Re-factored the code for checking spool and log partition space into a
886 function that finds that data and another that does the check. The former
887 is then used to implement four new variables: $spool_space, $log_space,
888 $spool_inodes, and $log_inodes.
890 32. The RFC2047 encoding function was originally intended for short strings
891 such as real names; it was not keeping to the 75-character limit for
892 encoded words that the RFC imposes. It now respects the limit, and
893 generates multiple encoded words if necessary. To be on the safe side, I
894 have increased the buffer size for the ${rfc2047: expansion operator from
897 33. It is now permitted to omit both strings after an "if" condition; if the
898 condition is true, the result is "true". As before, when the second string
899 is omitted, a false condition yields an empty string. This makes it less
900 cumbersome to write custom ACL and router conditions.
902 34. Failure to deliver a bounce message always caused it to be frozen, even if
903 there was an errors_to setting on the router. The errors_to setting is now
906 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
907 canonical form (fully expanded) before being placed in
908 $sender_host_address.
910 36. The table in the code that translates DNS record types into text (T_A to
911 "A" for instance) was missing entries for NS and CNAME. It is just possible
912 that this could have caused confusion if both these types were looked up
913 for the same domain, because the text type is used as part of Exim's
914 per-process caching. But the chance of anyone hitting this buglet seems
917 37. The dnsdb lookup has been extended in a number of ways.
919 (1) There is a new type, "zns", which walks up the domain tree until it
920 finds some nameserver records. It should be used with care.
922 (2) There is a new type, "mxh", which is like "mx" except that it returns
923 just the host names, not the priorities.
925 (3) It is now possible to give a list of domains (or IP addresses) to be
926 looked up. The behaviour when one of the lookups defers can be
927 controlled by a keyword.
929 (4) It is now possible to specify the separator character for use when
930 multiple records are returned.
932 38. The dnslists ACL condition has been extended: it is now possible to supply
933 a list of IP addresses and/or domains to be looked up in a particular DNS
936 39. Added log_selector=+queue_time_overall.
938 40. When running the queue in the test harness, wait just a tad after forking a
939 delivery process, to get repeatability of debugging output.
941 41. Include certificate and key file names in error message when GnuTLS fails
942 to set them up, because the GnuTLS error message doesn't include the name
943 of the failing file when there is a problem reading it.
945 42. Allow both -bf and -bF in the same test run.
947 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
949 44. The "Exiscan patch" is now merged into the mainline Exim source.
951 45. Sometimes the final signoff response after QUIT could fail to get
952 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
953 < 0 before doing a fflush(). This bug looks as though it goes back to the
954 introduction of TLS in release 3.20, but "sometimes" must have been rare
955 because the tests only now provoked it.
957 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
958 (this can affect the format of dates).
960 47. exim_tidydb, when checking for the continued existence of a message for
961 which it has found a message-specific retry record, was not finding
962 messages that were in split spool directories. Consequently, it was
963 deleting retry records that should have stayed in existence.
965 48. Steve fixed some bugs in eximstats.
967 49. The SPA authentication driver was not abandoning authentication and moving
968 on to the next authenticator when an expansion was forced to fail,
969 contradicting the general specification for all authenticators. Instead it
970 was generating a temporary error. It now behaves as specified.
972 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
973 (the order specifies the preference for clients). The order is now AES256,
974 AES128, 3DES, ARCFOUR128.
976 51. Small patch to Sieve code - explicitly set From: when generating an
979 52. Exim crashed if a remote delivery caused a very long error message to be
980 recorded - for instance if somebody sent an entire SpamAssassin report back
981 as a large number of 550 error lines. This bug was coincidentally fixed by
982 increasing the size of one of Exim's internal buffers (big_buffer) that
983 happened as part of the Exiscan merge. However, to be on the safe side, I
984 have made the code more robust (and fixed the comments that describe what
987 53. Now that there can be additional text after "Completed" in log lines (if
988 the queue_time_overall log selector is set), a one-byte patch to exigrep
989 was needed to allow it to recognize "Completed" as not the last thing in
992 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
993 patch that reportedly fixes this has been added. I am not expert enough to
994 create a test for it. This is what the patch creator wrote:
996 "I found a little strange behaviour of ldap code when working with
997 Windows 2003 AD Domain, where users was placed in more than one
998 Organization Units. When I tried to give exim partial DN, the exit code
999 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1000 But simultaneously result of request was absolutely normal ldap result,
1001 so I produce this patch..."
1003 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1004 REFERENCE, so I have modified the code to exclude the patch when that macro
1007 55. Some experimental protocols are using DNS PTR records for new purposes. The
1008 keys for these records are domain names, not reversed IP addresses. The
1009 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1010 leaves it alone. Component reversal etc. now happens only for IP addresses.
1013 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1015 57. Double the size of the debug message buffer (to 2048) so that more of very
1016 long debug lines gets shown.
1018 58. The exicyclog utility now does better if the number of log files to keep
1019 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1021 59. Two changes related to the smtp_active_hostname option:
1023 (1) $smtp_active_hostname is now available as a variable.
1024 (2) The default for smtp_banner uses $smtp_active_hostname instead
1025 of $primary_hostname.
1027 60. The host_aton() function is supposed to be passed a string that is known
1028 to be a valid IP address. However, in the case of IPv6 addresses, it was
1029 not checking this. This is a hostage to fortune. Exim now panics and dies
1030 if the condition is not met. A case was found where this could be provoked
1031 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1032 components; fortuitously, this particular loophole had already been fixed
1033 by change 4.50/55 above.
1035 If there are any other similar loopholes, the new check in host_aton()
1036 itself should stop them being exploited. The report I received stated that
1037 data on the command line could provoke the exploit when Exim was running as
1038 exim, but did not say which command line option was involved. All I could
1039 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1040 running as the user.
1043 61. There was a buffer overflow vulnerability in the SPA authentication code
1044 (which came originally from the Samba project). I have added a test to the
1045 spa_base64_to_bits() function which I hope fixes it.
1048 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1049 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1051 63. The daemon start-up calls getloadavg() while still root for those OS that
1052 need the first call to be done as root, but it missed one case: when
1053 deliver_queue_load_max is set with deliver_drop_privilege. This is
1054 necessary for the benefit of the queue runner, because there is no re-exec
1055 when deliver_drop_privilege is set.
1057 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1058 This has been fixed.
1060 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1061 was in use, was not putting the data itself into the right store pool;
1062 consequently, it could be overwritten for a subsequent message in the same
1063 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1066 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1068 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1069 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1070 the function were treating the return as a boolean value, which happened to
1071 work because 0=false and not-0=true, but is not correct code.
1073 68. The host_aton() function was not handling scoped IPv6 addresses (those
1074 with, for example, "%eth0" on the end) correctly.
1076 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1077 negated items (that is, ~something) in unsigned ints. Some compilers
1078 apparently mutter when there is no cast.
1080 70. If an address verification called from an ACL failed, and did not produce a
1081 user-specific message (i.e. there was only a "system" message), nothing was
1082 put in $acl_verify_message. In this situation, it now puts the system
1085 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1086 session; change 4.31/43 added the unwanted input to the log line - except
1087 that it did not do this in the start of session case. It now does.
1089 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1090 This is wrong and can cause the other end to generate a synchronization
1091 error if it is another Exim or anything else that does the synchronization
1092 check. A QUIT command is no longer sent after a timeout.
1094 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1095 during host lookups.
1097 74. The defer_ok option of callout verification was not working if it was used
1098 when verifying addresses in header lines, that is, for this case:
1100 verify = header_sender/callout=defer_ok
1102 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1103 those file descriptors could be used for SMTP connections. If anything
1104 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1105 could be sent to the SMTP client, causing chaos. The daemon now opens
1106 stdin, stdout, and stderr to /dev/null when it puts itself into the
1109 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1110 log by default. The user can override this with suitable Perl magic.
1112 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1113 the log message when discard triggers, was not working for the DATA ACL or
1114 for the non-SMTP ACL.
1116 78. Error message wording change in sieve.c.
1118 79. If smtp_accept_max_per_host was set, the number of connections could be
1119 restricted to fewer than expected, because the daemon was trying to set up
1120 a new connection before checking whether the processes handling previous
1121 connections had finished. The check for completed processes is now done
1122 earlier. On busy systems, this bug wouldn't be noticed because something
1123 else would have woken the daemon, and it would have reaped the completed
1126 80. If a message was submitted locally by a user whose login name contained one
1127 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1128 It caused a spool format error. I have fixed the spool reading code. A
1129 related problem was that the "from" clause in the Received: line became
1130 illegal because of the space(s). It is now covered by ${quote_local_part.
1132 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1135 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1137 83. Patch from Sieve maintainer to fix unterminated string problem in
1138 "vacation" handling.
1140 84. Some minor changes to the Linux configuration files to help with other
1141 OS variants using glibc.
1143 85. One more patch for Sieve to update vacation handling to latest spec.
1146 ----------------------------------------------------
1147 See the note above about the 4.44 and 4.50 releases.
1148 ----------------------------------------------------
1154 1. Change 4.43/35 introduced a bug that caused file counts to be
1155 incorrectly computed when quota_filecount was set in an appendfile
1158 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1159 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1162 3. Give more explanation in the error message when the command for a transport
1163 filter fails to execute.
1165 4. There are several places where Exim runs a non-Exim command in a
1166 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1167 was being done only for the command run by the queryprogram router. It is
1168 now done for all such subprocesses. The other cases are: ${run, transport
1169 filters, and the commands run by the lmtp and pipe transports.
1171 5. Some older OS have a limit of 256 on the maximum number of file
1172 descriptors. Exim was using setrlimit() to set 1000 as a large value
1173 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1174 systems. I've change it so that if it can't get 1000, it tries for 256.
1176 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1177 was an oversight, and furthermore, ever since the addition of extra
1178 controls (e.g. 4.43/32), the checks on when to allow different forms of
1179 "control" were broken. There should now be diagnostics for all cases when a
1180 control that does not make sense is encountered.
1182 7. $recipients is now available in the predata ACL (oversight).
1184 8. Tidy the search cache before the fork to do a delivery from a message
1185 received from the command line. Otherwise the child will trigger a lookup
1186 failure and thereby defer the delivery if it tries to use (for example) a
1187 cached ldap connection that the parent has called unbind on.
1189 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1190 of $address_data from the recipient verification was clobbered by the
1191 sender verification.
1193 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1194 was its contents. (It was OK if the option was not defined at all.)
1196 11. A "Completed" log line is now written for messages that are removed from
1197 the spool by the -Mrm option.
1199 12. $host_address is now set to the target address during the checking of
1200 ignore_target_hosts.
1202 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1203 being passed; this would have caused $sender_host_name to have been used if
1204 matching the list had actually called for a host name (not very likely,
1205 since this list is usually IP addresses). A host name is now passed as
1208 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1209 code that specifies a non-restarting handler (typically sigaction() in
1210 modern systems) in an attempt to fix a rare and obscure crash bug.
