1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.233 2005/09/15 16:02:07 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.
206 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
208 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
209 command itself ended in error, the underlying message about the transport
210 filter timeout was being overwritten with the pipe command error. Now the
211 underlying error message should be appended to the second error message.
213 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
215 PH/02 Reduce the amount of output that "make" produces by default. Full output
216 can still be requested.
218 PH/03 The warning log line about a condition test deferring for a "warn" verb
219 was being output only once per connection, rather than after each
220 occurrence (because it was using the same function as for successful
221 "warn" verbs). This seems wrong, so I have changed it.
223 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
224 it should not have, which might have caused a crash in the right
225 circumstances, but probably never did.
227 PH/04 Installed a modified version of Tony Finch's patch to make submission
228 mode fix the return path as well as the Sender: header line, and to
229 add a /name= option so that you can make the user's friendly name appear
232 TF/03 Added the control = fakedefer ACL modifier.
234 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
235 Mark Lowes for thorough testing.
237 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
239 TK/03 Merged latest SRS patch from Miles Wilton.
241 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
242 with the definition in sysexits.h (which is #included earlier).
243 Fortunately, Exim does not actually use EX_OK. The code used to try to
244 preserve the sysexits.h value, by assumimg that macro definitions were
245 scanned for macro replacements. I have been disabused of this notion,
246 so now the code just undefines EX_OK before #including unistd.h.
248 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
249 in the smtp transport. When a block could not be written in a single
250 write() function, the timeout was being re-applied to each part-write.
251 This seems wrong - if the receiver was accepting one byte at a time it
252 would take for ever. The timeout is now adjusted when this happens. It
253 doesn't have to be particularly precise.
255 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
256 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
258 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
259 without a domain if the check to <postmaster@domain> fails.
261 SC/01 Eximstats: added -xls and the ability to specify output files
262 (patch written by Frank Heydlauf).
264 SC/02 Eximstats: use FileHandles for outputing results.
266 SC/03 Eximstats: allow any combination of xls, txt, and html output.
268 SC/04 Eximstats: fixed display of large numbers with -nvr option
270 SC/05 Eximstats: fixed merging of reports with empty tables.
272 SC/06 Eximstats: added the -include_original_destination flag
274 SC/07 Eximstats: removed tabs and trailing whitespace.
276 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
278 TK/06 MBOX spool code: Add real "From " MBOX separator line
279 so the .eml file is really in mbox format (even though
280 most programs do not really care). Patch from Alex Miller.
282 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
283 The latter is generated from $received_to and is only set if the
284 message has one envelope recipient. SA can use these headers,
285 obviously out-of-the-box. Patch from Alex Miller.
287 PH/08 The ${def test on a variable was returning false if the variable's
288 value was "0", contrary to what the specification has always said!
289 The result should be true unless the variable is empty.
291 PH/09 The syntax error of a character other than { following "${if
292 def:variable_name" (after optional whitespace) was not being diagnosed.
293 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
294 accidental colon was present, for example, could give incorrect results.
296 PH/10 Tidied the code in a number of places where the st_size field of a stat()
297 result is used (not including appendfile, where other changes are about
300 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
301 This involved changing a lot of size variables from int to off_t. It
302 should work with maildirs and everything.
304 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
305 spamd dying while we are connected to it.
307 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
310 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
311 function for BDB 4.3.
313 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
315 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
316 the sources for PCRE have been re-arranged and more files are now
319 PH/15 The code I had for printing potentially long long variables in PH/11
320 above was not the best (it lost precision). The length of off_t variables
321 is now inspected at build time, and an appropriate printing format (%ld
322 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
323 to be "long long int" or "long int". This is needed for the internal
324 formatting function string_vformat().
326 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
327 the configuration file to be ":syslog", then the script "guesses" where
328 the logs files are, rather than using the compiled in default. In our
329 case the guess is not the same as the compiled default, so the script
330 suddenly stopped working when I started to use syslog. The patch checks
331 to see if log_file_path is "". If so, it attempts to read it from exim
332 with no configuration file to get the compiled in version, before it
333 falls back to the previous guessing code."
335 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
336 implementing BATV in an Exim configuration. See NewStuff for the gory
339 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
340 Makefile that are specific to HP-UX.
342 PH/18 If the "use_postmaster" option was set for a recipient callout together
343 with the "random" option, the postmaster address was used as the MAIL
344 FROM address for the random test, but not for the subsequent recipient
345 test. It is now used for both.
347 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
348 patch removes a few documentation additions to RFC 3028, because the
349 latest draft now contains them. It adds the new en;ascii-case comparator
350 and a new error check for 8bit text in MIME parts. Comparator and
351 require names are now matched exactly. I enabled the subaddress
352 extension, but it is not well tested yet (read: it works for me)."
354 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
355 rework some of the code of TK/09 above to avoid the hardwired use of
356 "%lld" and "long long". Replaced the call to snprintf() with a call to
359 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
360 records point to non-existent hosts", "retry timeout exceeded", and
361 "retry time not reached for any host after a long failure period".
363 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
364 experimental DomainKeys support:
366 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
367 (2) On an error such as an illegally used "control", the wrong name for
368 the control was given.
370 These problems did NOT occur unless DomainKeys support was compiled.
372 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
374 PH/24 Added ${if match_ip condition.
376 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
377 databases so that it will be absolutely obvious if a crash occurs in the
378 DB library. This is a regular occurrence (often caused by mis-matched
381 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
382 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
383 picked up on a user's system that detects such things. There doesn't seem
384 to be a gcc warning option for this - only an attribute that has to be
385 put on the function's prototype. It seems that in Fedora Core 4 they have
386 set this on a number of new functions. No doubt there will be more in due
389 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
390 specify a transport. However, if an address that was verified by such a
391 router was the subject of a callout, Exim crashed because it tried to
392 read the rcpt_include_affixes from the non-existent transport. Now it
393 just assumes that the setting of that option is false. This bug was
394 introduced by 4.51/PH/31.
396 PH/28 Changed -d+all to exclude +memory, because that information is very
397 rarely of interest, but it makes the output a lot bigger. People tend to
398 do -d+all out of habit.
400 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
401 code in os-type was giving problems when libc.so lives in lib64, like on
404 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
405 aren't the modern standard, and it seems that some systems' include files
406 don't always have them. Exim was already checking for some of the newer
407 ones like T_AAAA, and defining it itself. I've added checks for all the
408 record types that Exim uses.
410 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
411 not automatically generating a new one, as it is supposed to. This
412 prevented TLS from working. If the file did exist, but contained invalid
413 data, a new version was generated, as expected. It was only the case of a
414 non-existent file that was broken.
416 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
417 with a change in libdomainkeys > 0.64.
419 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
420 from DNS. If the selector record carries the flag, it now has
421 precedence over the domain-wide flag.
423 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
425 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
426 the use of a port name that isn't defined in /etc/services) occurred, the
427 message was deferred as in a normal delivery, and thus remained on the
428 spool, instead of being failed because of the mua_wrapper setting. This
429 is now fixed, and I tidied up some of the mua_wrapper messages at the
432 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
433 the messages in a hash of arrays rather than using individual hashes.
434 This is a bit cleaner and results in dramatic memory savings, albeit
435 at a slight CPU cost.
437 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
438 as requested by Marc Sherman.
440 SC/10 Eximstats: added histograms for user specified patterns as requested
443 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
445 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
446 fopen() in the content-scanning modules that did not already have it.
452 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
453 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
455 TK/02 Fix ACL "control" statement not being available in MIME ACL.
457 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
459 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
460 to test Sieve filters that use "vacation".
462 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
463 that changes the way the GnuTLS parameters are stored in the cache file.
464 The new format can be generated externally. For backward compatibility,
465 if the data in the cache doesn't make sense, Exim assumes it has read an
466 old-format file, and it generates new data and writes a new file. This
467 means that you can't go back to an older release without removing the
470 PH/03 A redirect router that has both "unseen" and "one_time" set does not
471 work if there are any delivery delays because "one_time" forces the
472 parent to be marked "delivered", so its unseen clone is never tried
473 again. For this reason, Exim now forbids the simultaneous setting of
476 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
477 redirected to themselves ("homonym" addresses). Read the long ChangeLog
478 entry if you want to know the details. The fix, however, neglected to
479 consider the case when local delivery batching is involved. The test for
480 "previously delivered" was not happening when checking to see if an
481 address could be batched with a previous (undelivered) one; under
482 certain circumstances this could lead to multiple deliveries to the same
485 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
486 in its include files, and this causes problems building Exim.
488 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
489 header_syntax) but Exim was just ignoring anything given after a slash.
490 In particular, this caused confusion with an attempt to use "verify =
491 reverse_host_lookup/defer_ok". An error is now given when options are
492 supplied for verify items that do not have them. (Maybe reverse_host_
493 lookup should have a defer_ok option, but that's a different point.)
495 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
496 defined by RFC 821) to 2048, because there were problems with some AUTH
497 commands, and RFC 1869 says the size should be increased for extended
498 SMTP commands that take arguments.
500 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
503 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
504 "unknown" error; now it says that the functionality isn't in the binary.
506 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
507 an address' error message when a string expansion fails (syntax or
508 whatever). Otherwise the password may appear in the log. Following change
509 PH/42 below, there is no longer a chance of it appearing in a bounce
512 PH/11 Installed exipick version 20050225.0 from John Jetmore.
514 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
515 of its addresses was ever tried. (Bugzilla bug #2.)
517 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
518 the result incorrectly in the debug output. (It correctly added a newline
519 to what was transported.)
521 TF/01 Added $received_time.
523 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
524 commented out examples of how to interface to a virus scanner and to
525 SpamAssassin. Also added commented examples of av_scanner and
526 spamd_address settings.
528 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
529 and controls are allowed in which ACLs. There were a couple of minor
530 errors. Some of the entries in the conditions table (which is a table of
531 where they are NOT allowed) were getting very unwieldy; rewrote them as a
532 negation of where the condition IS allowed.
534 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
536 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
537 header file does not have a version number, so I've had to invent a new
538 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
539 API. The code is untested by me (my Linux distribution still has 0.3.2 of
540 radiusclient), but it was contributed by a Radius user.
542 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
543 files or directories, for OpenSSL.
545 PH/19 When an Exim process that is running as root has to create an Exim log
546 file, it does so in a subprocess that runs as exim:exim so as to get the
547 ownership right at creation (otherwise, other Exim processes might see
548 the file with the wrong ownership). There was no test for failure of this
549 fork() call, which would lead to the process getting stuck as it waited
550 for a non-existent subprocess. Forks do occasionally fail when resources
551 run out. I reviewed all the other calls to fork(); they all seem to check
554 PH/20 When checking for unexpected SMTP input at connect time (before writing
555 the banner), Exim was not dealing correctly with a non-positive return
556 from the read() function. If the client had disconnected by this time,
557 the result was a log entry for a synchronization error with an empty
558 string after "input=" when read() returned zero. If read() returned -1
559 (an event I could not check), uninitialized data bytes were printed.
560 There were reports of junk text (parts of files, etc) appearing after
563 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
565 PH/22 Added support for macro redefinition, and (re)definition in between
566 driver and ACL definitions.
568 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
569 forgetting to use the resulting value; it was using the unexpanded value.