1212 15. Narrowed the window for a race in the daemon that could cause it to ignore
1213 SIGCHLD signals. This is not a major problem, because they are used only to
1214 wake it up if nothing else does.
1216 16. A malformed maildirsize file could cause Exim to calculate negative values
1217 for the mailbox size or file count. Odd effects could occur as a result.
1218 The maildirsize information is now recalculated if the size or filecount
1221 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1222 support for a long time. Removed HAVE_SYS_VFS_H.
1224 18. Updated exipick to current release from John Jetmore.
1226 19. Allow an empty sender to be matched against a lookup in an address list.
1227 Previously the only cases considered were a regular expression, or an
1230 20. Exim went into a mad DNS lookup loop when doing a callout where the
1231 host was specified on the transport, if the DNS lookup yielded more than
1234 21. The RFC2047 encoding function was originally intended for short strings
1235 such as real names; it was not keeping to the 75-character limit for
1236 encoded words that the RFC imposes. It now respects the limit, and
1237 generates multiple encoded words if necessary. To be on the safe side, I
1238 have increased the buffer size for the ${rfc2047: expansion operator from
1241 22. Failure to deliver a bounce message always caused it to be frozen, even if
1242 there was an errors_to setting on the router. The errors_to setting is now
1245 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1246 canonical form (fully expanded) before being placed in
1247 $sender_host_address.
1249 24. Updated eximstats to version 1.33
1251 25. Include certificate and key file names in error message when GnuTLS fails
1252 to set them up, because the GnuTLS error message doesn't include the name
1253 of the failing file when there is a problem reading it.
1255 26. Expand error message when OpenSSL has problems setting up cert/key files.
1258 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1259 (this can affect the format of dates).
1261 28. exim_tidydb, when checking for the continued existence of a message for
1262 which it has found a message-specific retry record, was not finding
1263 messages that were in split spool directories. Consequently, it was
1264 deleting retry records that should have stayed in existence.
1266 29. eximstats updated to version 1.35
1267 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1268 1.35 - bugfix such that pie charts by volume are generated correctly
1270 30. The SPA authentication driver was not abandoning authentication and moving
1271 on to the next authenticator when an expansion was forced to fail,
1272 contradicting the general specification for all authenticators. Instead it
1273 was generating a temporary error. It now behaves as specified.
1275 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1276 (the order specifies the preference for clients). The order is now AES256,
1277 AES128, 3DES, ARCFOUR128.
1279 31. Small patch to Sieve code - explicitly set From: when generating an
1282 32. Exim crashed if a remote delivery caused a very long error message to be
1283 recorded - for instance if somebody sent an entire SpamAssassin report back
1284 as a large number of 550 error lines. This bug was coincidentally fixed by
1285 increasing the size of one of Exim's internal buffers (big_buffer) that
1286 happened as part of the Exiscan merge. However, to be on the safe side, I
1287 have made the code more robust (and fixed the comments that describe what
1290 33. Some experimental protocols are using DNS PTR records for new purposes. The
1291 keys for these records are domain names, not reversed IP addresses. The
1292 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1293 leaves it alone. Component reversal etc. now happens only for IP addresses.
1296 34. The host_aton() function is supposed to be passed a string that is known
1297 to be a valid IP address. However, in the case of IPv6 addresses, it was
1298 not checking this. This is a hostage to fortune. Exim now panics and dies
1299 if the condition is not met. A case was found where this could be provoked
1300 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1301 components; fortuitously, this particular loophole had already been fixed
1302 by change 4.50/55 or 4.44/33 above.
1304 If there are any other similar loopholes, the new check in host_aton()
1305 itself should stop them being exploited. The report I received stated that
1306 data on the command line could provoke the exploit when Exim was running as
1307 exim, but did not say which command line option was involved. All I could
1308 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1309 running as the user.
1312 35. There was a buffer overflow vulnerability in the SPA authentication code
1313 (which came originally from the Samba project). I have added a test to the
1314 spa_base64_to_bits() function which I hope fixes it.
1317 36. The daemon start-up calls getloadavg() while still root for those OS that
1318 need the first call to be done as root, but it missed one case: when
1319 deliver_queue_load_max is set with deliver_drop_privilege. This is
1320 necessary for the benefit of the queue runner, because there is no re-exec
1321 when deliver_drop_privilege is set.
1323 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1324 was in use, was not putting the data itself into the right store pool;
1325 consequently, it could be overwritten for a subsequent message in the same
1326 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1329 38. Sometimes the final signoff response after QUIT could fail to get
1330 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1331 < 0 before doing a fflush(). This bug looks as though it goes back to the
1332 introduction of TLS in release 3.20, but "sometimes" must have been rare
1333 because the tests only now provoked it.
1339 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1340 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1341 Now it returns an integer. A number of calls were still expecting a T/F
1342 return. Fortuitously, in all cases, the tests worked in OK situations,
1343 which is the norm. However, things would have gone wrong on any write
1344 failures on the smtp file descriptor. This function is used when sending
1345 messages over SMTP and also when doing verify callouts.
1347 2. When Exim is called to do synchronous delivery of a locally submitted
1348 message (the -odf or -odi options), it no longer closes stderr before doing
1351 3. Implemented the mua_wrapper option.
1353 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1355 5. Implemented the functions header_remove(), header_testname(),
1356 header_add_at_position(), and receive_remove_recipient(), and exported them
1359 6. If an ACL "warn" statement specified the addition of headers, Exim already
1360 inserted X-ACL-Warn: at the start if there was no header name. However, it
1361 was not making this test for the second and subsequent header lines if
1362 there were newlines in the string. This meant that an invalid header could
1363 be inserted if Exim was badly configured.
1365 7. Allow an ACL "warn" statement to add header lines at the start or after all
1366 the Received: headers, as well as at the end.
1368 8. Added the rcpt_4xx retry error code.
1370 9. Added postmaster_mailfrom=xxx to callout verification option.
1372 10. Added mailfrom=xxxx to the callout verification option, for verify=
1375 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1376 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1378 12. Inserted some casts to stop certain compilers warning when using pointer
1379 differences as field lengths or precisions in printf-type calls (mostly
1380 affecting debugging statements).
1382 13. Added optional readline() support for -be (dynamically loaded).
1384 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1385 same clock tick as a message's arrival, so that its received time was the
1386 same as the "first fail" time on the retry record, and that message
1387 remained on the queue past the ultimate address timeout, every queue runner
1388 would try a delivery (because it was past the ultimate address timeout) but
1389 after another failure, the ultimate address timeout, which should have then
1390 bounced the address, did not kick in. This was a "< instead of <=" error;
1391 in most cases the first failure would have been in the next clock tick
1392 after the received time, and all would be well.
1394 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1395 being recognized when the domain list was tested by the match_domain
1396 condition in an expansion string.
1398 16. Added the ${str2b64: operator.
1400 17. Exim was always calling setrlimit() to set a large limit for the number of
1401 processes, without checking whether the existing limit was already
1402 adequate. (It did check for the limit on file descriptors.) Furthermore,
1403 errors from getrlimit() and setrlimit() were being ignored. Now they are
1404 logged to the main and panic logs, but Exim does carry on, to try to do its
1405 job under whatever limits there are.
1407 18. Imported PCRE 5.0.
1409 19. Trivial typo in log message " temporarily refused connection" (the leading
1412 20. If the log selector return_path_on_delivery was set and an address was
1413 redirected to /dev/null, the delivery process crashed because it assumed
1414 that a return path would always be set for a "successful" delivery. In this
1415 case, the whole delivery is bypassed as an optimization, and therefore no
1418 21. Internal re-arrangement: the function for sending a challenge and reading
1419 a response while authentication was assuming a zero-terminated challenge
1420 string. It's now changed to take a pointer and a length, to allow for
1421 binary data in such strings.
1423 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1425 23. Exim was not respecting finduser_retries when seeking the login of the
1426 uid under which it was called; it was always trying 10 times. (The default
1427 setting of finduser_retries is zero.) Also, it was sleeping after the final
1428 failure, which is pointless.
1430 24. Implemented tls_on_connect_ports.
1432 25. Implemented acl_smtp_predata.
1434 26. If the domain in control=submission is set empty, Exim assumes that the
1435 authenticated id is a complete email address when it generates From: or
1436 Sender: header lines.
1438 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1439 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1440 chown and chgrp in /bin and hostname in /usr/bin.
1442 28. Exim was keeping the "process log" file open after each use, just as it
1443 does for the main log. This opens the possibility of it remaining open for
1444 long periods when the USR1 signal hits a daemon. Occasional processlog
1445 errors were reported, that could have been caused by this. Anyway, it seems
1446 much more sensible not to leave this file open at all, so that is what now
1449 29. The long-running daemon process does not normally write to the log once it
1450 has entered its main loop, and it closes the log before doing so. This is
1451 so that log files can straightforwardly be renamed and moved. However,
1452 there are a couple of unusual error situations where the daemon does write
1453 log entries, and I had neglected to close the log afterwards.
1455 30. The text of an SMTP error response that was received during a remote
1456 delivery was being truncated at 512 bytes. This is too short for some of
1457 the long messages that one sometimes sees. I've increased the limit to
1460 31. It is now possible to make retry rules that apply only when a message has a
1461 specific sender, in particular, an empty sender.
1463 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1464 it possible to be selective about when SMTP synchronization is enforced.
1466 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1468 32. Implemented hosts_connection_nolog.
1470 33. Added an ACL for QUIT.
1472 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1475 35. Added mailbox_size and mailbox_filecount to appendfile.
1477 36. Added control = no_multiline_responses to ACLs.
1479 37. There was a bug in the logic of the code that waits for the clock to tick
1480 in the case where the clock went backwards by a substantial amount such
1481 that the microsecond fraction of "now" was more than the microsecond
1482 fraction of "then" (but the whole seconds number was less).
1484 38. Added support for the libradius Radius client library this is found on
1485 FreeBSD (previously only the radiusclient library was supported).
1491 1. When certain lookups returned multiple values in the form name=value, the
1492 quoting of the values was not always being done properly. Specifically:
1493 (a) If the value started with a double quote, but contained no whitespace,
1495 (b) If the value contained whitespace other than a space character (i.e.
1496 tabs or newlines or carriage returns) it was not quoted.
1497 This fix has been applied to the mysql and pgsql lookups by writing a
1498 separate quoting function and calling it from the lookup code. The fix
1499 should probably also be applied to nisplus, ibase and oracle lookups, but
1500 since I cannot test any of those, I have not disturbed their existing code.
1502 2. A hit in the callout cache for a specific address caused a log line with no
1503 reason for rejecting RCPT. Now it says "Previous (cached) callout
1504 verification failure".