571 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
572 hadn't been configured. The fix is from Juergen Kreileder, who
573 understands it better than I do:
575 "Here's what I see happening with three configured cyrus_sasl
576 authenticators configured (plain, login, cram-md5):
578 On startup auth_cyrus_sasl_init() gets called for each of these.
579 This means three calls to sasl_listmech() without a specified mech_list.
580 => SASL tests which mechs of all available mechs actually work
581 => three warnings about OTP not working
582 => the returned list contains: plain, login, cram-md5, digest-md5, ...
584 With the patch, sasl_listmech() also gets called three times. But now
585 SASL's mech_list option is set to the server_mech specified in the the
586 authenticator. Or in other words, the answer from sasl_listmech()
587 gets limited to just the mech you're testing for (which is different
589 => the return list contains just 'plain' or 'login', 'cram-md5' or
590 nothing depending on the value of ob->server_mech.
592 I've just tested the patch: Authentication still works fine,
593 unavailable mechs specified in the exim configuration are still
594 caught, and the auth.log warnings about OTP are gone."
596 PH/25 When debugging is enabled, the contents of the command line are added
597 to the debugging output, even when log_selector=+arguments is not
600 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
601 answer is "GNU", and only if the return is "GNU/something" is the answer
604 PH/27 $acl_verify_message is now set immediately after the failure of a
605 verification in an ACL, and so is available in subsequent modifiers. In
606 particular, the message can be preserved by coding like this:
608 warn !verify = sender
609 set acl_m0 = $acl_verify_message
611 Previously, $acl_verify_message was set only while expanding "message"
612 and "log_message" when a very denied access.
614 PH/28 Modified OS/os.c-Linux with
616 -#ifndef OS_LOAD_AVERAGE
617 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
619 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
620 nomenclature these days.)
622 PH/29 Installed patch from the Sieve maintainer that adds the options
623 sieve_useraddress and sieve_subaddress to the redirect router.
625 PH/30 In these circumstances:
626 . Two addresses routed to the same list of hosts;
627 . First host does not offer TLS;
628 . First host accepts first address;
629 . First host gives temporary error to second address;
630 . Second host offers TLS and a TLS session is established;
631 . Second host accepts second address.
632 Exim incorrectly logged both deliveries with the TLS parameters (cipher
633 and peerdn, if requested) that were in fact used only for the second
636 PH/31 When doing a callout as part of verifying an address, Exim was not paying
637 attention to any local part prefix or suffix that was matched by the
638 router that accepted the address. It now behaves in the same way as it
639 does for delivery: the affixes are removed from the local part unless
640 rcpt_include_affixes is set on the transport.
642 PH/32 Add the sender address, as F=<...>, to the log line when logging a
643 timeout during the DATA phase of an incoming message.
645 PH/33 Sieve envelope tests were broken for match types other than :is. I have
646 applied a patch sanctioned by the Sieve maintainer.
648 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
649 the uid or gid is negative. A case of a negative gid caused this to be
650 noticed. The fix allows for either to be negative.
652 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
653 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
656 PH/36 The change PH/12 above was broken. Fixed it.
658 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
659 the grounds that routing the same address twice would always produce the
660 same answer. This might have been true once, but it is certainly no
661 longer true now. Routing a child address may depend on the previous
662 routing that produced that child. Some complicated redirection strategies
663 went wrong when messages had multiple recipients, and made Exim's
664 behaviour dependent on the order in which the addresses were given.
666 I have moved the duplicate checking until after the routing is complete.
667 Exim scans the addresses that are assigned to local and remote
668 transports, and removes any duplicates. This means that more work will be
669 done, as duplicates will always all be routed, but duplicates are
670 presumably rare, so I don't expect this is of any significance.
672 For deliveries to pipes, files, and autoreplies, the duplicate checking
673 still happens during the routing process, since they are not going to be
676 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
677 It corrects a timeout issue with spamd. This is Ian's comment: "The
678 background is that sometimes spamd either never reads data from a
679 connection it has accepted, or it never writes response data. The exiscan
680 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
681 blindly assumes that writes won't block so it may never time out."
683 PH/39 Allow G after quota size as well as K and M.
685 PH/40 The value set for $authenticated_id in an authenticator may not contain
686 binary zeroes or newlines because the value is written to log lines and
687 to spool files. There was no check on this. Now the value is run through
688 the string_printing() function so that such characters are converted to
689 printable escape sequences.
691 PH/41 $message_linecount is a new variable that contains the total number of
692 lines in the message. Compare $body_linecount, which is the count for the
695 PH/42 Exim no longer gives details of delivery errors for specific addresses in
696 bounce and delay warning messages, except in certain special cases, which
699 (a) An SMTP error message from a remote host;
700 (b) A message specified in a :fail: redirection;
701 (c) A message specified in a "fail" command in a system filter;
702 (d) A message specified in a FAIL return from the queryprogram router;
703 (e) A message specified by the cannot_route_message router option.
705 In these cases only, Exim does include the error details in bounce and
706 warning messages. There are also a few cases where bland messages such
707 as "unrouteable address" or "local delivery error" are given.
709 PH/43 $value is now also set for the "else" part of a ${run expansion.
711 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
712 being worked on, but at least Exim now implements the latest version to
715 PH/45 In a pipe transport, although a timeout while waiting for the pipe
716 process to complete was treated as a delivery failure, a timeout while
717 writing the message to the pipe was logged, but erroneously treated as a
718 successful delivery. Such timeouts include transport filter timeouts. For
719 consistency with the overall process timeout, these timeouts are now
720 treated as errors, giving rise to delivery failures by default. However,
721 there is now a new Boolean option for the pipe transport called
722 timeout_defer, which, if set TRUE, converts the failures into defers for
723 both kinds of timeout. A transport filter timeout is now identified in
726 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
727 systems where "make" and "gmake" are different, calling "gmake" at top
728 level broke things. I've arranged for the value of $(MAKE) to be passed
729 from the Makefile to this script so that it can call the same version of
733 A note about Exim versions 4.44 and 4.50
734 ----------------------------------------
736 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
737 changes of various kinds. As a consequence, a big documentation update was
738 needed. This delayed the release for rather longer than seemed good, especially
739 in the light of a couple of (minor) security issues. Therefore, the changes
740 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
741 release. So 4.44 and 4.50 are in effect two different branches that both start
744 I have left the 4.50 change log unchanged; it contains all the changes since
745 4.43. The change log for 4.44 is below; many of its items are identical to
746 those for 4.50. This seems to be the most sensible way to preserve the
747 historical information.
753 1. Minor wording change to the doc/README.SIEVE file.
755 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
756 computation of the current number of files was incorrect.
758 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
759 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
762 4. Give more explanation in the error message when the command for a transport
763 filter fails to execute.
765 5. There are several places where Exim runs a non-Exim command in a
766 subprocess. The SIGUSR1 signal should be disabled for these processes. This
767 was being done only for the command run by the queryprogram router. It is
768 now done for all such subprocesses. The other cases are: ${run, transport
769 filters, and the commands run by the lmtp and pipe transports.
771 6. Added CONFIGURE_GROUP build-time option.
773 7. Some older OS have a limit of 256 on the maximum number of file
774 descriptors. Exim was using setrlimit() to set 1000 as a large value
775 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
776 systems. I've change it so that if it can't get 1000, it tries for 256.
778 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
779 was an oversight, and furthermore, ever since the addition of extra
780 controls (e.g. 4.43/32), the checks on when to allow different forms of
781 "control" were broken. There should now be diagnostics for all cases when a
782 control that does not make sense is encountered.
784 9. Added the /retain_sender option to "control=submission".
786 10. $recipients is now available in the predata ACL (oversight).
788 11. Tidy the search cache before the fork to do a delivery from a message
789 received from the command line. Otherwise the child will trigger a lookup
790 failure and thereby defer the delivery if it tries to use (for example) a
791 cached ldap connection that the parent has called unbind on.
793 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
794 of $address_data from the recipient verification was clobbered by the
797 13. The value of address_data from a sender verification is now available in
798 $sender_address_data in subsequent conditions in the ACL statement.
800 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
802 15. Added a new option "connect=<time>" to callout options, to set a different
805 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
806 was its contents. (It was OK if the option was not defined at all.)
808 17. A "Completed" log line is now written for messages that are removed from
809 the spool by the -Mrm option.
811 18. New variables $sender_verify_failure and $recipient_verify_failure contain
812 information about exactly what failed.
814 19. Added -dd to debug only the daemon process.
816 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
817 handles timeouts, both on the server side and network timeouts. Renamed the
818 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
820 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
821 It is now set to "smtps".
823 22. $host_address is now set to the target address during the checking of
826 23. When checking ignore_target_hosts for an ipliteral router, no host name was
827 being passed; this would have caused $sender_host_name to have been used if
828 matching the list had actually called for a host name (not very likely,
829 since this list is usually IP addresses). A host name is now passed as
832 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
833 code that specifies a non-restarting handler (typically sigaction() in
834 modern systems) in an attempt to fix a rare and obscure crash bug.
836 25. Narrowed the window for a race in the daemon that could cause it to ignore
837 SIGCHLD signals. This is not a major problem, because they are used only to
838 wake it up if nothing else does.
840 26. A malformed maildirsize file could cause Exim to calculate negative values
841 for the mailbox size or file count. Odd effects could occur as a result.
842 The maildirsize information is now recalculated if the size or filecount
845 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
846 support for a long time. Removed HAVE_SYS_VFS_H.
848 28. Installed the latest version of exipick from John Jetmore.
850 29. In an address list, if the pattern was not a regular expression, an empty
851 subject address (from a bounce message) matched only if the pattern was an
852 empty string. Non-empty patterns were not even tested. This was the wrong
853 because it is perfectly reasonable to use an empty address as part of a
854 database query. An empty address is now tested by patterns that are
855 lookups. However, all the other forms of pattern expect the subject to
856 contain a local part and a domain, and therefore, for them, an empty
857 address still always fails if the pattern is not itself empty.
859 30. Exim went into a mad DNS loop when attempting to do a callout where the
860 host was specified on an smtp transport, and looking it up yielded more
863 31. Re-factored the code for checking spool and log partition space into a
864 function that finds that data and another that does the check. The former
865 is then used to implement four new variables: $spool_space, $log_space,
866 $spool_inodes, and $log_inodes.
868 32. The RFC2047 encoding function was originally intended for short strings
869 such as real names; it was not keeping to the 75-character limit for
870 encoded words that the RFC imposes. It now respects the limit, and
871 generates multiple encoded words if necessary. To be on the safe side, I
872 have increased the buffer size for the ${rfc2047: expansion operator from
875 33. It is now permitted to omit both strings after an "if" condition; if the
876 condition is true, the result is "true". As before, when the second string
877 is omitted, a false condition yields an empty string. This makes it less
878 cumbersome to write custom ACL and router conditions.
880 34. Failure to deliver a bounce message always caused it to be frozen, even if
881 there was an errors_to setting on the router. The errors_to setting is now
884 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
885 canonical form (fully expanded) before being placed in
886 $sender_host_address.
888 36. The table in the code that translates DNS record types into text (T_A to
889 "A" for instance) was missing entries for NS and CNAME. It is just possible
890 that this could have caused confusion if both these types were looked up
891 for the same domain, because the text type is used as part of Exim's
892 per-process caching. But the chance of anyone hitting this buglet seems
895 37. The dnsdb lookup has been extended in a number of ways.
897 (1) There is a new type, "zns", which walks up the domain tree until it
898 finds some nameserver records. It should be used with care.
900 (2) There is a new type, "mxh", which is like "mx" except that it returns
901 just the host names, not the priorities.
903 (3) It is now possible to give a list of domains (or IP addresses) to be
904 looked up. The behaviour when one of the lookups defers can be
905 controlled by a keyword.
907 (4) It is now possible to specify the separator character for use when
908 multiple records are returned.
910 38. The dnslists ACL condition has been extended: it is now possible to supply
911 a list of IP addresses and/or domains to be looked up in a particular DNS
914 39. Added log_selector=+queue_time_overall.
916 40. When running the queue in the test harness, wait just a tad after forking a
917 delivery process, to get repeatability of debugging output.
919 41. Include certificate and key file names in error message when GnuTLS fails
920 to set them up, because the GnuTLS error message doesn't include the name
921 of the failing file when there is a problem reading it.
923 42. Allow both -bf and -bF in the same test run.
925 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
927 44. The "Exiscan patch" is now merged into the mainline Exim source.
929 45. Sometimes the final signoff response after QUIT could fail to get
930 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
931 < 0 before doing a fflush(). This bug looks as though it goes back to the
932 introduction of TLS in release 3.20, but "sometimes" must have been rare
933 because the tests only now provoked it.
935 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
936 (this can affect the format of dates).
938 47. exim_tidydb, when checking for the continued existence of a message for
939 which it has found a message-specific retry record, was not finding
940 messages that were in split spool directories. Consequently, it was
941 deleting retry records that should have stayed in existence.
943 48. Steve fixed some bugs in eximstats.
945 49. The SPA authentication driver was not abandoning authentication and moving
946 on to the next authenticator when an expansion was forced to fail,
947 contradicting the general specification for all authenticators. Instead it
948 was generating a temporary error. It now behaves as specified.
950 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
951 (the order specifies the preference for clients). The order is now AES256,
952 AES128, 3DES, ARCFOUR128.
954 51. Small patch to Sieve code - explicitly set From: when generating an
957 52. Exim crashed if a remote delivery caused a very long error message to be
958 recorded - for instance if somebody sent an entire SpamAssassin report back
959 as a large number of 550 error lines. This bug was coincidentally fixed by
960 increasing the size of one of Exim's internal buffers (big_buffer) that
961 happened as part of the Exiscan merge. However, to be on the safe side, I
962 have made the code more robust (and fixed the comments that describe what
965 53. Now that there can be additional text after "Completed" in log lines (if
966 the queue_time_overall log selector is set), a one-byte patch to exigrep
967 was needed to allow it to recognize "Completed" as not the last thing in
970 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
971 patch that reportedly fixes this has been added. I am not expert enough to
972 create a test for it. This is what the patch creator wrote:
974 "I found a little strange behaviour of ldap code when working with
975 Windows 2003 AD Domain, where users was placed in more than one
976 Organization Units. When I tried to give exim partial DN, the exit code
977 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
978 But simultaneously result of request was absolutely normal ldap result,
979 so I produce this patch..."
981 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
982 REFERENCE, so I have modified the code to exclude the patch when that macro
985 55. Some experimental protocols are using DNS PTR records for new purposes. The
986 keys for these records are domain names, not reversed IP addresses. The
987 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
988 leaves it alone. Component reversal etc. now happens only for IP addresses.
991 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
993 57. Double the size of the debug message buffer (to 2048) so that more of very
994 long debug lines gets shown.
996 58. The exicyclog utility now does better if the number of log files to keep
997 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
999 59. Two changes related to the smtp_active_hostname option:
1001 (1) $smtp_active_hostname is now available as a variable.
1002 (2) The default for smtp_banner uses $smtp_active_hostname instead
1003 of $primary_hostname.
1005 60. The host_aton() function is supposed to be passed a string that is known
1006 to be a valid IP address. However, in the case of IPv6 addresses, it was
1007 not checking this. This is a hostage to fortune. Exim now panics and dies
1008 if the condition is not met. A case was found where this could be provoked
1009 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1010 components; fortuitously, this particular loophole had already been fixed
1011 by change 4.50/55 above.
1013 If there are any other similar loopholes, the new check in host_aton()
1014 itself should stop them being exploited. The report I received stated that
1015 data on the command line could provoke the exploit when Exim was running as
1016 exim, but did not say which command line option was involved. All I could
1017 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1018 running as the user.
1021 61. There was a buffer overflow vulnerability in the SPA authentication code
1022 (which came originally from the Samba project). I have added a test to the
1023 spa_base64_to_bits() function which I hope fixes it.
1026 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1027 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1029 63. The daemon start-up calls getloadavg() while still root for those OS that
1030 need the first call to be done as root, but it missed one case: when
1031 deliver_queue_load_max is set with deliver_drop_privilege. This is
1032 necessary for the benefit of the queue runner, because there is no re-exec
1033 when deliver_drop_privilege is set.
1035 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1036 This has been fixed.
1038 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1039 was in use, was not putting the data itself into the right store pool;
1040 consequently, it could be overwritten for a subsequent message in the same
1041 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1044 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1046 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1047 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1048 the function were treating the return as a boolean value, which happened to
1049 work because 0=false and not-0=true, but is not correct code.
1051 68. The host_aton() function was not handling scoped IPv6 addresses (those
1052 with, for example, "%eth0" on the end) correctly.
1054 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1055 negated items (that is, ~something) in unsigned ints. Some compilers
1056 apparently mutter when there is no cast.
1058 70. If an address verification called from an ACL failed, and did not produce a
1059 user-specific message (i.e. there was only a "system" message), nothing was
1060 put in $acl_verify_message. In this situation, it now puts the system
1063 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1064 session; change 4.31/43 added the unwanted input to the log line - except
1065 that it did not do this in the start of session case. It now does.
1067 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1068 This is wrong and can cause the other end to generate a synchronization
1069 error if it is another Exim or anything else that does the synchronization
1070 check. A QUIT command is no longer sent after a timeout.
1072 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1073 during host lookups.
1075 74. The defer_ok option of callout verification was not working if it was used
1076 when verifying addresses in header lines, that is, for this case:
1078 verify = header_sender/callout=defer_ok
1080 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1081 those file descriptors could be used for SMTP connections. If anything
1082 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1083 could be sent to the SMTP client, causing chaos. The daemon now opens
1084 stdin, stdout, and stderr to /dev/null when it puts itself into the
1087 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1088 log by default. The user can override this with suitable Perl magic.
1090 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1091 the log message when discard triggers, was not working for the DATA ACL or
1092 for the non-SMTP ACL.
1094 78. Error message wording change in sieve.c.
1096 79. If smtp_accept_max_per_host was set, the number of connections could be
1097 restricted to fewer than expected, because the daemon was trying to set up
1098 a new connection before checking whether the processes handling previous
1099 connections had finished. The check for completed processes is now done
1100 earlier. On busy systems, this bug wouldn't be noticed because something
1101 else would have woken the daemon, and it would have reaped the completed
1104 80. If a message was submitted locally by a user whose login name contained one
1105 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1106 It caused a spool format error. I have fixed the spool reading code. A
1107 related problem was that the "from" clause in the Received: line became
1108 illegal because of the space(s). It is now covered by ${quote_local_part.
1110 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1113 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1115 83. Patch from Sieve maintainer to fix unterminated string problem in
1116 "vacation" handling.
1118 84. Some minor changes to the Linux configuration files to help with other
1119 OS variants using glibc.
1121 85. One more patch for Sieve to update vacation handling to latest spec.
1124 ----------------------------------------------------
1125 See the note above about the 4.44 and 4.50 releases.
1126 ----------------------------------------------------
1132 1. Change 4.43/35 introduced a bug that caused file counts to be
1133 incorrectly computed when quota_filecount was set in an appendfile
1136 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1137 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1140 3. Give more explanation in the error message when the command for a transport
1141 filter fails to execute.
1143 4. There are several places where Exim runs a non-Exim command in a
1144 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1145 was being done only for the command run by the queryprogram router. It is
1146 now done for all such subprocesses. The other cases are: ${run, transport
1147 filters, and the commands run by the lmtp and pipe transports.
1149 5. Some older OS have a limit of 256 on the maximum number of file
1150 descriptors. Exim was using setrlimit() to set 1000 as a large value
1151 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1152 systems. I've change it so that if it can't get 1000, it tries for 256.
1154 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1155 was an oversight, and furthermore, ever since the addition of extra
1156 controls (e.g. 4.43/32), the checks on when to allow different forms of
1157 "control" were broken. There should now be diagnostics for all cases when a
1158 control that does not make sense is encountered.
1160 7. $recipients is now available in the predata ACL (oversight).
1162 8. Tidy the search cache before the fork to do a delivery from a message
1163 received from the command line. Otherwise the child will trigger a lookup
1164 failure and thereby defer the delivery if it tries to use (for example) a
1165 cached ldap connection that the parent has called unbind on.
1167 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1168 of $address_data from the recipient verification was clobbered by the
1169 sender verification.
1171 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1172 was its contents. (It was OK if the option was not defined at all.)
1174 11. A "Completed" log line is now written for messages that are removed from
1175 the spool by the -Mrm option.
1177 12. $host_address is now set to the target address during the checking of
1178 ignore_target_hosts.
1180 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1181 being passed; this would have caused $sender_host_name to have been used if
1182 matching the list had actually called for a host name (not very likely,
1183 since this list is usually IP addresses). A host name is now passed as
1186 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1187 code that specifies a non-restarting handler (typically sigaction() in
1188 modern systems) in an attempt to fix a rare and obscure crash bug.
1190 15. Narrowed the window for a race in the daemon that could cause it to ignore
1191 SIGCHLD signals. This is not a major problem, because they are used only to
1192 wake it up if nothing else does.
1194 16. A malformed maildirsize file could cause Exim to calculate negative values
1195 for the mailbox size or file count. Odd effects could occur as a result.
1196 The maildirsize information is now recalculated if the size or filecount
1199 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1200 support for a long time. Removed HAVE_SYS_VFS_H.
1202 18. Updated exipick to current release from John Jetmore.
1204 19. Allow an empty sender to be matched against a lookup in an address list.
1205 Previously the only cases considered were a regular expression, or an
1208 20. Exim went into a mad DNS lookup loop when doing a callout where the
1209 host was specified on the transport, if the DNS lookup yielded more than
1212 21. The RFC2047 encoding function was originally intended for short strings
1213 such as real names; it was not keeping to the 75-character limit for
1214 encoded words that the RFC imposes. It now respects the limit, and
1215 generates multiple encoded words if necessary. To be on the safe side, I
1216 have increased the buffer size for the ${rfc2047: expansion operator from
1219 22. Failure to deliver a bounce message always caused it to be frozen, even if
1220 there was an errors_to setting on the router. The errors_to setting is now
1223 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1224 canonical form (fully expanded) before being placed in
1225 $sender_host_address.
1227 24. Updated eximstats to version 1.33
1229 25. Include certificate and key file names in error message when GnuTLS fails
1230 to set them up, because the GnuTLS error message doesn't include the name
1231 of the failing file when there is a problem reading it.
1233 26. Expand error message when OpenSSL has problems setting up cert/key files.