1506 3. There was an off-by-one bug in the queryprogram router. An over-long
1507 return line was truncated at 256 instead of 255 characters, thereby
1508 overflowing its buffer with the terminating zero. As well as fixing this, I
1509 have increased the buffer size to 1024 (and made a note to document this).
1511 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1512 when Exim is waiting for an SMTP response from a remote server, Exim
1513 restarts its select() call on the socket, thereby resetting its timeout.
1514 This is not a problem when such interrupts are rare. Somebody set up a cron
1515 job to run exiwhat every 2 minutes, which is less than the normal select()
1516 timeout (5 or 10 minutes). This meant that the select() timeout never
1517 kicked in because it was always reset. I have fixed this by comparing the
1518 time when an interrupt arrives with the time at the start of the first call
1519 to select(). If more time than the timeout has elapsed, the interrupt is
1520 treated as a timeout.
1522 5. Some internal re-factoring in preparation for the addition of Sieve
1523 extensions (by MH). In particular, the "personal" test is moved to a
1524 separate function, and given an option for scanning Cc: and Bcc: (which is
1525 not set for Exim filters).
1527 6. When Exim created an email address using the login of the caller as the
1528 local part (e.g. when creating a From: or Sender: header line), it was not
1529 quoting the local part when it contained special characters such as @.
1531 7. Installed new OpenBSD configuration files.
1533 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1534 try to make them clearer.
1536 9. Callout options, other than the timeout value, were being ignored when
1537 verifying sender addresses in header lines. For example, when using
1539 verify = header_sender/callout=no_cache
1541 the cache was (incorrectly) being used.
1543 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1544 only the Cygwin environment.
1546 11. When return_path_on_delivery was set as a log selector, if different remote
1547 addresses in the same message used different return paths and parallel
1548 remote delivery occurred, the wrong values would sometimes be logged.
1549 (Whenever a remote delivery process finished, the return path value from
1550 the most recently started remote delivery process was logged.)
1552 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1553 lines when AUTH and/or TLS are in use. This is the "received protocol"
1554 field. Exim used to use "asmtp" for authenticated SMTP, without any
1555 indication (in the protocol name) for TLS use. Now it follows the RFC and
1556 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1557 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1558 names appear in log lines as well as in Received: header lines.
1560 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1561 extensions, and comparison tests, and to fix some bugs.
1563 14. Changes to the "personal" filter test:
1565 (1) The test was buggy in that it was just doing the equivalent of
1566 "contains" tests on header lines. For example, if a user's address was
1567 anne@some.where, the "personal" test would incorrectly be true for
1569 To: susanne@some.where
1571 This test is now done by extracting each address from the header in turn,
1572 and checking the entire address. Other tests that are part of "personal"
1573 are now done using regular expressions (for example, to check local parts
1574 of addresses in From: header lines).
1576 (2) The list of non-personal local parts in From: addresses has been
1577 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1578 taken from the Sieve specification recommendations.
1580 (3) If the message contains any header line starting with "List-" it is
1581 treated as non-personal.
1583 (4) The test for "circular" in the Subject: header line has been removed
1584 because it now seems ill-conceived.
1586 15. Minor typos in src/EDITME comments corrected.
1588 16. Installed latest exipick from John Jetmore.
1590 17. If headers_add on a router specified a text string that was too long for
1591 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1592 of string_sprintf() is now avoided.
1594 18. $message_body_size was not set (it was always zero) when running the DATA
1595 ACL and the local_scan() function.
1597 19. For the "mail" command in an Exim filter, no default was being set for
1598 the once_repeat time, causing a random time value to be used if "once" was
1599 specified. (If the value happened to be <= 0, no repeat happened.) The
1600 default is now 0s, meaning "never repeat". The "vacation" command was OK
1601 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1602 (I found it when inspecting the code).
1604 20. There is now an overall timeout for performing a callout verification. It
1605 defaults to 4 times the callout timeout, which applies to individual SMTP
1606 commands during the callout. The overall timeout applies when there is more
1607 than one host that can be tried. The timeout is checked before trying the
1608 next host. This prevents very long delays if there are a large number of
1609 hosts and all are timing out (e.g. when the network connections are timing
1610 out). The value of the overall timeout can be changed by specifying an
1611 additional sub-option for "callout", called "maxwait". For example:
1613 verify = sender/callout=5s,maxwait=20s
1615 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1616 to the end before writing, but this should make it even safer).
1618 22. Exim was forgetting that it had advertised PIPELINING for the second and
1619 subsequent messages on an SMTP connection. It was also not resetting its
1620 memory on STARTTLS and an internal HELO.
1622 23. When Exim logs an SMTP synchronization error within a session, it now
1623 records whether PIPELINING has been advertised or not.
1625 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1626 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1627 rather than long int.
1629 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1631 26. Added the never_mail option to autoreply.
1637 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1638 crash if the getsockname() call failed; this can happen if a connection is
1639 closed very soon after it is established. The problem was simply in the
1640 order in which certain operations were done, causing Exim to try to write
1641 to the SMTP stream before it had set up the file descriptor. The bug has
1642 been fixed by making things happen in the correct order.
1648 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1649 before the connection was closed, thus losing the rejection response.
1651 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1652 some early Solaris releases, but causes trouble in current releases where
1653 socklen_t is defined.
1655 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1658 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1661 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1662 that releasing the top of it at the end releases what was used for sub-
1663 expansions (unless the block got too big). However, discard this block if
1664 the first thing is a variable or header, so that we can use its block when
1665 it is dynamic (useful for very large $message_headers, for example).
1667 6. Lookups now cache *every* query, not just the most recent. A new, separate
1668 store pool is used for this. It can be recovered when all lookup caches are
1669 flushed. Lookups now release memory at the end of their result strings.
1670 This has involved some general refactoring of the lookup sources.
1672 7. Some code has been added to the store_xxx() functions to reduce the amount
1673 of flapping under certain conditions.
1675 8. log_incoming_interface used to affect only the <= reception log lines. Now
1676 it causes the local interface and port to be added to several more SMTP log
1677 lines, for example "SMTP connection from", and rejection lines.
1679 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1681 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1683 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1684 could be overwritten at the end of the current message (or the start of a
1685 new message if it was set in a HELO ACL). The value is now preserved for
1686 the duration of the SMTP connection.
1688 12. If a transport had a headers_rewrite setting, and a matching header line
1689 contained an unqualified address, that address was qualified, even if it
1690 did not match any rewriting rules. The underlying bug was that the values
1691 of the flags that permit the existence of unqualified sender and recipient
1692 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1693 non-local messages, and by -bnq for local messages) were not being
1694 preserved with the message after it was received.
1696 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1697 "next input=" as part of the text comprising the host identity instead of
1698 the correct text. The code was using the same buffer for two different
1699 strings. However, depending on which order the printing function evaluated
1700 its arguments, the bug did not always show up. Under Linux, for example, my
1701 test suite worked just fine.
1703 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1704 doesn't work with some older versions of Perl. It has been changed to "my",
1705 which in any case is probably the better facility to use.
1707 15. A really picky compiler found some instances of statements for creating
1708 error messages that either had too many or two few arguments for the format
1711 16. The size of the buffer for calls to the DNS resolver has been increased
1712 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1713 for addresses that have a lot of PTR records. This alleviates a problem; it
1714 does not fully solve it.
1716 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1717 buffer now truncates the list and logs the incident, which is the same
1718 action as happens when Exim is looking up a host name and its aliases.
1719 Previously in this situation something unpredictable would happen;
1720 sometimes it was "internal error: store_reset failed".
1722 18. If a server dropped the connection unexpectedly when an Exim client was
1723 using GnuTLS and trying to read a response, the client delivery process
1724 crashed while trying to generate an error log message.
1726 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1727 string, for example:
1729 warn message = H1: something\nH2: something
1731 the text was added as a single header line from Exim's point of view
1732 though it ended up OK in the delivered message. However, searching for the
1733 second and subsequent header lines using $h_h2: did not work. This has been
1734 fixed. Similarly, if a system filter added multiple headers in this way,
1735 the routers could not see them.
1737 20. Expanded the error message when iplsearch is called with an invalid key to
1738 suggest using net-iplsearch in a host list.
1740 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1741 ACLs are no longer actually imposed (and a message to that effect is
1744 22. If a "gecos" field in a passwd entry contained escaped characters, in
1745 particular, if it contained a \" sequence, Exim got it wrong when building
1746 a From: or a Sender: header from that name. A second bug also caused
1747 incorrect handling when an unquoted " was present following a character
1748 that needed quoting.
1750 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1751 was not being matched caselessly.
1753 24. Arranged for all hyphens in the exim.8 source to be escaped with
1756 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1757 properly. Recipient callout cache records were still being keyed to include
1758 the sender, even when use_sender was set false. This led to far more
1759 callouts that were necessary. The sender is no longer included in the key
1760 when use_sender is false.
1762 26. Added "control = submission" modifier to ACLs.
1764 27. Added the ${base62d: operator to decode base 62 numbers.
1766 28. dnsdb lookups can now access SRV records.
1768 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1769 the configuration file.
1771 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1772 (-v) mode. This makes the output for a verbose queue run more intelligible.
1774 31. Added a use_postmaster feature to recipient callouts.
1776 32. Added the $body_zerocount variable, containing the number of binary zero
1777 bytes in the message body.
1779 33. The time of last modification of the "new" subdirectory is now used as the
1780 "mailbox time last read" when there is a quota error for a maildir
1783 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1785 35. Added +ignore_unknown as a special item in host lists.
1787 36. Code for decoding IPv6 addresses in host lists is now included, even if
1788 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1789 address was recognized as an IP address, but was then not correctly decoded
1790 into binary, causing unexpected and incorrect effects when compared with
1797 1. Very minor rewording of debugging text in manualroute to say "list of
1798 hosts" instead of "hostlist".
1800 2. If verify=header_syntax was set, and a header line with an unqualified
1801 address (no domain) and a large number of spaces between the end of the
1802 name and the colon was received, the reception process suffered a buffer
1803 overflow, and (when I tested it) crashed. This was caused by some obsolete
1804 code that should have been removed. The fix is to remove it!
1806 3. When running in the test harness, delay a bit after writing a bounce
1807 message to get a bit more predictability in the log output.
1809 4. Added a call to search_tidyup() just before forking a reception process. In
1810 theory, someone could use a lookup in the expansion of smtp_accept_max_
1811 per_host which, without the tidyup, could leave open a database connection.
1813 5. Added the variables $recipient_data and $sender_data which get set from a
1814 lookup success in an ACL "recipients" or "senders" condition, or a router
1815 "senders" option, similar to $domain_data and $local_part_data.
1817 6. Moved the writing of debug_print from before to after the "senders" test
1820 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1821 problems for message scanning, either using a data ACL, or using
1822 local_scan() because the Received: header was not generated till after they
1823 were called (in order to set the time as the time of reception completion).
1824 I have revised the way this works. The header is now generated after the
1825 body is received, but before the ACL or local_scan() are called. After they
1826 are run, the timestamp in the header is updated.