1236 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1237 (this can affect the format of dates).
1239 28. exim_tidydb, when checking for the continued existence of a message for
1240 which it has found a message-specific retry record, was not finding
1241 messages that were in split spool directories. Consequently, it was
1242 deleting retry records that should have stayed in existence.
1244 29. eximstats updated to version 1.35
1245 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1246 1.35 - bugfix such that pie charts by volume are generated correctly
1248 30. The SPA authentication driver was not abandoning authentication and moving
1249 on to the next authenticator when an expansion was forced to fail,
1250 contradicting the general specification for all authenticators. Instead it
1251 was generating a temporary error. It now behaves as specified.
1253 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1254 (the order specifies the preference for clients). The order is now AES256,
1255 AES128, 3DES, ARCFOUR128.
1257 31. Small patch to Sieve code - explicitly set From: when generating an
1260 32. Exim crashed if a remote delivery caused a very long error message to be
1261 recorded - for instance if somebody sent an entire SpamAssassin report back
1262 as a large number of 550 error lines. This bug was coincidentally fixed by
1263 increasing the size of one of Exim's internal buffers (big_buffer) that
1264 happened as part of the Exiscan merge. However, to be on the safe side, I
1265 have made the code more robust (and fixed the comments that describe what
1268 33. Some experimental protocols are using DNS PTR records for new purposes. The
1269 keys for these records are domain names, not reversed IP addresses. The
1270 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1271 leaves it alone. Component reversal etc. now happens only for IP addresses.
1274 34. The host_aton() function is supposed to be passed a string that is known
1275 to be a valid IP address. However, in the case of IPv6 addresses, it was
1276 not checking this. This is a hostage to fortune. Exim now panics and dies
1277 if the condition is not met. A case was found where this could be provoked
1278 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1279 components; fortuitously, this particular loophole had already been fixed
1280 by change 4.50/55 or 4.44/33 above.
1282 If there are any other similar loopholes, the new check in host_aton()
1283 itself should stop them being exploited. The report I received stated that
1284 data on the command line could provoke the exploit when Exim was running as
1285 exim, but did not say which command line option was involved. All I could
1286 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1287 running as the user.
1290 35. There was a buffer overflow vulnerability in the SPA authentication code
1291 (which came originally from the Samba project). I have added a test to the
1292 spa_base64_to_bits() function which I hope fixes it.
1295 36. The daemon start-up calls getloadavg() while still root for those OS that
1296 need the first call to be done as root, but it missed one case: when
1297 deliver_queue_load_max is set with deliver_drop_privilege. This is
1298 necessary for the benefit of the queue runner, because there is no re-exec
1299 when deliver_drop_privilege is set.
1301 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1302 was in use, was not putting the data itself into the right store pool;
1303 consequently, it could be overwritten for a subsequent message in the same
1304 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1307 38. Sometimes the final signoff response after QUIT could fail to get
1308 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1309 < 0 before doing a fflush(). This bug looks as though it goes back to the
1310 introduction of TLS in release 3.20, but "sometimes" must have been rare
1311 because the tests only now provoked it.
1317 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1318 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1319 Now it returns an integer. A number of calls were still expecting a T/F
1320 return. Fortuitously, in all cases, the tests worked in OK situations,
1321 which is the norm. However, things would have gone wrong on any write
1322 failures on the smtp file descriptor. This function is used when sending
1323 messages over SMTP and also when doing verify callouts.
1325 2. When Exim is called to do synchronous delivery of a locally submitted
1326 message (the -odf or -odi options), it no longer closes stderr before doing
1329 3. Implemented the mua_wrapper option.
1331 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1333 5. Implemented the functions header_remove(), header_testname(),
1334 header_add_at_position(), and receive_remove_recipient(), and exported them
1337 6. If an ACL "warn" statement specified the addition of headers, Exim already
1338 inserted X-ACL-Warn: at the start if there was no header name. However, it
1339 was not making this test for the second and subsequent header lines if
1340 there were newlines in the string. This meant that an invalid header could
1341 be inserted if Exim was badly configured.
1343 7. Allow an ACL "warn" statement to add header lines at the start or after all
1344 the Received: headers, as well as at the end.
1346 8. Added the rcpt_4xx retry error code.
1348 9. Added postmaster_mailfrom=xxx to callout verification option.
1350 10. Added mailfrom=xxxx to the callout verification option, for verify=
1353 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1354 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1356 12. Inserted some casts to stop certain compilers warning when using pointer
1357 differences as field lengths or precisions in printf-type calls (mostly
1358 affecting debugging statements).
1360 13. Added optional readline() support for -be (dynamically loaded).
1362 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1363 same clock tick as a message's arrival, so that its received time was the
1364 same as the "first fail" time on the retry record, and that message
1365 remained on the queue past the ultimate address timeout, every queue runner
1366 would try a delivery (because it was past the ultimate address timeout) but
1367 after another failure, the ultimate address timeout, which should have then
1368 bounced the address, did not kick in. This was a "< instead of <=" error;
1369 in most cases the first failure would have been in the next clock tick
1370 after the received time, and all would be well.
1372 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1373 being recognized when the domain list was tested by the match_domain
1374 condition in an expansion string.
1376 16. Added the ${str2b64: operator.
1378 17. Exim was always calling setrlimit() to set a large limit for the number of
1379 processes, without checking whether the existing limit was already
1380 adequate. (It did check for the limit on file descriptors.) Furthermore,
1381 errors from getrlimit() and setrlimit() were being ignored. Now they are
1382 logged to the main and panic logs, but Exim does carry on, to try to do its
1383 job under whatever limits there are.
1385 18. Imported PCRE 5.0.
1387 19. Trivial typo in log message " temporarily refused connection" (the leading
1390 20. If the log selector return_path_on_delivery was set and an address was
1391 redirected to /dev/null, the delivery process crashed because it assumed
1392 that a return path would always be set for a "successful" delivery. In this
1393 case, the whole delivery is bypassed as an optimization, and therefore no
1396 21. Internal re-arrangement: the function for sending a challenge and reading
1397 a response while authentication was assuming a zero-terminated challenge
1398 string. It's now changed to take a pointer and a length, to allow for
1399 binary data in such strings.
1401 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1403 23. Exim was not respecting finduser_retries when seeking the login of the
1404 uid under which it was called; it was always trying 10 times. (The default
1405 setting of finduser_retries is zero.) Also, it was sleeping after the final
1406 failure, which is pointless.
1408 24. Implemented tls_on_connect_ports.
1410 25. Implemented acl_smtp_predata.
1412 26. If the domain in control=submission is set empty, Exim assumes that the
1413 authenticated id is a complete email address when it generates From: or
1414 Sender: header lines.
1416 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1417 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1418 chown and chgrp in /bin and hostname in /usr/bin.
1420 28. Exim was keeping the "process log" file open after each use, just as it
1421 does for the main log. This opens the possibility of it remaining open for
1422 long periods when the USR1 signal hits a daemon. Occasional processlog
1423 errors were reported, that could have been caused by this. Anyway, it seems
1424 much more sensible not to leave this file open at all, so that is what now
1427 29. The long-running daemon process does not normally write to the log once it
1428 has entered its main loop, and it closes the log before doing so. This is
1429 so that log files can straightforwardly be renamed and moved. However,
1430 there are a couple of unusual error situations where the daemon does write
1431 log entries, and I had neglected to close the log afterwards.
1433 30. The text of an SMTP error response that was received during a remote
1434 delivery was being truncated at 512 bytes. This is too short for some of
1435 the long messages that one sometimes sees. I've increased the limit to
1438 31. It is now possible to make retry rules that apply only when a message has a
1439 specific sender, in particular, an empty sender.
1441 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1442 it possible to be selective about when SMTP synchronization is enforced.
1444 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1446 32. Implemented hosts_connection_nolog.
1448 33. Added an ACL for QUIT.
1450 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1453 35. Added mailbox_size and mailbox_filecount to appendfile.
1455 36. Added control = no_multiline_responses to ACLs.
1457 37. There was a bug in the logic of the code that waits for the clock to tick
1458 in the case where the clock went backwards by a substantial amount such
1459 that the microsecond fraction of "now" was more than the microsecond
1460 fraction of "then" (but the whole seconds number was less).
1462 38. Added support for the libradius Radius client library this is found on
1463 FreeBSD (previously only the radiusclient library was supported).
1469 1. When certain lookups returned multiple values in the form name=value, the
1470 quoting of the values was not always being done properly. Specifically:
1471 (a) If the value started with a double quote, but contained no whitespace,
1473 (b) If the value contained whitespace other than a space character (i.e.
1474 tabs or newlines or carriage returns) it was not quoted.
1475 This fix has been applied to the mysql and pgsql lookups by writing a
1476 separate quoting function and calling it from the lookup code. The fix
1477 should probably also be applied to nisplus, ibase and oracle lookups, but
1478 since I cannot test any of those, I have not disturbed their existing code.
1480 2. A hit in the callout cache for a specific address caused a log line with no
1481 reason for rejecting RCPT. Now it says "Previous (cached) callout
1482 verification failure".
1484 3. There was an off-by-one bug in the queryprogram router. An over-long
1485 return line was truncated at 256 instead of 255 characters, thereby
1486 overflowing its buffer with the terminating zero. As well as fixing this, I
1487 have increased the buffer size to 1024 (and made a note to document this).
1489 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1490 when Exim is waiting for an SMTP response from a remote server, Exim
1491 restarts its select() call on the socket, thereby resetting its timeout.
1492 This is not a problem when such interrupts are rare. Somebody set up a cron
1493 job to run exiwhat every 2 minutes, which is less than the normal select()
1494 timeout (5 or 10 minutes). This meant that the select() timeout never
1495 kicked in because it was always reset. I have fixed this by comparing the
1496 time when an interrupt arrives with the time at the start of the first call
1497 to select(). If more time than the timeout has elapsed, the interrupt is
1498 treated as a timeout.
1500 5. Some internal re-factoring in preparation for the addition of Sieve
1501 extensions (by MH). In particular, the "personal" test is moved to a
1502 separate function, and given an option for scanning Cc: and Bcc: (which is
1503 not set for Exim filters).
1505 6. When Exim created an email address using the login of the caller as the
1506 local part (e.g. when creating a From: or Sender: header line), it was not
1507 quoting the local part when it contained special characters such as @.
1509 7. Installed new OpenBSD configuration files.
1511 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1512 try to make them clearer.
1514 9. Callout options, other than the timeout value, were being ignored when
1515 verifying sender addresses in header lines. For example, when using
1517 verify = header_sender/callout=no_cache
1519 the cache was (incorrectly) being used.
1521 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1522 only the Cygwin environment.
1524 11. When return_path_on_delivery was set as a log selector, if different remote
1525 addresses in the same message used different return paths and parallel
1526 remote delivery occurred, the wrong values would sometimes be logged.
1527 (Whenever a remote delivery process finished, the return path value from
1528 the most recently started remote delivery process was logged.)
1530 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1531 lines when AUTH and/or TLS are in use. This is the "received protocol"
1532 field. Exim used to use "asmtp" for authenticated SMTP, without any
1533 indication (in the protocol name) for TLS use. Now it follows the RFC and
1534 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1535 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1536 names appear in log lines as well as in Received: header lines.