1832 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1833 before starting a queue runner without re-exec. This happened only when
1834 deliver_drop_privilege was set or when the Exim user was set to root. The
1835 effect of the bug was that timeouts during subsequent deliveries caused
1836 crashes instead of being properly handled. The handler is now left at its
1837 default (and expected) setting.
1839 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1840 message, again when deliver_drop_privilege is set or Exim is run as root.
1841 The bug described in (1) was not present in this case, but the tidying up
1842 of the other signals was missing. I have made the two cases consistent.
1844 3. The ignore_target_hosts setting on a manualroute router was being ignored
1845 for hosts that were looked up using the /MX notation.
1847 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1850 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1851 operated on the sender address. After changing the $sender_address to <>
1852 for the sender address verify, Exim was re-instated it as the original
1853 (before rewriting) address, but remembering that it had rewritten it, so it
1854 wasn't rewriting it again. This bug also had the effect of breaking the
1855 sender address verification caching when the sender address was rewritten.
1857 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1858 This has been changed so that if the ip literal address matches
1859 ignore_target_hosts, the router declines.
1861 7. Added expansion conditions match_domain, match_address, and match_local_
1862 part (NOT match_host).
1864 8. The placeholder for the Received: header didn't have a length field set.
1866 9. Added code to Exim itself and to exim_lock to test for a specific race
1867 condition that could lead to file corruption when using MBX delivery. The
1868 issue is with the lockfile that is created in /tmp. If this file is removed
1869 after a process has opened it but before that process has acquired a lock,
1870 there is the potential for a second process to recreate the file and also
1871 acquire a lock. This could lead to two Exim processes writing to the file
1872 at the same time. The added code performs the same test as UW imapd; it
1873 checks after acquiring the lock that its file descriptor still refers to
1874 the same named file.
1876 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1877 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1880 11. Added the smtp_active_hostname option. If used, this will typically be made
1881 to depend on the incoming interface address. Because $interface_address is
1882 not set up until the daemon has forked a reception process, error responses
1883 that can happen earlier (such as "too many connections") no longer contain
1886 12. If an expansion in a condition on a "warn" statement fails because a lookup
1887 defers, the "warn" statement is abandoned, and the next ACL statement is
1888 processed. Previously this caused the whole ACL to be aborted.
1890 13. Added the iplsearch lookup type.
1892 14. Added ident_timeout as a log selector.
1894 15. Added tls_certificate_verified as a log selector.
1896 16. Added a global option tls_require_ciphers (compare the smtp transport
1897 option of the same name). This controls incoming TLS connections.
1899 17. I finally figured out how to make tls_require_ciphers do a similar thing
1900 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1901 before starting the TLS session.
1903 18. Tabs are now shown as \t in -bP output.
1905 19. If the log selector return_path_on_delivery was set, Exim crashed when
1906 bouncing a message because it had too many Received: header lines.
1908 20. If two routers both had headers_remove settings, and the first one included
1909 a superfluous trailing colon, the final name in the first list and the
1910 first name in the second list were incorrectly joined into one item (with a
1911 colon in the middle).
1917 1. Added -C and -D options to the exinext utility, mainly to make it easier
1918 to include in the automated testing, but these could be helpful when
1919 multiple configurations are in use.
1921 2. The exinext utility was not formatting the output nicely when there was
1922 an alternate port involved in the retry record key, nor when there was a
1923 message id as well (for retries that were specific to a specific message
1924 and a specific host). It was also confused by IPv6 addresses, because of
1925 the additional colons they contain. I have fixed the IPv4 problem, and
1926 patched it up to do a reasonable job for IPv6.
1928 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1929 delivery, the log line now contains "pipelined" if PIPELINING was used.
1931 4. An SMTP transport process used to panic and die if the bind() call to set
1932 an explicit outgoing interface failed. This has been changed; it is now
1933 treated in the same way as a connect() failure.
1935 5. A reference to $sender_host_name in the part of a conditional expansion
1936 that was being skipped was still causing a DNS lookup. This no longer
1939 6. The def: expansion condition was not recognizing references to header lines
1940 that used bh_ and bheader_.
1942 7. Added the _cache feature to named lists.
1944 8. The code for checking quota_filecount in the appendfile transport was
1945 allowing one more file than it should have been.
1947 9. For compatibility with Sendmail, the command line option
1955 and sets the incoming protocol and host name (for trusted callers). The
1956 host name and its colon can be omitted when only the protocol is to be set.
1957 Note the Exim already has two private options, -pd and -ps, that refer to
1958 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1959 "s", but I don't think that's a major issue.
1961 10. A number of refactoring changes to the code, none of which should affect
1964 (a) The number of logging options was getting close to filling up the
1965 32-bit word that was used as a bit map. I have split them into two classes:
1966 those that are passed in the argument to log_write(), and those that are
1967 only ever tested independently outside of that function. These are now in
1968 separate 32-bit words, so there is plenty of room for expansion again.
1969 There is no change in the user interface or the logging behaviour.
1971 (b) When building, for example, log lines, the code previously used a
1972 macro that called string_cat() twice, in order to add two strings. This is
1973 not really sufficiently general. Furthermore, there was one instance where
1974 it was actually wrong because one of the argument was used twice, and in
1975 one call a function was used. (As it happened, calling the function twice
1976 did not affect the overall behaviour.) The macro has been replaced by a
1977 function that can join an arbitrary number of extra strings onto a growing
1980 (c) The code for expansion conditions now uses a table and a binary chop
1981 instead of a serial search (which was left over from when there were very
1982 few conditions). Also, it now recognizes conditions like "pam" even when
1983 the relevant support is not compiled in: a suitably worded error message is
1984 given if an attempt is made to use such a condition.
1986 11. Added ${time_interval:xxxxx}.
1988 12. A bug was causing one of the ddress fields not to be passed back correctly
1989 from remote delivery subprocesses. The field in question was not being
1990 subsequently used, so this caused to problems in practice.
1992 13. Added new log selectors queue_time and deliver_time.
1994 14. Might have fixed a bug in maildirsizefile handling that threw up
1995 "unexpected character" debug warnings, and recalculated the data
1996 unnecessarily. In any case, I expanded the warning message to give more
1999 15. Added the message "Restricted characters in address" to the statements in
2000 the default ACL that block characters like @ and % in local parts.
2002 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2003 Three changes have been made:
2005 (a) There was a serious bug; a negative response to MAIL caused the whole
2006 recipient domain to be cached as invalid, thereby blocking all messages
2007 to all local parts at the same domain, from all senders. This bug has
2008 been fixed. The domain is no longer cached after a negative response to
2009 MAIL if the sender used is not empty.
2011 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2014 (c) A new callout option, "use_sender" has been added for people who want
2015 the modified behaviour.
2021 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2024 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2025 indeed breaks things for older releases.
2027 3. Added additional logging to the case where there is a problem reading data
2028 from a filter that is running in a subprocess using a pipe, in order to
2029 try to track down a specific problem.
2031 4. Testing facility fudge: when running in the test harness and attempting
2032 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2033 getting "No route to host". Convert this to a timeout.
2035 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2038 6. Some OS don't have socklen_t but use size_t instead. This affects the
2039 fifth argument of getsockopt() amongst other things. This is now
2040 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2041 can be set for individual OS. I have set it for SunOS5, OSF1, and
2042 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2043 some earlier ones do not.
2045 7. Change 4.30/15 was not doing the test caselessly.
2047 8. The standard form for an IPv6 address literal was being rejected by address
2048 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2049 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2050 this, as well as the form without the "IPv6" on the front (but only when
2051 address literals are enabled, of course).
2053 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2055 10. Exim crashed if a message with an empty sender address specified by -f
2056 encountered a router with an errors_to setting. This could be provoked only
2057 by a command such as
2061 where an empty string was supplied; "<>" did not hit this bug.
2063 11. Installed PCRE release 4.5.
2065 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2066 remained set. It is now erased.
2068 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2069 times from message ids (which are base 36 rather than the normal 62).
2071 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2072 were being counted as actual protocol errors, and logged if the log
2073 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2074 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2075 and DATA following a set of rejected RCPTs do not count as protocol errors.
2076 In other words, Exim assumes they were pipelined, though this may not
2077 actually be the case. Of course, in all cases the client gets an
2078 appropriate error code.
2080 15. If a lookup fails in an ACL condition, a message about the failure may
2081 be available; it is used if testing the ACL cannot continue, because most
2082 such messages specify what the cause of the deferral is. However, some
2083 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2084 that caused an old message to be retained and used if a later statement
2085 caused a defer, replacing the real cause of the deferral.
2087 16. If an IP address had so many PTR records that the DNS lookup buffer
2088 was not large enough to hold them, Exim could crash while trying to process
2089 the truncated data. It now detects and logs this case.
2091 17. Further to 4.21/58, another change has been made: if (and only if) the
2092 first line of a message (the first header line) ends with CRLF, a bare LF
2093 in a subsequent header line has a space inserted after it, so as not to
2094 terminate the header.
2096 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2097 unnecessarily, used atoi() instead of strtol(), and didn't check the
2098 termination when getting file sizes from file names by regex.
2100 19. Completely re-implemented the support for maildirsize files, in the light
2101 of a number of problems with the previous contributed implementation
2102 (4.30/29). In particular:
2104 . If the quota is zero, the maildirsize file is maintained, but no quota is
2107 . If the maildir directory does not exist, it is created before any attempt
2108 to write a maildirsize file.
2110 . The quota value in the file is just a cache; if the quota is changed in
2111 the transport, the new value overrides.
2113 . A regular expression is available for excluding directories from the
2116 20. The autoreply transport checks the characters in options that define the
2117 message's headers; it allows continued headers, but it was checking with
2118 isspace() after an embedded newline instead of explicitly looking for a
2121 21. If all the "regular" hosts to which an address was routed had passed their
2122 expiry times, and had not reached their retry times, the address was
2123 bounced, even if fallback hosts were defined. Now Exim should go on to try
2126 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2127 equivalent code in the SMTP transport. Some hosts send humungous responses
2128 to HELO/EHLO, more than 1024 it seems.
2130 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2131 gives compiler warnings in some environments. I've now done it "properly",
2134 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2135 (because of problems with the built-in one) was declared to return uschar *
2136 instead of char *, causing compiler failure.
2138 25. Fixed a file descriptor leak when processing alias/forward files.
2140 26. Fixed a minor format string issue in dbfn.c.
2142 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2144 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2145 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2148 29. When a sender address is verified, it is cached, to save repeating the test
2149 when there is more than one recipient in a message. However, when the
2150 verification involves a callout, it is possible for different callout
2151 options to be set for different recipients. It is too complicated to keep
2152 track of this in the cache, so now Exim always runs a verification when a
2153 callout is required, relying on the callout cache for the optimization.
2154 The overhead is duplication of the address routing, but this should not be
2157 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2158 to be verified with callout=postmaster, and the main callout worked but the
2159 postmaster check failed, the verification correctly failed. However, if a
2160 subsequent RCPT command asked for sender verification *without* the
2161 postmaster check, incorrect caching caused this verification also to fail,
2164 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2165 it was not caching the DNS options (qualify_single, search_parents) that
2166 were used when the lookup failed. A subsequent lookup with different
2167 options therefore always gave the same answer, though there were cases
2168 where it should not have. (Example: a "domains = !$mx_any" option on a
2169 dnslookup router: the "domains" option is always processed without any
2170 widening, but the router might have qualify_single set.) Now Exim uses the
2171 cached value only when the same options are set.
2173 32. Added John Jetmore's "exipick" utility to the distribution.
2175 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2176 than a timeout (e.g. a certificate is required, and is not provided), an
2177 Exim server now closes the connection immediately. Previously it waited for
2178 the client to close - but if the client is SSL, it seems that they each
2179 wait for each other, leading to a delay before one of them times out.
2181 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2182 maintained 0.8.x compatibility because I don't think many are using it, and
2183 it is clearly obsolete.
2185 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2188 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2189 client certificate was expired. A simple patch fixes this, though I don't
2190 understand the full logic of why the verify callback is called multiple
2193 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2194 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2195 which causes problems with some clients (such as the Certicom SSL Plus
2196 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2197 disables the coutermeasure allowing Eudora to connect."
2199 38. Exim was not checking that a write() to a log file succeeded. This could
2200 lead to Bad Things if a log got too big, in particular if it hit a file
2201 size limit. Exim now panics and dies if it cannot write to a log file, just
2202 as it does if it cannot open a log file.
2204 39. Modified OS/Makefile-Linux so that it now contains
2206 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2208 The two -D definitions ensure that Exim is compiled with large file
2209 support, which makes it possible to handle log files that are bigger than
2212 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2213 instance) a domain was checked against a named list that involved a lookup,
2214 causing $domain_data to be set, then another domain was checked against the
2215 same list, then the first domain was re-checked, the value of $domain_data
2216 after the final check could be wrong. In particular, if the second check
2217 failed, it could be set empty. This bug probably also applied to
2220 41. The strip_trailing_dot option was not being applied to the address given
2221 with the -f command-line option.
2223 42. The code for reading a message's header from the spool was incrementing
2224 $received_count, but never initializing it. This meant that the value was
2225 incorrect (doubled) while delivering a message in the same process in which
2226 it was received. In the most common configuration of Exim, this never
2227 happens - a fresh exec is done - but it can happen when
2228 deliver_drop_privilege is set.
2230 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2231 it now includes up to 150 characters of the unexpected data in the log
2234 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2235 and building data strings. The size of both of these buffers was 10 000
2236 bytes - far larger than anybody would *ever* want, thought I. Needless to
2237 say, somebody hit the limit. I have increased the maximum line length to
2238 20 000 and the maximum data length of concatenated lines to 100 000. I have
2239 also fixed two bugs, because there was no checking on these buffers. Tsk,
2240 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2241 buffer is too small.
2243 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2244 lsearch lookups. Now it does.
2246 46. When parsing a route_list item in a manualroute router, a fixed-length
2247 buffer was used for the list of hosts. I made this 1024 bytes long,
2248 thinking that nobody would ever have a list of hosts that long. Wrong.
2249 Somebody had a whole pile of complicated expansion conditions, and the
2250 string was silently truncated, leading to an expansion error. It turns out
2251 that it is easier to change to an unlimited length (owing to other changes
2252 that have happened since this code was originally written) than to build
2253 structure for giving a limitation error. The length of the item that
2254 expands into the list of hosts is now unlimited.
2256 47. The lsearch lookup could not handle data where the length of text line was
2257 more than 4095 characters. Such lines were truncated, leading to shortened
2258 data being returned. It should now handle lines of any length.
2260 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2261 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2264 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2265 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2266 no longer happens in comment lines. A list of replacements is now placed
2267 at the head of all of the source files, except those whose only change is
2268 to replace PERL_COMMAND in the very first #! line.
2270 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2271 messages on the queue, with a bottom-up merge sort, using code contributed
2272 by Michael Haardt. This should make operations like -bp somewhat faster on
2273 large queues. It won't affect queue runners, except when queue_run_in_order
2276 51. Installed eximstats 1.31 in the distribution.
2278 52. Added support for SRV lookups to the dnslookup router.
2280 53. If an ACL referred to $message_body or $message_body_end, the value was not
2281 reset for any messages that followed in the same SMTP session.
2283 54. The store-handling optimization for building very long strings was not
2284 differentiating between the different store pools. I don't think this
2285 actually made any difference in practice, but I've tidied it.
2287 55. While running the routers to verify a sender address, $sender_address
2288 was still set to the sender address. This is wrong, because when routing to
2289 send a bounce to the sender, it would be empty. Therefore, I have changed
2290 it so that, while verifying a sender address, $sender_address is set to <>.
2291 (There is no change to what happens when verifying a recipient address.)
2293 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2294 target A or AAAA records (if not already returned) without resetting the
2295 qualify_single or search_parents options of the DNS resolver. These are
2296 inappropriate in this case because the targets of MX and SRV records must
2297 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2298 target that, when qualified, matched something in the local domain. These
2299 two options are now turned off when doing these lookups.
2301 57. It seems that at least some releases of Reiserfs (which does not have the
2302 concept of a fixed number of inodes) returns zero and not -1 for the
2303 number of available inodes. This interacted badly with check_spool_inodes,
2304 which assumed that -1 was the "no such thing" setting. What I have done is
2305 to check that the total number of inodes is greater than zero before doing
2306 the test of how many are available.
2308 58. When a "warn" ACL statement has a log_message modifier, the message is
2309 remembered, and not repeated. This is to avoid a lot of repetition when a
2310 message has many recipients that cause the same warning to be written.
2311 Howewer, Exim was preserving the list of already written lines for an
2312 entire SMTP session, which doesn't seem right. The memory is now reset if a
2313 new message is started.
2315 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2316 debugging output unless log_rewrite was also set.
2318 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2319 of (int)(handle) when we know that handle contains (void *)(-1).
2321 61. The Exim daemon panic-logs an error return when it closes the incoming
2322 connection. However "connection reset by peer" seems to be common, and
2323 isn't really an error worthy of noting specially, so that particular error
2326 62. When Exim is trying to find all the local interfaces, it used to panic and
2327 die if the ioctl to get the interface flags failed. However, it seems that
2328 on at least one OS (Solaris 9) it is possible to have an interface that is
2329 included in the list of interfaces, but for which you get a failure error
2330 for this call. This happens when the interface is not "plumbed" into a
2331 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2332 failure of the "get flags" call assumes that the interface is down.
2334 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2335 makes life easier for people who are doing arithmetic on fields extracted
2336 from dates, where you often get leading zeros that should not be
2337 interpreted as octal.
2339 64. Added qualify_domain to the redirect router, to override the global
2342 65. If a pathologically long header line contained very many addresses (the
2343 report of this problem mentioned 10 000) and each of them was rewritten,
2344 Exim could use up a very large amount of memory. (It kept on making new
2345 copies of the header line as it rewrote, and never released the old ones.)
2346 At the expense of a bit more processing, the header rewriting function has
2347 been changed so that it no longer eats memory in this way.
2349 66. The generation of the Received: header has been moved from the time that a
2350 message starts to be received, to the time that it finishes. The timestamp
2351 in the Received: header should now be very close to that of the <= log
2352 line. There are two side-effects of this change:
2354 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2355 logged header lines no longer include the local Received: line, because
2356 it has not yet been created. The same applies to a copy of the message
2357 that is returned to a non-SMTP sender when a message is rejected.
2359 (b) When a filter file is tested using -bf, no additional Received: header
2360 is added to the test message. After some thought, I decided that this
2363 This change does not affect the value of $received_for. It is still set
2364 after address rewriting, but before local_scan() is called.
2366 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2368 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2369 gave an unhelpful panic error message, and a defer error. I have managed to
2370 change this behaviour so that it now rejects any supplied certificate,
2371 which seems right, as the list of acceptable certificates is empty.
2373 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2374 gave an unhelpful defer error. I have not managed to make this reject any
2375 supplied certificates, but the error message it gives is "no certificate
2376 supplied", which is not helpful.
2378 70. exigrep's output now also includes lines that are not associated with any
2379 message, but which match the given pattern. Implemented by a patch from
2380 Martin Sluka, which also tidied up the Perl a bit.
2382 71. Recipient callout verification, like sender verification, was using <> in
2383 the MAIL FROM command. This isn't really the right thing, since the actual
2384 sender may affect whether the remote host accepts the recipient or not. I
2385 have changed it to use the actual sender in the callout; this means that
2386 the cache record is now keyed on a recipient/sender pair, not just the
2387 recipient address. There doesn't seem to be a real danger of callout loops,
2388 since a callout by the remote host to check the sender would use <>.
2389 [SEE ABOVE: changed after hitting problems.]
2391 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2392 temporary errors. However, in the case of such a code being given after
2393 the end of a data transmission (i.e. after ".") Exim was failing to write
2394 a retry record for the message. (Yes, there was some broken host that was
2395 actually sending 8xx at this point.)
2397 73. An unknown lookup type in a host list could cause Exim to panic-die when
2398 the list was checked. (An example that provoked this was putting <; in the
2399 middle of a list instead of at the start.) If this happened during a DATA
2400 ACL check, a -D file could be left lying around. This kind of configuration
2401 error no longer causes Exim to die; instead it causes a defer errror. The
2402 incident is still logged to the main and panic logs.
2404 74. Buglet left over from Exim 3 conversion. The message "too many messages
2405 in one connection" was written to the rejectlog but not the mainlog, except
2406 when address rewriting (yes!) was being logged.
2408 75. Added write_rejectlog option.
2410 76. When a system filter was run not as root (that is, when system_filter_user
2411 was set), the values of the $n variables were not being returned to the
2412 main process; thus, they were not subsequently available in the $sn
2415 77. Added +return_path_on_delivery log selector.
2417 78. A connection timeout was being treated differently from recipients deferred
2418 when testing hosts_max_try with a message that was older than the host's
2419 retry timeout. (The host should not be counted, thus allowing all hosts to
2420 be tried at least once before bouncing.) This may have been the cause of an
2421 occasionally reported bug whereby a message would remain on the queue
2422 longer than the retry timeout, but would be bounced if a delivery was
2423 forced. I say "may" because I never totally pinned down the problem;
2424 setting up timeout/retry tests is difficult. See also the next item.
2426 79. The ultimate address timeout was not being applied to errors that involved
2427 a combination of host plus message (for example, a timeout on a MAIL
2428 command). When an address resolved to a number of possible hosts, and they
2429 were not all tried for each delivery (e.g. because of hosts_max_try), a
2430 message could remain on the queue longer than the retry timeout.