1538 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1539 extensions, and comparison tests, and to fix some bugs.
1541 14. Changes to the "personal" filter test:
1543 (1) The test was buggy in that it was just doing the equivalent of
1544 "contains" tests on header lines. For example, if a user's address was
1545 anne@some.where, the "personal" test would incorrectly be true for
1547 To: susanne@some.where
1549 This test is now done by extracting each address from the header in turn,
1550 and checking the entire address. Other tests that are part of "personal"
1551 are now done using regular expressions (for example, to check local parts
1552 of addresses in From: header lines).
1554 (2) The list of non-personal local parts in From: addresses has been
1555 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1556 taken from the Sieve specification recommendations.
1558 (3) If the message contains any header line starting with "List-" it is
1559 treated as non-personal.
1561 (4) The test for "circular" in the Subject: header line has been removed
1562 because it now seems ill-conceived.
1564 15. Minor typos in src/EDITME comments corrected.
1566 16. Installed latest exipick from John Jetmore.
1568 17. If headers_add on a router specified a text string that was too long for
1569 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1570 of string_sprintf() is now avoided.
1572 18. $message_body_size was not set (it was always zero) when running the DATA
1573 ACL and the local_scan() function.
1575 19. For the "mail" command in an Exim filter, no default was being set for
1576 the once_repeat time, causing a random time value to be used if "once" was
1577 specified. (If the value happened to be <= 0, no repeat happened.) The
1578 default is now 0s, meaning "never repeat". The "vacation" command was OK
1579 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1580 (I found it when inspecting the code).
1582 20. There is now an overall timeout for performing a callout verification. It
1583 defaults to 4 times the callout timeout, which applies to individual SMTP
1584 commands during the callout. The overall timeout applies when there is more
1585 than one host that can be tried. The timeout is checked before trying the
1586 next host. This prevents very long delays if there are a large number of
1587 hosts and all are timing out (e.g. when the network connections are timing
1588 out). The value of the overall timeout can be changed by specifying an
1589 additional sub-option for "callout", called "maxwait". For example:
1591 verify = sender/callout=5s,maxwait=20s
1593 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1594 to the end before writing, but this should make it even safer).
1596 22. Exim was forgetting that it had advertised PIPELINING for the second and
1597 subsequent messages on an SMTP connection. It was also not resetting its
1598 memory on STARTTLS and an internal HELO.
1600 23. When Exim logs an SMTP synchronization error within a session, it now
1601 records whether PIPELINING has been advertised or not.
1603 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1604 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1605 rather than long int.
1607 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1609 26. Added the never_mail option to autoreply.
1615 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1616 crash if the getsockname() call failed; this can happen if a connection is
1617 closed very soon after it is established. The problem was simply in the
1618 order in which certain operations were done, causing Exim to try to write
1619 to the SMTP stream before it had set up the file descriptor. The bug has
1620 been fixed by making things happen in the correct order.
1626 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1627 before the connection was closed, thus losing the rejection response.
1629 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1630 some early Solaris releases, but causes trouble in current releases where
1631 socklen_t is defined.
1633 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1636 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1639 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1640 that releasing the top of it at the end releases what was used for sub-
1641 expansions (unless the block got too big). However, discard this block if
1642 the first thing is a variable or header, so that we can use its block when
1643 it is dynamic (useful for very large $message_headers, for example).
1645 6. Lookups now cache *every* query, not just the most recent. A new, separate
1646 store pool is used for this. It can be recovered when all lookup caches are
1647 flushed. Lookups now release memory at the end of their result strings.
1648 This has involved some general refactoring of the lookup sources.
1650 7. Some code has been added to the store_xxx() functions to reduce the amount
1651 of flapping under certain conditions.
1653 8. log_incoming_interface used to affect only the <= reception log lines. Now
1654 it causes the local interface and port to be added to several more SMTP log
1655 lines, for example "SMTP connection from", and rejection lines.
1657 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1659 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1661 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1662 could be overwritten at the end of the current message (or the start of a
1663 new message if it was set in a HELO ACL). The value is now preserved for
1664 the duration of the SMTP connection.
1666 12. If a transport had a headers_rewrite setting, and a matching header line
1667 contained an unqualified address, that address was qualified, even if it
1668 did not match any rewriting rules. The underlying bug was that the values
1669 of the flags that permit the existence of unqualified sender and recipient
1670 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1671 non-local messages, and by -bnq for local messages) were not being
1672 preserved with the message after it was received.
1674 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1675 "next input=" as part of the text comprising the host identity instead of
1676 the correct text. The code was using the same buffer for two different
1677 strings. However, depending on which order the printing function evaluated
1678 its arguments, the bug did not always show up. Under Linux, for example, my
1679 test suite worked just fine.
1681 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1682 doesn't work with some older versions of Perl. It has been changed to "my",
1683 which in any case is probably the better facility to use.
1685 15. A really picky compiler found some instances of statements for creating
1686 error messages that either had too many or two few arguments for the format
1689 16. The size of the buffer for calls to the DNS resolver has been increased
1690 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1691 for addresses that have a lot of PTR records. This alleviates a problem; it
1692 does not fully solve it.
1694 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1695 buffer now truncates the list and logs the incident, which is the same
1696 action as happens when Exim is looking up a host name and its aliases.
1697 Previously in this situation something unpredictable would happen;
1698 sometimes it was "internal error: store_reset failed".
1700 18. If a server dropped the connection unexpectedly when an Exim client was
1701 using GnuTLS and trying to read a response, the client delivery process
1702 crashed while trying to generate an error log message.
1704 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1705 string, for example:
1707 warn message = H1: something\nH2: something
1709 the text was added as a single header line from Exim's point of view
1710 though it ended up OK in the delivered message. However, searching for the
1711 second and subsequent header lines using $h_h2: did not work. This has been
1712 fixed. Similarly, if a system filter added multiple headers in this way,
1713 the routers could not see them.
1715 20. Expanded the error message when iplsearch is called with an invalid key to
1716 suggest using net-iplsearch in a host list.
1718 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1719 ACLs are no longer actually imposed (and a message to that effect is
1722 22. If a "gecos" field in a passwd entry contained escaped characters, in
1723 particular, if it contained a \" sequence, Exim got it wrong when building
1724 a From: or a Sender: header from that name. A second bug also caused
1725 incorrect handling when an unquoted " was present following a character
1726 that needed quoting.
1728 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1729 was not being matched caselessly.
1731 24. Arranged for all hyphens in the exim.8 source to be escaped with
1734 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1735 properly. Recipient callout cache records were still being keyed to include
1736 the sender, even when use_sender was set false. This led to far more
1737 callouts that were necessary. The sender is no longer included in the key
1738 when use_sender is false.
1740 26. Added "control = submission" modifier to ACLs.
1742 27. Added the ${base62d: operator to decode base 62 numbers.
1744 28. dnsdb lookups can now access SRV records.
1746 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1747 the configuration file.
1749 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1750 (-v) mode. This makes the output for a verbose queue run more intelligible.
1752 31. Added a use_postmaster feature to recipient callouts.
1754 32. Added the $body_zerocount variable, containing the number of binary zero
1755 bytes in the message body.
1757 33. The time of last modification of the "new" subdirectory is now used as the
1758 "mailbox time last read" when there is a quota error for a maildir
1761 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1763 35. Added +ignore_unknown as a special item in host lists.
1765 36. Code for decoding IPv6 addresses in host lists is now included, even if
1766 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1767 address was recognized as an IP address, but was then not correctly decoded
1768 into binary, causing unexpected and incorrect effects when compared with
1775 1. Very minor rewording of debugging text in manualroute to say "list of
1776 hosts" instead of "hostlist".
1778 2. If verify=header_syntax was set, and a header line with an unqualified
1779 address (no domain) and a large number of spaces between the end of the
1780 name and the colon was received, the reception process suffered a buffer
1781 overflow, and (when I tested it) crashed. This was caused by some obsolete
1782 code that should have been removed. The fix is to remove it!
1784 3. When running in the test harness, delay a bit after writing a bounce
1785 message to get a bit more predictability in the log output.
1787 4. Added a call to search_tidyup() just before forking a reception process. In
1788 theory, someone could use a lookup in the expansion of smtp_accept_max_
1789 per_host which, without the tidyup, could leave open a database connection.
1791 5. Added the variables $recipient_data and $sender_data which get set from a
1792 lookup success in an ACL "recipients" or "senders" condition, or a router
1793 "senders" option, similar to $domain_data and $local_part_data.
1795 6. Moved the writing of debug_print from before to after the "senders" test
1798 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1799 problems for message scanning, either using a data ACL, or using
1800 local_scan() because the Received: header was not generated till after they
1801 were called (in order to set the time as the time of reception completion).
1802 I have revised the way this works. The header is now generated after the
1803 body is received, but before the ACL or local_scan() are called. After they
1804 are run, the timestamp in the header is updated.
1810 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1811 before starting a queue runner without re-exec. This happened only when
1812 deliver_drop_privilege was set or when the Exim user was set to root. The
1813 effect of the bug was that timeouts during subsequent deliveries caused
1814 crashes instead of being properly handled. The handler is now left at its
1815 default (and expected) setting.
1817 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1818 message, again when deliver_drop_privilege is set or Exim is run as root.
1819 The bug described in (1) was not present in this case, but the tidying up
1820 of the other signals was missing. I have made the two cases consistent.
1822 3. The ignore_target_hosts setting on a manualroute router was being ignored
1823 for hosts that were looked up using the /MX notation.
1825 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1828 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1829 operated on the sender address. After changing the $sender_address to <>
1830 for the sender address verify, Exim was re-instated it as the original
1831 (before rewriting) address, but remembering that it had rewritten it, so it
1832 wasn't rewriting it again. This bug also had the effect of breaking the
1833 sender address verification caching when the sender address was rewritten.
1835 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1836 This has been changed so that if the ip literal address matches
1837 ignore_target_hosts, the router declines.
1839 7. Added expansion conditions match_domain, match_address, and match_local_
1840 part (NOT match_host).
1842 8. The placeholder for the Received: header didn't have a length field set.
1844 9. Added code to Exim itself and to exim_lock to test for a specific race
1845 condition that could lead to file corruption when using MBX delivery. The
1846 issue is with the lockfile that is created in /tmp. If this file is removed
1847 after a process has opened it but before that process has acquired a lock,
1848 there is the potential for a second process to recreate the file and also
1849 acquire a lock. This could lead to two Exim processes writing to the file
1850 at the same time. The added code performs the same test as UW imapd; it
1851 checks after acquiring the lock that its file descriptor still refers to
1852 the same named file.
1854 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1855 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1858 11. Added the smtp_active_hostname option. If used, this will typically be made
1859 to depend on the incoming interface address. Because $interface_address is
1860 not set up until the daemon has forked a reception process, error responses
1861 that can happen earlier (such as "too many connections") no longer contain
1864 12. If an expansion in a condition on a "warn" statement fails because a lookup
1865 defers, the "warn" statement is abandoned, and the next ACL statement is
1866 processed. Previously this caused the whole ACL to be aborted.
1868 13. Added the iplsearch lookup type.
1870 14. Added ident_timeout as a log selector.
1872 15. Added tls_certificate_verified as a log selector.
1874 16. Added a global option tls_require_ciphers (compare the smtp transport
1875 option of the same name). This controls incoming TLS connections.
1877 17. I finally figured out how to make tls_require_ciphers do a similar thing
1878 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1879 before starting the TLS session.
1881 18. Tabs are now shown as \t in -bP output.
1883 19. If the log selector return_path_on_delivery was set, Exim crashed when
1884 bouncing a message because it had too many Received: header lines.
1886 20. If two routers both had headers_remove settings, and the first one included
1887 a superfluous trailing colon, the final name in the first list and the
1888 first name in the second list were incorrectly joined into one item (with a
1889 colon in the middle).
1895 1. Added -C and -D options to the exinext utility, mainly to make it easier
1896 to include in the automated testing, but these could be helpful when
1897 multiple configurations are in use.
1899 2. The exinext utility was not formatting the output nicely when there was
1900 an alternate port involved in the retry record key, nor when there was a
1901 message id as well (for retries that were specific to a specific message
1902 and a specific host). It was also confused by IPv6 addresses, because of
1903 the additional colons they contain. I have fixed the IPv4 problem, and
1904 patched it up to do a reasonable job for IPv6.
1906 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1907 delivery, the log line now contains "pipelined" if PIPELINING was used.
1909 4. An SMTP transport process used to panic and die if the bind() call to set
1910 an explicit outgoing interface failed. This has been changed; it is now
1911 treated in the same way as a connect() failure.
1913 5. A reference to $sender_host_name in the part of a conditional expansion
1914 that was being skipped was still causing a DNS lookup. This no longer
1917 6. The def: expansion condition was not recognizing references to header lines
1918 that used bh_ and bheader_.
1920 7. Added the _cache feature to named lists.
1922 8. The code for checking quota_filecount in the appendfile transport was
1923 allowing one more file than it should have been.
1925 9. For compatibility with Sendmail, the command line option
1933 and sets the incoming protocol and host name (for trusted callers). The
1934 host name and its colon can be omitted when only the protocol is to be set.
1935 Note the Exim already has two private options, -pd and -ps, that refer to
1936 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1937 "s", but I don't think that's a major issue.
1939 10. A number of refactoring changes to the code, none of which should affect
1942 (a) The number of logging options was getting close to filling up the
1943 32-bit word that was used as a bit map. I have split them into two classes:
1944 those that are passed in the argument to log_write(), and those that are
1945 only ever tested independently outside of that function. These are now in
1946 separate 32-bit words, so there is plenty of room for expansion again.
1947 There is no change in the user interface or the logging behaviour.
1949 (b) When building, for example, log lines, the code previously used a
1950 macro that called string_cat() twice, in order to add two strings. This is
1951 not really sufficiently general. Furthermore, there was one instance where
1952 it was actually wrong because one of the argument was used twice, and in
1953 one call a function was used. (As it happened, calling the function twice
1954 did not affect the overall behaviour.) The macro has been replaced by a
1955 function that can join an arbitrary number of extra strings onto a growing
1958 (c) The code for expansion conditions now uses a table and a binary chop
1959 instead of a serial search (which was left over from when there were very
1960 few conditions). Also, it now recognizes conditions like "pam" even when
1961 the relevant support is not compiled in: a suitably worded error message is
1962 given if an attempt is made to use such a condition.
1964 11. Added ${time_interval:xxxxx}.
1966 12. A bug was causing one of the ddress fields not to be passed back correctly
1967 from remote delivery subprocesses. The field in question was not being
1968 subsequently used, so this caused to problems in practice.
1970 13. Added new log selectors queue_time and deliver_time.
1972 14. Might have fixed a bug in maildirsizefile handling that threw up
1973 "unexpected character" debug warnings, and recalculated the data
1974 unnecessarily. In any case, I expanded the warning message to give more
1977 15. Added the message "Restricted characters in address" to the statements in
1978 the default ACL that block characters like @ and % in local parts.
1980 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1981 Three changes have been made:
1983 (a) There was a serious bug; a negative response to MAIL caused the whole
1984 recipient domain to be cached as invalid, thereby blocking all messages
1985 to all local parts at the same domain, from all senders. This bug has
1986 been fixed. The domain is no longer cached after a negative response to
1987 MAIL if the sender used is not empty.
1989 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1992 (c) A new callout option, "use_sender" has been added for people who want
1993 the modified behaviour.
1999 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2002 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2003 indeed breaks things for older releases.
2005 3. Added additional logging to the case where there is a problem reading data
2006 from a filter that is running in a subprocess using a pipe, in order to
2007 try to track down a specific problem.
2009 4. Testing facility fudge: when running in the test harness and attempting
2010 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2011 getting "No route to host". Convert this to a timeout.
2013 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2016 6. Some OS don't have socklen_t but use size_t instead. This affects the
2017 fifth argument of getsockopt() amongst other things. This is now
2018 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2019 can be set for individual OS. I have set it for SunOS5, OSF1, and
2020 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2021 some earlier ones do not.
2023 7. Change 4.30/15 was not doing the test caselessly.
2025 8. The standard form for an IPv6 address literal was being rejected by address
2026 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2027 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2028 this, as well as the form without the "IPv6" on the front (but only when
2029 address literals are enabled, of course).
2031 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2033 10. Exim crashed if a message with an empty sender address specified by -f
2034 encountered a router with an errors_to setting. This could be provoked only
2035 by a command such as
2039 where an empty string was supplied; "<>" did not hit this bug.
2041 11. Installed PCRE release 4.5.
2043 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2044 remained set. It is now erased.
2046 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2047 times from message ids (which are base 36 rather than the normal 62).
2049 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2050 were being counted as actual protocol errors, and logged if the log
2051 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2052 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2053 and DATA following a set of rejected RCPTs do not count as protocol errors.
2054 In other words, Exim assumes they were pipelined, though this may not
2055 actually be the case. Of course, in all cases the client gets an
2056 appropriate error code.
2058 15. If a lookup fails in an ACL condition, a message about the failure may
2059 be available; it is used if testing the ACL cannot continue, because most
2060 such messages specify what the cause of the deferral is. However, some
2061 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2062 that caused an old message to be retained and used if a later statement
2063 caused a defer, replacing the real cause of the deferral.
2065 16. If an IP address had so many PTR records that the DNS lookup buffer
2066 was not large enough to hold them, Exim could crash while trying to process
2067 the truncated data. It now detects and logs this case.
2069 17. Further to 4.21/58, another change has been made: if (and only if) the
2070 first line of a message (the first header line) ends with CRLF, a bare LF
2071 in a subsequent header line has a space inserted after it, so as not to
2072 terminate the header.
2074 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2075 unnecessarily, used atoi() instead of strtol(), and didn't check the
2076 termination when getting file sizes from file names by regex.
2078 19. Completely re-implemented the support for maildirsize files, in the light
2079 of a number of problems with the previous contributed implementation
2080 (4.30/29). In particular:
2082 . If the quota is zero, the maildirsize file is maintained, but no quota is
2085 . If the maildir directory does not exist, it is created before any attempt
2086 to write a maildirsize file.
2088 . The quota value in the file is just a cache; if the quota is changed in
2089 the transport, the new value overrides.
2091 . A regular expression is available for excluding directories from the
2094 20. The autoreply transport checks the characters in options that define the
2095 message's headers; it allows continued headers, but it was checking with
2096 isspace() after an embedded newline instead of explicitly looking for a
2099 21. If all the "regular" hosts to which an address was routed had passed their
2100 expiry times, and had not reached their retry times, the address was
2101 bounced, even if fallback hosts were defined. Now Exim should go on to try
2104 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2105 equivalent code in the SMTP transport. Some hosts send humungous responses
2106 to HELO/EHLO, more than 1024 it seems.
2108 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2109 gives compiler warnings in some environments. I've now done it "properly",
2112 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2113 (because of problems with the built-in one) was declared to return uschar *
2114 instead of char *, causing compiler failure.
2116 25. Fixed a file descriptor leak when processing alias/forward files.
2118 26. Fixed a minor format string issue in dbfn.c.
2120 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2122 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2123 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2126 29. When a sender address is verified, it is cached, to save repeating the test
2127 when there is more than one recipient in a message. However, when the
2128 verification involves a callout, it is possible for different callout
2129 options to be set for different recipients. It is too complicated to keep
2130 track of this in the cache, so now Exim always runs a verification when a
2131 callout is required, relying on the callout cache for the optimization.
2132 The overhead is duplication of the address routing, but this should not be
2135 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2136 to be verified with callout=postmaster, and the main callout worked but the
2137 postmaster check failed, the verification correctly failed. However, if a
2138 subsequent RCPT command asked for sender verification *without* the
2139 postmaster check, incorrect caching caused this verification also to fail,
2142 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2143 it was not caching the DNS options (qualify_single, search_parents) that
2144 were used when the lookup failed. A subsequent lookup with different
2145 options therefore always gave the same answer, though there were cases
2146 where it should not have. (Example: a "domains = !$mx_any" option on a
2147 dnslookup router: the "domains" option is always processed without any
2148 widening, but the router might have qualify_single set.) Now Exim uses the
2149 cached value only when the same options are set.
2151 32. Added John Jetmore's "exipick" utility to the distribution.
2153 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2154 than a timeout (e.g. a certificate is required, and is not provided), an
2155 Exim server now closes the connection immediately. Previously it waited for
2156 the client to close - but if the client is SSL, it seems that they each
2157 wait for each other, leading to a delay before one of them times out.
2159 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2160 maintained 0.8.x compatibility because I don't think many are using it, and
2161 it is clearly obsolete.
2163 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2166 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2167 client certificate was expired. A simple patch fixes this, though I don't
2168 understand the full logic of why the verify callback is called multiple
2171 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2172 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2173 which causes problems with some clients (such as the Certicom SSL Plus
2174 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2175 disables the coutermeasure allowing Eudora to connect."
2177 38. Exim was not checking that a write() to a log file succeeded. This could
2178 lead to Bad Things if a log got too big, in particular if it hit a file
2179 size limit. Exim now panics and dies if it cannot write to a log file, just
2180 as it does if it cannot open a log file.
2182 39. Modified OS/Makefile-Linux so that it now contains
2184 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2186 The two -D definitions ensure that Exim is compiled with large file
2187 support, which makes it possible to handle log files that are bigger than
2190 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2191 instance) a domain was checked against a named list that involved a lookup,
2192 causing $domain_data to be set, then another domain was checked against the
2193 same list, then the first domain was re-checked, the value of $domain_data
2194 after the final check could be wrong. In particular, if the second check
2195 failed, it could be set empty. This bug probably also applied to
2198 41. The strip_trailing_dot option was not being applied to the address given
2199 with the -f command-line option.
2201 42. The code for reading a message's header from the spool was incrementing
2202 $received_count, but never initializing it. This meant that the value was
2203 incorrect (doubled) while delivering a message in the same process in which
2204 it was received. In the most common configuration of Exim, this never
2205 happens - a fresh exec is done - but it can happen when
2206 deliver_drop_privilege is set.
2208 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2209 it now includes up to 150 characters of the unexpected data in the log
2212 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2213 and building data strings. The size of both of these buffers was 10 000
2214 bytes - far larger than anybody would *ever* want, thought I. Needless to
2215 say, somebody hit the limit. I have increased the maximum line length to
2216 20 000 and the maximum data length of concatenated lines to 100 000. I have
2217 also fixed two bugs, because there was no checking on these buffers. Tsk,
2218 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2219 buffer is too small.