2432 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2435 81. Fixed an obscure SMTP outgoing bug which required at least the following
2436 conditions: (a) there was another message waiting for the same server;
2437 (b) the server returned 5xx to all RCPT commands in the first message so
2438 that the message was not completed; (c) the server dropped the connection
2439 or gave a negative response to the RSET that Exim sends to abort the
2440 transaction. The observed case was a dropped connection after DATA that had
2441 been sent in pipelining mode. That is, the server had advertised PIPELINING
2442 but was not implementing it correctly. The effect of the bug was incorrect
2443 behaviour, such as trying another host, and this could lead to a crash.
2449 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2450 and daemon.c were passed as pointers to ints; they should have been
2451 pointers to socklen_t variables (which are typically unsigned ints).
2453 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2456 3. Fixed a really odd bug that affected only the testing scheme; patching a
2457 certain fixed string in the binary changed the value of another string that
2458 happened to be identical to the end of the original first string.
2460 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2461 name", it returns that address as the IP address. On some operating
2462 systems (e.g. Solaris), it also passes back the IP address string as the
2463 "host name". However, on others (e.g. Linux), it passes back an empty
2464 string. Exim wasn't checking for this, and was changing the host name to an
2465 empty string, assuming it had been canonicized.
2467 5. Although rare, it is permitted to have more than one PTR record for a given
2468 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2469 all the names associated with an address, because they do in Solaris.
2470 However, it seems that they do not in Linux for data that comes from the
2471 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2472 I found this out when I moved to a new Linux workstation and tried to run
2473 the Exim test suite.
2475 To get round this problem I have changed the code so that it now does its
2476 own call to the DNS to look up PTR records when searching for a host name.
2477 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2478 addresses that are only in /etc/hosts are still found.
2480 This behaviour is, however, controlled by an option called host_lookup_
2481 order, which defaults to "bydns:byaddr". If people want to use the other
2482 order, or indeed, just use one or the other means of lookup, they can
2483 specify it in this variable.
2485 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2486 some operating systems, this comes back from gethostbyaddr() as an empty
2487 string, and this is what Exim used to test for. However, it seems that in
2488 other systems, "." is yielded. Exim now tests for this case too.
2490 7. The values of check_spool_space and check_log_space are now held internally
2491 as a number of kilobytes instead of an absolute number of bytes. If a
2492 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2493 kilobyte. This means that much larger values can be stored.
2495 8. Exim monitor: an attempt to get the action menu when not actually pointing
2496 at a message produces an empty menu entitled "No message selected". This
2497 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2498 no entries in it ("Shell widget menu has zero width and/or height"). So I
2499 have added a single, blank menu entry in this case.
2501 9. Added ${quote_local_part.
2503 10. MIME decoding is now applied to the contents of Subject: header lines when
2506 11. Now that a reference to $sender_host_address automatically causes a reverse
2507 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2508 host lookup before query-style lookups in lists that might use this
2509 variable. This has therefore been abolished, and the "net-" prefix is no
2510 longer necessary for query-style lookups.
2512 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2513 have been a typo for LFLAGS, so it has been changed.
2515 13. The install script calls Exim with "-C /dev/null" in order to find the
2516 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2517 to be output. Howeve, since Exim outputs its version number before the
2518 error, it didn't break the script. It just looked ugly. I fixed this by
2519 always allowing "-C /dev/null" if the caller is root.
2521 14. Ignore overlarge ACL variable number when reading spool file - insurance
2522 against a later release with more variables having written the file.
2524 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2525 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2526 this, as well as the form without the "IPv6" on the front.
2528 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2529 OS/Makefile-Darwin file.
2531 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2532 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2534 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2535 variable. However, some OS use ints for inodes. I've added cast to long int
2536 to get rid of the compiler warning.
2538 19. I had forgotten to lock out "/../" in configuration file names when
2539 ALT_CONFIG_PREFIX was set.
2541 20. Routers used for verification do not need to specify transports. However,
2542 if such a router generated a host list, and callout was configured, Exim
2543 crashed, because it could not find a port number from the (non-existent)
2544 transport. It now assumes port 25 in this circumstance.
2546 21. Added the -t option to exigrep.
2548 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2549 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2550 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2551 them is set, LOOKUP_LSEARCH is forced.
2553 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2555 24. Added sender and host information to the "rejected by local_scan()" log
2556 line; previously there was no indication of these.
2558 25. Added .include_if_exists.
2560 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2561 Linux. It turns out that not all file systems support this. Apparently some
2562 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2563 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2564 not supported on the file descriptor, is now ignored when Exim is trying to
2565 sync a directory. This applies only to Linux.
2567 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2569 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2570 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2573 29. Added support for maildirsize files from supplied patch (modified a bit).
2575 30. The use of :fail: followed by an empty string could lead Exim to respond to
2576 sender verification failures with (e.g.):
2578 550 Verification failed for <xxx>
2579 550 Sender verify failed
2581 where the first response line was missing the '-' that indicates it is not
2582 the final line of the response.
2584 31. The loop for finding the name of the user that called Exim had a hardwired
2585 limit of 10; it now uses the value of finduser_retries, which is used for
2586 all other user lookups.
2588 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2591 33. Exim was neglecting to zero errno before one call of strtol() when
2592 expanding a string and expecting an integer value. On some systems this
2593 resulted in spurious "integer overflow" errors. Also, it was casting the
2594 result into an int without checking.
2596 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2597 did not work. The code looks as if it has *never* worked, though it appears
2598 to have been documented since at least releast 1.62. I have made it work.
2600 35. The "timeout_DNS" error in retry rules, also documented since at least
2601 1.62, also never worked. As it isn't clear exactly what this means, and
2602 clearly it isn't a major issue, I have abolished the feature by treating it
2603 as "timeout", and writing a warning to the main and panic logs.
2605 36. The display of retry rules for -brt wasn't always showing the error code
2608 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2609 timeout_connect_A, timeout_connect_MX.
2611 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2612 to the empty sender.
2614 39. The daemon was not analysing the content of -oX till after it had closed
2615 stderr and disconnected from the controlling terminal. This meant that any
2616 syntax errors were only noted on the panic log, and the return code from
2617 the command was 0. By re-arranging the code a little, I've made the
2618 decoding happen first, so such errors now appear on stderr, and the return
2619 code is 1. However, the actual setting up of the sockets still happens in
2620 the disconnected process, so errors there are still only recorded on the
2623 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2624 connections (as happens on some IP stacks) was logged at start up time as
2625 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2626 it from "IPv6 and IPv4", which means that two separate sockets are being
2629 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2630 says whether AF_INET or AF_INET6 was passed as an argument.
2632 42. Exiwhat output was messed up when time zones were included in log
2635 43. Exiwhat now gives more information about the daemon's listening ports,
2636 and whether -tls-on-connect was used.
2638 44. The "port" option of the smtp transport is now expanded.
2640 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2641 silently ignored. Now an error message is written to the main and panic
2644 46. There's a new ACL modifier called "logwrite" which writes to a log file
2645 as soon as it is encountered.
2647 47. Added $local_user_uid and $local_user_gid at routing time.
2649 48. Exim crashed when trying to verify a sender address that was being
2652 49. Exim was recognizing only a space character after ".include". It now also
2653 recognizes a tab character.
2655 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2656 extracting the relevant information from the specification. The man page no
2657 longer contains scrambled data for the -d option, and I've added a section
2658 at the front about calling Exim under different names.
2660 51. Added "extra_headers" argument to the "mail" command in filter files.
2662 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2665 53. Installed eximstats 1.29.
2667 54. Added transport_filter_timeout as a generic transport option.
2669 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2670 Cc: header lines. This was required by RFC 822, but it not required by RFC
2673 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2674 incoming messages that did not have them. Now it does so only if the
2675 message originates locally, that is, if there is no associated remote host
2676 address. When Resent- header lines are present, this applies to the Resent-
2677 lines rather than the non-Resent- lines.
2679 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2680 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2682 58. Messages for configuration errors now include the name of the main
2683 configuration file - useful now that there may be more than one file in a
2684 list (.included file names were always shown).
2686 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2687 for those rare installations that do not start the daemon as root or run it
2688 setuid root. I've cut out the call to initgroups() if the daemon is not
2691 60. The Exim user and group can now be bound into the binary as text strings
2692 that are looked up at the start of Exim's processing.
2694 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2696 62. Added $mailstore_basename variable.
2698 63. Installed patch to sieve.c from Michael Haardt.
2700 64. When Exim failed to open the panic log after failing to open the main log,
2701 the original message it was trying to log was written to stderr and debug
2702 output, but if they were not available (the usual case in production), it
2703 was lost. Now it is written to syslog before the two lines that record the
2704 failures to open the logs.
2706 65. Users' Exim filters run in subprocesses under the user's uid. It is
2707 possible for a "deliver" command or an alias in a "personal" command to
2708 provoke an address rewrite. If logging of address rewriting is configured,
2709 this fails because the process is not running as root or exim. There may be
2710 a better way of dealing with this, but for the moment (because 4.30 needs
2711 to be released), I have disabled address rewrite logging when running a
2712 filter in a non-root, non-exim process.
2718 1. The buildconfig auxiliary program wasn't quoting the value set for
2719 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2720 not defined. This bug was masked in 4.22 by the effect that was fixed in
2723 2. Some messages that were rejected after a message id was allocated were
2724 shown as "incomplete" by exigrep. It no longer does this for messages that
2725 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2727 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2728 have allow_domain_literals set, the ID did not get logged in the <= line.
2729 Domain literals are now always recognized in Message-ID: header lines.
2731 4. The first argument for a ${extract expansion item is the key name or field
2732 number. Leading and trailing spaces in this item were not being ignored,
2733 causing some misleading effects.
2735 5. When deliver_drop_privilege was set, single queue runner processes started
2736 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2737 same command in the process it spins off) were not dropping privilege.
2739 6. When the daemon running as "exim" started a queue runner, it always
2740 re-executed Exim in the spun-off process. This is a waste of effort when
2741 deliver_drop_privilege is set. The new process now just calls the
2742 queue-runner function directly.
2748 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2751 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2752 ignored. Though the use of -oP was forcing the writing of a pid file, it
2753 was always written to the default place.
2755 3. If the message "no IP address found for host xxxx" is generated during
2756 incoming verification, it is now followed by identification of the incoming
2757 connection (so you can more easily find what provoked it).
2759 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2761 5. Added some features to "harden" Exim a bit more against certain attacks:
2763 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2764 be put in Local/Makefile. This is like the never_users runtime option,
2765 but it cannot be overridden. The default setting is "root".
2767 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2768 prefix string with which any file named in a -C command line option
2771 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2772 is retained for -C and -D only if the caller of Exim is root. Without
2773 it, the exim user may also use -C and -D and retain privilege.