2221 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2222 lsearch lookups. Now it does.
2224 46. When parsing a route_list item in a manualroute router, a fixed-length
2225 buffer was used for the list of hosts. I made this 1024 bytes long,
2226 thinking that nobody would ever have a list of hosts that long. Wrong.
2227 Somebody had a whole pile of complicated expansion conditions, and the
2228 string was silently truncated, leading to an expansion error. It turns out
2229 that it is easier to change to an unlimited length (owing to other changes
2230 that have happened since this code was originally written) than to build
2231 structure for giving a limitation error. The length of the item that
2232 expands into the list of hosts is now unlimited.
2234 47. The lsearch lookup could not handle data where the length of text line was
2235 more than 4095 characters. Such lines were truncated, leading to shortened
2236 data being returned. It should now handle lines of any length.
2238 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2239 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2242 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2243 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2244 no longer happens in comment lines. A list of replacements is now placed
2245 at the head of all of the source files, except those whose only change is
2246 to replace PERL_COMMAND in the very first #! line.
2248 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2249 messages on the queue, with a bottom-up merge sort, using code contributed
2250 by Michael Haardt. This should make operations like -bp somewhat faster on
2251 large queues. It won't affect queue runners, except when queue_run_in_order
2254 51. Installed eximstats 1.31 in the distribution.
2256 52. Added support for SRV lookups to the dnslookup router.
2258 53. If an ACL referred to $message_body or $message_body_end, the value was not
2259 reset for any messages that followed in the same SMTP session.
2261 54. The store-handling optimization for building very long strings was not
2262 differentiating between the different store pools. I don't think this
2263 actually made any difference in practice, but I've tidied it.
2265 55. While running the routers to verify a sender address, $sender_address
2266 was still set to the sender address. This is wrong, because when routing to
2267 send a bounce to the sender, it would be empty. Therefore, I have changed
2268 it so that, while verifying a sender address, $sender_address is set to <>.
2269 (There is no change to what happens when verifying a recipient address.)
2271 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2272 target A or AAAA records (if not already returned) without resetting the
2273 qualify_single or search_parents options of the DNS resolver. These are
2274 inappropriate in this case because the targets of MX and SRV records must
2275 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2276 target that, when qualified, matched something in the local domain. These
2277 two options are now turned off when doing these lookups.
2279 57. It seems that at least some releases of Reiserfs (which does not have the
2280 concept of a fixed number of inodes) returns zero and not -1 for the
2281 number of available inodes. This interacted badly with check_spool_inodes,
2282 which assumed that -1 was the "no such thing" setting. What I have done is
2283 to check that the total number of inodes is greater than zero before doing
2284 the test of how many are available.
2286 58. When a "warn" ACL statement has a log_message modifier, the message is
2287 remembered, and not repeated. This is to avoid a lot of repetition when a
2288 message has many recipients that cause the same warning to be written.
2289 Howewer, Exim was preserving the list of already written lines for an
2290 entire SMTP session, which doesn't seem right. The memory is now reset if a
2291 new message is started.
2293 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2294 debugging output unless log_rewrite was also set.
2296 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2297 of (int)(handle) when we know that handle contains (void *)(-1).
2299 61. The Exim daemon panic-logs an error return when it closes the incoming
2300 connection. However "connection reset by peer" seems to be common, and
2301 isn't really an error worthy of noting specially, so that particular error
2304 62. When Exim is trying to find all the local interfaces, it used to panic and
2305 die if the ioctl to get the interface flags failed. However, it seems that
2306 on at least one OS (Solaris 9) it is possible to have an interface that is
2307 included in the list of interfaces, but for which you get a failure error
2308 for this call. This happens when the interface is not "plumbed" into a
2309 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2310 failure of the "get flags" call assumes that the interface is down.
2312 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2313 makes life easier for people who are doing arithmetic on fields extracted
2314 from dates, where you often get leading zeros that should not be
2315 interpreted as octal.
2317 64. Added qualify_domain to the redirect router, to override the global
2320 65. If a pathologically long header line contained very many addresses (the
2321 report of this problem mentioned 10 000) and each of them was rewritten,
2322 Exim could use up a very large amount of memory. (It kept on making new
2323 copies of the header line as it rewrote, and never released the old ones.)
2324 At the expense of a bit more processing, the header rewriting function has
2325 been changed so that it no longer eats memory in this way.
2327 66. The generation of the Received: header has been moved from the time that a
2328 message starts to be received, to the time that it finishes. The timestamp
2329 in the Received: header should now be very close to that of the <= log
2330 line. There are two side-effects of this change:
2332 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2333 logged header lines no longer include the local Received: line, because
2334 it has not yet been created. The same applies to a copy of the message
2335 that is returned to a non-SMTP sender when a message is rejected.
2337 (b) When a filter file is tested using -bf, no additional Received: header
2338 is added to the test message. After some thought, I decided that this
2341 This change does not affect the value of $received_for. It is still set
2342 after address rewriting, but before local_scan() is called.
2344 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2346 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2347 gave an unhelpful panic error message, and a defer error. I have managed to
2348 change this behaviour so that it now rejects any supplied certificate,
2349 which seems right, as the list of acceptable certificates is empty.
2351 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2352 gave an unhelpful defer error. I have not managed to make this reject any
2353 supplied certificates, but the error message it gives is "no certificate
2354 supplied", which is not helpful.
2356 70. exigrep's output now also includes lines that are not associated with any
2357 message, but which match the given pattern. Implemented by a patch from
2358 Martin Sluka, which also tidied up the Perl a bit.
2360 71. Recipient callout verification, like sender verification, was using <> in
2361 the MAIL FROM command. This isn't really the right thing, since the actual
2362 sender may affect whether the remote host accepts the recipient or not. I
2363 have changed it to use the actual sender in the callout; this means that
2364 the cache record is now keyed on a recipient/sender pair, not just the
2365 recipient address. There doesn't seem to be a real danger of callout loops,
2366 since a callout by the remote host to check the sender would use <>.
2367 [SEE ABOVE: changed after hitting problems.]
2369 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2370 temporary errors. However, in the case of such a code being given after
2371 the end of a data transmission (i.e. after ".") Exim was failing to write
2372 a retry record for the message. (Yes, there was some broken host that was
2373 actually sending 8xx at this point.)
2375 73. An unknown lookup type in a host list could cause Exim to panic-die when
2376 the list was checked. (An example that provoked this was putting <; in the
2377 middle of a list instead of at the start.) If this happened during a DATA
2378 ACL check, a -D file could be left lying around. This kind of configuration
2379 error no longer causes Exim to die; instead it causes a defer errror. The
2380 incident is still logged to the main and panic logs.
2382 74. Buglet left over from Exim 3 conversion. The message "too many messages
2383 in one connection" was written to the rejectlog but not the mainlog, except
2384 when address rewriting (yes!) was being logged.
2386 75. Added write_rejectlog option.
2388 76. When a system filter was run not as root (that is, when system_filter_user
2389 was set), the values of the $n variables were not being returned to the
2390 main process; thus, they were not subsequently available in the $sn
2393 77. Added +return_path_on_delivery log selector.
2395 78. A connection timeout was being treated differently from recipients deferred
2396 when testing hosts_max_try with a message that was older than the host's
2397 retry timeout. (The host should not be counted, thus allowing all hosts to
2398 be tried at least once before bouncing.) This may have been the cause of an
2399 occasionally reported bug whereby a message would remain on the queue
2400 longer than the retry timeout, but would be bounced if a delivery was
2401 forced. I say "may" because I never totally pinned down the problem;
2402 setting up timeout/retry tests is difficult. See also the next item.
2404 79. The ultimate address timeout was not being applied to errors that involved
2405 a combination of host plus message (for example, a timeout on a MAIL
2406 command). When an address resolved to a number of possible hosts, and they
2407 were not all tried for each delivery (e.g. because of hosts_max_try), a
2408 message could remain on the queue longer than the retry timeout.
2410 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2413 81. Fixed an obscure SMTP outgoing bug which required at least the following
2414 conditions: (a) there was another message waiting for the same server;
2415 (b) the server returned 5xx to all RCPT commands in the first message so
2416 that the message was not completed; (c) the server dropped the connection
2417 or gave a negative response to the RSET that Exim sends to abort the
2418 transaction. The observed case was a dropped connection after DATA that had
2419 been sent in pipelining mode. That is, the server had advertised PIPELINING
2420 but was not implementing it correctly. The effect of the bug was incorrect
2421 behaviour, such as trying another host, and this could lead to a crash.
2427 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2428 and daemon.c were passed as pointers to ints; they should have been
2429 pointers to socklen_t variables (which are typically unsigned ints).
2431 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2434 3. Fixed a really odd bug that affected only the testing scheme; patching a
2435 certain fixed string in the binary changed the value of another string that
2436 happened to be identical to the end of the original first string.
2438 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2439 name", it returns that address as the IP address. On some operating
2440 systems (e.g. Solaris), it also passes back the IP address string as the
2441 "host name". However, on others (e.g. Linux), it passes back an empty
2442 string. Exim wasn't checking for this, and was changing the host name to an
2443 empty string, assuming it had been canonicized.
2445 5. Although rare, it is permitted to have more than one PTR record for a given
2446 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2447 all the names associated with an address, because they do in Solaris.
2448 However, it seems that they do not in Linux for data that comes from the
2449 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2450 I found this out when I moved to a new Linux workstation and tried to run
2451 the Exim test suite.
2453 To get round this problem I have changed the code so that it now does its
2454 own call to the DNS to look up PTR records when searching for a host name.
2455 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2456 addresses that are only in /etc/hosts are still found.
2458 This behaviour is, however, controlled by an option called host_lookup_
2459 order, which defaults to "bydns:byaddr". If people want to use the other
2460 order, or indeed, just use one or the other means of lookup, they can
2461 specify it in this variable.
2463 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2464 some operating systems, this comes back from gethostbyaddr() as an empty
2465 string, and this is what Exim used to test for. However, it seems that in
2466 other systems, "." is yielded. Exim now tests for this case too.
2468 7. The values of check_spool_space and check_log_space are now held internally
2469 as a number of kilobytes instead of an absolute number of bytes. If a
2470 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2471 kilobyte. This means that much larger values can be stored.
2473 8. Exim monitor: an attempt to get the action menu when not actually pointing
2474 at a message produces an empty menu entitled "No message selected". This
2475 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2476 no entries in it ("Shell widget menu has zero width and/or height"). So I
2477 have added a single, blank menu entry in this case.
2479 9. Added ${quote_local_part.
2481 10. MIME decoding is now applied to the contents of Subject: header lines when
2484 11. Now that a reference to $sender_host_address automatically causes a reverse
2485 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2486 host lookup before query-style lookups in lists that might use this
2487 variable. This has therefore been abolished, and the "net-" prefix is no
2488 longer necessary for query-style lookups.
2490 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2491 have been a typo for LFLAGS, so it has been changed.
2493 13. The install script calls Exim with "-C /dev/null" in order to find the
2494 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2495 to be output. Howeve, since Exim outputs its version number before the
2496 error, it didn't break the script. It just looked ugly. I fixed this by
2497 always allowing "-C /dev/null" if the caller is root.
2499 14. Ignore overlarge ACL variable number when reading spool file - insurance
2500 against a later release with more variables having written the file.
2502 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2503 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2504 this, as well as the form without the "IPv6" on the front.
2506 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2507 OS/Makefile-Darwin file.
2509 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2510 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2512 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2513 variable. However, some OS use ints for inodes. I've added cast to long int
2514 to get rid of the compiler warning.
2516 19. I had forgotten to lock out "/../" in configuration file names when
2517 ALT_CONFIG_PREFIX was set.
2519 20. Routers used for verification do not need to specify transports. However,
2520 if such a router generated a host list, and callout was configured, Exim
2521 crashed, because it could not find a port number from the (non-existent)
2522 transport. It now assumes port 25 in this circumstance.
2524 21. Added the -t option to exigrep.
2526 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2527 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2528 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2529 them is set, LOOKUP_LSEARCH is forced.
2531 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2533 24. Added sender and host information to the "rejected by local_scan()" log
2534 line; previously there was no indication of these.
2536 25. Added .include_if_exists.
2538 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2539 Linux. It turns out that not all file systems support this. Apparently some
2540 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2541 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2542 not supported on the file descriptor, is now ignored when Exim is trying to
2543 sync a directory. This applies only to Linux.
2545 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2547 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2548 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2551 29. Added support for maildirsize files from supplied patch (modified a bit).
2553 30. The use of :fail: followed by an empty string could lead Exim to respond to
2554 sender verification failures with (e.g.):
2556 550 Verification failed for <xxx>
2557 550 Sender verify failed
2559 where the first response line was missing the '-' that indicates it is not
2560 the final line of the response.
2562 31. The loop for finding the name of the user that called Exim had a hardwired
2563 limit of 10; it now uses the value of finduser_retries, which is used for
2564 all other user lookups.
2566 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2569 33. Exim was neglecting to zero errno before one call of strtol() when
2570 expanding a string and expecting an integer value. On some systems this
2571 resulted in spurious "integer overflow" errors. Also, it was casting the
2572 result into an int without checking.
2574 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2575 did not work. The code looks as if it has *never* worked, though it appears
2576 to have been documented since at least releast 1.62. I have made it work.
2578 35. The "timeout_DNS" error in retry rules, also documented since at least
2579 1.62, also never worked. As it isn't clear exactly what this means, and
2580 clearly it isn't a major issue, I have abolished the feature by treating it
2581 as "timeout", and writing a warning to the main and panic logs.
2583 36. The display of retry rules for -brt wasn't always showing the error code
2586 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2587 timeout_connect_A, timeout_connect_MX.
2589 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2590 to the empty sender.
2592 39. The daemon was not analysing the content of -oX till after it had closed
2593 stderr and disconnected from the controlling terminal. This meant that any
2594 syntax errors were only noted on the panic log, and the return code from
2595 the command was 0. By re-arranging the code a little, I've made the
2596 decoding happen first, so such errors now appear on stderr, and the return
2597 code is 1. However, the actual setting up of the sockets still happens in
2598 the disconnected process, so errors there are still only recorded on the
2601 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2602 connections (as happens on some IP stacks) was logged at start up time as
2603 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2604 it from "IPv6 and IPv4", which means that two separate sockets are being
2607 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2608 says whether AF_INET or AF_INET6 was passed as an argument.
2610 42. Exiwhat output was messed up when time zones were included in log
2613 43. Exiwhat now gives more information about the daemon's listening ports,
2614 and whether -tls-on-connect was used.
2616 44. The "port" option of the smtp transport is now expanded.
2618 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2619 silently ignored. Now an error message is written to the main and panic
2622 46. There's a new ACL modifier called "logwrite" which writes to a log file
2623 as soon as it is encountered.
2625 47. Added $local_user_uid and $local_user_gid at routing time.
2627 48. Exim crashed when trying to verify a sender address that was being
2630 49. Exim was recognizing only a space character after ".include". It now also
2631 recognizes a tab character.
2633 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2634 extracting the relevant information from the specification. The man page no
2635 longer contains scrambled data for the -d option, and I've added a section
2636 at the front about calling Exim under different names.
2638 51. Added "extra_headers" argument to the "mail" command in filter files.
2640 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2643 53. Installed eximstats 1.29.
2645 54. Added transport_filter_timeout as a generic transport option.
2647 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2648 Cc: header lines. This was required by RFC 822, but it not required by RFC
2651 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2652 incoming messages that did not have them. Now it does so only if the
2653 message originates locally, that is, if there is no associated remote host
2654 address. When Resent- header lines are present, this applies to the Resent-
2655 lines rather than the non-Resent- lines.
2657 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2658 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2660 58. Messages for configuration errors now include the name of the main
2661 configuration file - useful now that there may be more than one file in a
2662 list (.included file names were always shown).
2664 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2665 for those rare installations that do not start the daemon as root or run it
2666 setuid root. I've cut out the call to initgroups() if the daemon is not
2669 60. The Exim user and group can now be bound into the binary as text strings
2670 that are looked up at the start of Exim's processing.
2672 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2674 62. Added $mailstore_basename variable.
2676 63. Installed patch to sieve.c from Michael Haardt.
2678 64. When Exim failed to open the panic log after failing to open the main log,
2679 the original message it was trying to log was written to stderr and debug
2680 output, but if they were not available (the usual case in production), it
2681 was lost. Now it is written to syslog before the two lines that record the
2682 failures to open the logs.
2684 65. Users' Exim filters run in subprocesses under the user's uid. It is
2685 possible for a "deliver" command or an alias in a "personal" command to
2686 provoke an address rewrite. If logging of address rewriting is configured,
2687 this fails because the process is not running as root or exim. There may be
2688 a better way of dealing with this, but for the moment (because 4.30 needs
2689 to be released), I have disabled address rewrite logging when running a
2690 filter in a non-root, non-exim process.
2696 1. The buildconfig auxiliary program wasn't quoting the value set for
2697 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2698 not defined. This bug was masked in 4.22 by the effect that was fixed in
2701 2. Some messages that were rejected after a message id was allocated were
2702 shown as "incomplete" by exigrep. It no longer does this for messages that
2703 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2705 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2706 have allow_domain_literals set, the ID did not get logged in the <= line.
2707 Domain literals are now always recognized in Message-ID: header lines.
2709 4. The first argument for a ${extract expansion item is the key name or field
2710 number. Leading and trailing spaces in this item were not being ignored,
2711 causing some misleading effects.
2713 5. When deliver_drop_privilege was set, single queue runner processes started
2714 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2715 same command in the process it spins off) were not dropping privilege.
2717 6. When the daemon running as "exim" started a queue runner, it always
2718 re-executed Exim in the spun-off process. This is a waste of effort when
2719 deliver_drop_privilege is set. The new process now just calls the
2720 queue-runner function directly.
2726 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2729 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2730 ignored. Though the use of -oP was forcing the writing of a pid file, it
2731 was always written to the default place.
2733 3. If the message "no IP address found for host xxxx" is generated during
2734 incoming verification, it is now followed by identification of the incoming
2735 connection (so you can more easily find what provoked it).
2737 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2739 5. Added some features to "harden" Exim a bit more against certain attacks:
2741 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2742 be put in Local/Makefile. This is like the never_users runtime option,
2743 but it cannot be overridden. The default setting is "root".
2745 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2746 prefix string with which any file named in a -C command line option
2749 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2750 is retained for -C and -D only if the caller of Exim is root. Without
2751 it, the exim user may also use -C and -D and retain privilege.
2753 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2754 command line option is disabled.
2756 6. Macro names set by the -D option must start with an upper case letter, just
2757 like macro names defined in the configuration file.
2759 7. Added "dereference=" facility to LDAP.
2761 8. Two instances of the typo "uknown" in the source files are fixed.
2763 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2764 the Configure-Makefile script screwed up while processing it.
2766 10. Incorporated PCRE 4.4.
2768 11. The SMTP synchronization check was not operating right at the start of an
2769 SMTP session. For example, it could not catch a HELO sent before the client
2770 waited for the greeting. There is now a check for outstanding input at the
2771 point when the greeting is written. Because of the duplex, asynchronous
2772 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2773 way, but not yet received, when the check is performed.
2775 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2776 on TCP/IP sockets, because this apparently causes some broken clients to
2779 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2780 unchanged) from the Cygwin maintainer.
2782 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2783 was supported instead of testing for mbx. Effectively a typo.
2785 15. The spa authenticator server code was not checking that the input it
2786 received was valid base64.
2788 16. The debug output line for the "set" modifier in ACLs was not showing the
2789 name of the variable that was being set.
2791 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2793 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2794 lookup on its own. Something else was needed to trigger the lookup. For
2795 example, a match in host_lookup or the need for a host name in a host list.
2796 Now, if $sender_host_name is referenced and the host name has not yet been
2797 looked up, a lookup is performed. If the lookup fails, the variable remains
2798 empty, and $host_lookup_failed is set to "1".
2800 19. Added "eqi" as a case-independent comparison operator.
2802 20. The saslauthd authentication condition could segfault if neither service
2803 nor realm was specified.
2805 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2806 error message that was logged was misleading, and incoming SMTP
2807 connections were dropped. The message is now more accurate, and temporary
2808 errors are given to SMTP connections.
2810 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2811 (except RSET) with a 421 error, until QUIT is received. However, it was
2812 failing to send a response to QUIT.
2814 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2815 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2816 ACL is now run after the helo_try_verify_hosts code.
2818 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2819 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2820 case-independent, so other case variants are also recognized). Apparently
2821 some systems use these upper case variants.
2823 25. If more than two messages were waiting for the same host, and a transport
2824 filter was specified for the transport, Exim sent two messages over the
2825 same TCP/IP connection, and then failed with "socket operation on non-
2826 socket" when it tried to send the third.
2828 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2830 27. The extern definition of crypt16() in expand.c was not being excluded when
2831 the OS had its own crypt16() function.
2833 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2834 as a preferred synonym for return_size_limit, both as an option and as an
2837 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2839 30. Changed the default configuration ACL to relax the local part checking rule
2840 for addresses that are not in any local domains. For these addresses,
2841 slashes and pipe symbols are allowed within local parts, but the sequence
2842 /../ is explicitly forbidden.
2844 31. SPA server authentication was not clearing the challenge buffer before
2847 32. log_message in a "warn" ACL statement was writing to the reject log as
2848 well as to the main log, which contradicts the documentation and doesn't
2849 seem right (because no rejection is happening). So I have stopped it.
2851 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2852 However, I am unable to do any testing of this.
2854 34. Fixed an infelicity in the appendfile transport. When checking directories
2855 for a mailbox, to see if any needed to be created, it was accidentally
2856 using path names with one or more superfluous leading slashes; tracing
2857 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2859 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2860 contents are added to the log line that is written for every discarded
2861 recipient. (Previously a log_message setting was ignored.)
2863 36. The ${quote: operator now quotes the string if it is empty.
2865 37. The install script runs exim in order to find its version number. If for
2866 some reason other than non-existence or emptiness, which it checks, it
2867 could not run './exim', it was installing it with an empty version number,
2868 i.e. as "exim-". This error state is now caught, and the installation is
2871 38. An argument was missing from the function that creates an error message
2872 when Exim fails to connect to the socket for saslauthd authentication.
2873 This could cause Exim to crash