2775 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2776 command line option is disabled.
2778 6. Macro names set by the -D option must start with an upper case letter, just
2779 like macro names defined in the configuration file.
2781 7. Added "dereference=" facility to LDAP.
2783 8. Two instances of the typo "uknown" in the source files are fixed.
2785 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2786 the Configure-Makefile script screwed up while processing it.
2788 10. Incorporated PCRE 4.4.
2790 11. The SMTP synchronization check was not operating right at the start of an
2791 SMTP session. For example, it could not catch a HELO sent before the client
2792 waited for the greeting. There is now a check for outstanding input at the
2793 point when the greeting is written. Because of the duplex, asynchronous
2794 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2795 way, but not yet received, when the check is performed.
2797 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2798 on TCP/IP sockets, because this apparently causes some broken clients to
2801 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2802 unchanged) from the Cygwin maintainer.
2804 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2805 was supported instead of testing for mbx. Effectively a typo.
2807 15. The spa authenticator server code was not checking that the input it
2808 received was valid base64.
2810 16. The debug output line for the "set" modifier in ACLs was not showing the
2811 name of the variable that was being set.
2813 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2815 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2816 lookup on its own. Something else was needed to trigger the lookup. For
2817 example, a match in host_lookup or the need for a host name in a host list.
2818 Now, if $sender_host_name is referenced and the host name has not yet been
2819 looked up, a lookup is performed. If the lookup fails, the variable remains
2820 empty, and $host_lookup_failed is set to "1".
2822 19. Added "eqi" as a case-independent comparison operator.
2824 20. The saslauthd authentication condition could segfault if neither service
2825 nor realm was specified.
2827 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2828 error message that was logged was misleading, and incoming SMTP
2829 connections were dropped. The message is now more accurate, and temporary
2830 errors are given to SMTP connections.
2832 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2833 (except RSET) with a 421 error, until QUIT is received. However, it was
2834 failing to send a response to QUIT.
2836 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2837 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2838 ACL is now run after the helo_try_verify_hosts code.
2840 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2841 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2842 case-independent, so other case variants are also recognized). Apparently
2843 some systems use these upper case variants.
2845 25. If more than two messages were waiting for the same host, and a transport
2846 filter was specified for the transport, Exim sent two messages over the
2847 same TCP/IP connection, and then failed with "socket operation on non-
2848 socket" when it tried to send the third.
2850 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2852 27. The extern definition of crypt16() in expand.c was not being excluded when
2853 the OS had its own crypt16() function.
2855 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2856 as a preferred synonym for return_size_limit, both as an option and as an
2859 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2861 30. Changed the default configuration ACL to relax the local part checking rule
2862 for addresses that are not in any local domains. For these addresses,
2863 slashes and pipe symbols are allowed within local parts, but the sequence
2864 /../ is explicitly forbidden.
2866 31. SPA server authentication was not clearing the challenge buffer before
2869 32. log_message in a "warn" ACL statement was writing to the reject log as
2870 well as to the main log, which contradicts the documentation and doesn't
2871 seem right (because no rejection is happening). So I have stopped it.
2873 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2874 However, I am unable to do any testing of this.
2876 34. Fixed an infelicity in the appendfile transport. When checking directories
2877 for a mailbox, to see if any needed to be created, it was accidentally
2878 using path names with one or more superfluous leading slashes; tracing
2879 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2881 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2882 contents are added to the log line that is written for every discarded
2883 recipient. (Previously a log_message setting was ignored.)
2885 36. The ${quote: operator now quotes the string if it is empty.
2887 37. The install script runs exim in order to find its version number. If for
2888 some reason other than non-existence or emptiness, which it checks, it
2889 could not run './exim', it was installing it with an empty version number,
2890 i.e. as "exim-". This error state is now caught, and the installation is
2893 38. An argument was missing from the function that creates an error message
2894 when Exim fails to connect to the socket for saslauthd authentication.
2895 This could cause Exim to crash, or give a corrupted message.
2897 39. Added isip, isip4, and isip6 to ${if conditions.
2899 40. The ACL variables $acl_xx are now saved with the message, and can be
2900 accessed later in routers, transports, and filters.
2902 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2903 strings in the file are not string-expanded.
2905 42. If a MAIL command specified a SIZE value that was too large to fit into an
2906 int variable, the check against message_size_limit failed. Such values are
2907 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2908 day this will have to be increased, but I don't think I want to be around
2909 when emails are that large.
2916 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2917 iconv() is not standard in FreeBSD.
2919 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2920 IPv6 enabled. The observed symptom was a segmentation fault on return from
2921 the function os_common_find_running_interfaces() in src/os.c.
2923 3. In the check_special_case() function in daemon.c I had used "errno" as an
2924 argument name, which causes warnings on some systems. This was basically a
2925 typo, since it was named "eno" in the comments!
2927 4. The code that waits for the clock to tick (at a resolution of some fraction
2928 of a second) so as to ensure message-id uniqueness was always waiting for
2929 at least one whole tick, when it could have waited for less. [This is
2930 almost certainly not relevant at current processor speeds, where it is
2931 unlikely to ever wait at all. But we try to future-proof.]
2933 5. The function that sleeps for a time interval that includes fractions of a
2934 second contained a race. It did not block SIGALRM between setting the
2935 timer, and suspending (a couple of lines later). If the interval was short
2936 and the sigsuspend() was delayed until after it had expired, the suspension
2937 never ended. On busy systems this could lead to processes getting stuck for
2940 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2941 process, before it forks any delivery processes. The open lookup caching
2942 mechanism meant that the open file or database connection was passed into
2943 the delivery process. The problem was that delivery processes always tidy
2944 up cached lookup data. This could cause a problem for the next delivery
2945 process started by the queue runner, because the external queue runner
2946 process does not know about the closure. So the next delivery process
2947 still has data in the lookup cache. In the case of a file lookup, there was
2948 no problem because closing a file descriptor in a subprocess doesn't affect
2949 the parent. However, if the lookup was caching a connection to a database,
2950 the connection was closed, and the second delivery process was likely to
2951 see errors such as "PGSQL: query failed: server closed the connection
2952 unexpectedly". The problem has been fixed by closing all cached lookups
2953 in a queue runner before running a delivery process.
2955 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2956 seem to have the "const" qualifier which it has on other OS. I've
2959 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2960 *of the same type* (client or server) with the same public name that an
2961 error should be diagnosed.
2963 9. When Exim looked up a host name for an IP address, but failed to find the
2964 original IP address when looking up the host name (a safety check), it
2965 output the message "<ip address> does not match any IP for NULL", which was
2966 confusing, to say the least. The bug was that the host name should have
2967 appeared instead of "NULL".
2969 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2970 user that is built into the binary, and the -C or -D options is used, root
2971 privilege is dropped before the configuration file is read. In addition,
2972 logging is switched to stderr instead of the normal log files. If the
2973 configuration then re-defines the Exim user, the unprivileged environment
2974 is probably not what is expected, so Exim logs a panic warning message (but
2977 However, if deliver_drop_privilege is set, the unprivileged state may well
2978 be exactly what is intended, so the warning has been cut out in that case,
2979 and Exim is allowed to try to write to its normal log files.
2985 1. smtp_return_error_details was not giving details for temporary sender
2986 or receiver verification errors.
2988 2. Diagnose a configuration error if two authenticators have the same public
2991 3. Exim used not to create the message log file for a message until the first
2992 delivery attempt. This could be confusing when incoming messages were held
2993 for policy or load reasons. The message log file is now created at the time
2994 the message is received, and an initial "Received" line is written to it.
2996 4. The automatically generated man page for command line options had a minor
2997 bug that caused no ill effects; however, a more serious problem was that
2998 the procedure for building the man page automatically didn't always
2999 operate. Consequently, release 4.20 contains an out-of-date version. This
3000 shouldn't happen again.
3002 5. When building Exim with embedded Perl support, the script that builds the
3003 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3004 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3006 6. The freeze_tell option was not being used for messages that were frozen on
3007 arrival, either by an ACL or by local_scan().
3009 7. Added the smtp_incomplete_transaction log selector.
3011 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3012 was accepting AUTH without a new EHLO.
3014 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3015 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3018 10. Logging of TCP/IP connections (when configured) now happens in the main
3019 daemon process instead of the child process, so that the TCP/IP connection
3020 count is more accurate (but it can never be perfect).
3022 11. The use of "drop" in a nested ACL was not being handled correctly in the
3023 outer ACL. Now, if condition failure induced by the nested "drop" causes
3024 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3025 or "require"), the connection is dropped.
3027 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3028 that yield "discard" can now be used with an "accept" or a "discard" verb,
3029 but an error is generated for any others (because I can't see a useful way
3030 to define what should happen).
3032 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3033 are now processed in the same way as lines in the Exim configuration file.
3034 In particular, continuation lines are supported.
3036 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3038 15. Added -ti meaning -t -i.
3040 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3041 domains, and warn by logging if others are found.
3043 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3044 returned from GIFCONF when Exim is trying to find the list of interfaces on
3045 a host. The code in os.c has been modified to copy each ifreq to an aligned
3046 structure in all cases.
3048 Also, in some cases, the returned ifreq's were being copied to a 'struct
3049 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3050 means the last couple of bytes of an IPv6 address could be chopped if the
3051 ifreq contained only a normal sockaddr (14 bytes storage).
3053 18. Named domain lists were not supported in the hosts_treat_as_local option.
3054 An entry such as +xxxx was not recognized, and was treated as a literal
3057 19. Ensure that header lines added by a DATA ACL are included in the reject log
3058 if the ACL subsequently rejects the message.
3060 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3061 MD5 (which is deprecated).
3063 21. When testing a filter file using -bf, Exim was writing a message when it
3064 took the sender from a "From " line in the message, but it was not doing so
3065 when it took $return_path from a Return-Path: header line. It now does.
3067 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3068 with a valid header line field name (a series of printing characters
3069 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3071 23. Changed "disc" in the source to "disk" to conform to the documentation and
3072 the book and for uniformity.
3074 24. Ignore Sendmail's -Ooption=value command line item.
3076 25. When execve() failed while trying to run a command in a pipe transport,
3077 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3078 could be confused with a return value of 69 from the command itself. This
3079 has been changed to 127, the value the shell returns if it is asked to run
3080 a non-existent command. The wording for the related log line suggests a
3081 non-existent command as the problem.
3083 26. If received_header_text expands to an empty string, do not add a Received:
3084 header line to the message. (Well, it adds a token one on the spool, but
3085 marks it "old" so that it doesn't get used or transmitted.)
3087 27. Installed eximstats 1.28 (addition of -nt option).
3089 28. There was no check for failure on the call to getsockname() in the daemon
3090 code. This can fail if there is a shortage of resources on the system, with
3091 ENOMEM, for example. A temporary error is now given on failure.
3093 29. Contrary to the C standard, it seems that in some environments, the
3094 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3095 program. Exim now does this explicitly; it affects the formatting of
3096 timestamps using strftime().
3098 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3099 complaints. I've now initialized all the variables, to avoid this.
3101 32. Header lines added by a system filter were not being "seen" during
3102 transport-time rewrites.
3104 33. The info_callback() function passed to OpenSSL is set up with type void
3105 (*)(SSL *, int, int), as described somewhere. However, when calling the
3106 function (actually a macro) that sets it up, the type void(*)() is
3107 expected. I've put in a cast to prevent warnings from picky compilers.
3109 34. If a DNS black list lookup found a CNAME record, but there were no A
3110 records associated with the domain it pointed at, Exim crashed.
3112 35. If a DNS black list lookup returned more than one A record, Exim ignored
3113 all but the first. It now scans all returned addresses if a particular IP
3114 value is being sought. In this situation, the contents of the
3115 $dnslist_value variable are a list of all the addresses, separated by a
3118 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3119 to accept a host name and all its aliases if the forward lookup for any of
3120 them yielded the IP address of the incoming connection. Now it accepts only
3121 those names whose forward lookup yields the correct IP address. Any other
3122 names are discarded. This closes a loophole whereby a rogue DNS
3123 administrator could create reverse DNS records to break through a
3124 wildcarded host restriction in an ACL.
3126 37. If a user filter or a system filter that ran in a subprocess used any of
3127 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3128 the wrong values were passed to the pipe command ($thisaddress had the
3129 value of $0, $0 had the value of $1, etc). This bug was introduced by
3130 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3132 38. Improved the line breaking for long SMTP error messages from ACLs.
3133 Previously, if there was no break point between 40 and 75 characters, Exim
3134 left the rest of the message alone. Two changes have been made: (a) I've
3135 reduced the minimum length to 35 characters; (b) if it can't find a break
3136 point between 35 and 75 characters, it looks ahead and uses the first one
3137 that it finds. This may give the occasional overlong line, but at least the
3138 remaining text gets split now.
3140 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3141 file descriptors might be low, and that setting 1000 would always raise it.
3142 It turns out that in some environments, the limit is already over 1000 and
3143 that lowering it causes trouble. So now Exim takes care not to decrease it.
3145 40. When delivering a message, the value of $return_path is set to $sender_
3146 address at the start of routing (routers may change the value). By an
3147 oversight, this default was not being set up when an address was tested by
3148 -bt or -bv, which affected the outcome if any router or filter referred to
3151 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3152 log only when "message" or "log_message" are set. However, if one of the
3153 conditions was an address verification, or a call to a nested ACL, the
3154 messages generated by the underlying test were being passed through. This
3155 no longer happens. The underlying message is available in $acl_verify_
3156 message for both "message" and "log_message" expansions, so it can be
3157 passed through if needed.
3159 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3160 new expansion $bh_ to give the encoded byte string without charset
3161 translation. Translation happens only if iconv() is available; HAVE_ICONV
3162 indicates this at build time. HEADERS_CHARSET gives the charset to
3163 translate to; headers_charset can change it in the configuration, and
3164 "headers charset" can change it in an individual filter file.
3166 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3167 that creates RFC 2047 encoded "words" labels them as that charset instead
3168 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3169 expansion operator; (ii) when Exim creates a From: line for a local
3170 message; (iii) when a header line is rewritten to include a "phrase" part.
3172 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3173 buggy, causing it to skip the first lines of messages whose message ID
3174 ended in 'D'. This would not have bitten before Exim release 4.14, because
3175 message IDs were unlikely to end in 'D' before then. The effect was to have
3176 incorrect size information for certain domains.
3178 45. #include "config.h" was missing at the start of the crypt16.c module. This
3179 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3182 46. If there was a timeout during a "random" callout check, Exim treated it as
3183 a failure of the random address, and carried on sending RSET and the real
3184 address. If the delay was just some slowness somewhere, the response to the
3185 original RCPT would be taken as a response to RSET and so on, causing
3186 mayhem of various kinds.
3188 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3189 when I implemented it. It didn't allow for the fact that some option values
3190 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3191 the right test for positive values.
3193 48. Domain names in DNS records are case-independent. Exim always looks them up
3194 in lower case. Some resolvers return domain names in exactly the case they
3195 appear in the zone file, that is, they may contain uppercase letters. Not
3196 all resolvers do this - some return always lower case. Exim was treating a
3197 change of case by a resolver as a change of domain, similar to a widening
3198 of a domain abbreviation. This triggered its re-routing code and so it was
3199 trying to route what was effectively the same domain again. This normally
3200 caused routing to fail (because the router wouldn't handle the domain
3201 twice). Now Exim checks for this case specially, and just changes the
3202 casing of the domain that it ultimately uses when it transmits the message
3205 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3208 50. If a filter generated a file delivery with a non-absolute name (possible if
3209 no home directory exists for the router), the forbid_file option was not
3212 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3213 the existing equality matching.
3215 52. Exim was using ints instead of ino_t variables in some places where it was
3216 dealing with inode numbers.
3218 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3219 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3220 called TMPDIR, and if it finds it is different, it changes its value.
3222 54. The smtp_printf() function is now made available to local_scan() so
3223 additional output lines can be written before returning. There is also an
3224 smtp_fflush() function to enable the detection of a dropped connection.
3225 The variables smtp_input and smtp_batched_input are exported to
3228 55. Changed the default runtime configuration: the message "Unknown user"
3229 has been removed from the ACL, and instead placed on the localuser router,
3230 using the cannot_route_message feature. This means that any verification
3231 failures that generate their own messages won't get overridden. Similarly,
3232 the "Unrouteable address" message that was in the ACL for unverifiable
3233 relay addresses has also been removed.
3235 56. Added hosts_avoid_esmtp to the smtp transport.
3237 57. The exicyclog script was not checking for the esoteric option
3238 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3239 will work only if exicyclog is run under the appropriate euid.
3241 58. Following a discussion on the list, the rules by which Exim recognises line
3242 endings on incoming messages have been changed. The -dropcr and drop_cr
3243 options are now no-ops, retained only for backwards compatibility. The
3244 following line terminators are recognized: LF CRLF CR. However, special
3245 processing applies to CR:
3247 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3248 nor a local message in the state where . is a terminator.
3250 (ii) If a bare CR is encountered in a header line, an extra space is added
3251 after the line terminator so as not to end the header. The reasoning
3252 behind this is that bare CRs in header lines are most likely either
3253 to be mistakes, or people trying to play silly games.
3255 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3256 was being incorrectly given as 18 bytes larger than it should have been.
3257 This is a VOB (very old bug).
3259 60. This may never have affected anything current, but just in case it has:
3260 When the local host is found other than at the start of a list of hosts,
3261 the local host, those with the same MX, and any that follow, are discarded.
3262 When the list in question was part of a longer list of hosts, the following
3263 hosts (not currently being processed) were also being discarded. This no
3264 longer happens. I'm not sure if this situation could ever has previously
3267 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3270 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3271 header. This is something that is recommended in a new Internet Draft, and
3272 is something that is documented as being done by Sendmail. There are two
3273 possible values. For messages generated by the autoreply transport, Exim
3276 Auto-Submitted: auto-replied
3278 whereas for all other generated messages (e.g. bounces) it adds
3280 Auto-Submitted: auto-generated
3282 63. The "personal" condition in filters now includes a test for the
3283 Auto-Submitted: header. If it contains the string "auto-" the message it
3284 not considered personal.
3286 64. Added rcpt_include_affixes as a generic transport option.
3288 65. Added queue_only_override (default true).
3290 66. Added the syslog_duplication option.
3292 67. If what should have been the first header line of a message consisted of
3293 a space followed by a colon, Exim was mis-interpreting it as a header line.
3294 It isn't of course - it is syntactically invalid and should therefore be
3295 treated as the start of the message body. The misbehaviour could have
3296 caused a number of strange effects, including loss of data in subsequent
3297 header lines, and spool format errors.
3299 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3300 client host had authenticated. This control can now be exercised by an ACL
3301 for more flexibility.
3303 69. By default, callouts do not happen when testing with -bh. There is now a
3304 variant, -bhc, which does actually run the callout code, including
3305 consulting and updating the callout cache.
3307 70. Added support for saslauthd authentication, courtesy of Alexander
3310 71. If statvfs() failed on the spool or log directories while checking their
3311 size for availability, Exim confusingly gave the error "space shortage".
3312 Furthermore, in debugging mode it crashed with a floating point exception.
3313 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3314 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3315 serious problem, Exim now writes to the main and panic logs when this
3316 happens, with details of the failure. It then refuses to accept the
3317 incoming message, giving the message "spool directory problem" or "log
3318 directory problem" with a 421 code for SMTP messages.
3320 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3321 0, 1, and 2 exist, because some OS complain for execs without them (see
3322 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3323 descriptors. However, the code omitted to check that the open succeeded,
3324 causing mysterious errors if for some reason the permissions on /dev/null
3325 got screwed. Now Exim writes a message to the main and panic logs, and
3326 bombs out if it can't open /dev/null.
3328 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3329 interact so that it is all more flexible. It is supposed to remain
3330 backwards compatible. Also added extra_local_interfaces.
3332 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3333 to bomb out with an assertion failure - to the client this appears as a
3334 connection drop. This problem occurs in the part of the code that was taken
3335 from the Samba project. Fortunately, the assertion is in a very simple
3336 function, so I have fixed this by reproducing the function inline in the
3337 one place where it is called, and arranging for authentication to fail
3338 instead of killing the process with assert().
3340 75. The SPA client code was not working when the server requested OEM rather
3341 than Unicode encoding.
3343 76. Added code to make require_files with a specific uid setting more usable in
3344 the case where statting the file as root fails - usually a non-root-mounted
3345 NFS file system. When this happens and the failure is EACCES, Exim now
3346 forks a subprocess and does the per-uid checking as the relevant uid.
3348 77. Added process_log_path.
3350 78. If log_file_path was not explicitly set, a setting of check_log_space or
3351 check_log_inodes was ignored.
3353 79. If a space check for the spool or log partitions fails, the incident is now
3354 logged. Of course, in the latter case the data may get lost...
3356 80. Added the %p formatting code to string_format() so that it can be used to
3357 print addresses in debug_print(). Adjusted all the address printing in the
3358 debugging in store.c to use %p rather than %d.
3360 81. There was a concern that a line of code in smtp_in.c could overflow a
3361 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3362 initially expressed, the concern was not well-founded, because trailing
3363 spaces are removed early. However, if the trailing spaces were followed by
3364 a NULL, they did not get removed, so the overflow was possible. Two fixes
3367 (a) I re-wrote the offending code in a cleaner fashion.
3368 (b) If an incoming SMTP command contains a NULL character, it is rejected
3371 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3372 runs initgroups(), so that if the Exim user is in any additional groups,
3373 they will be used during message reception.
3379 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob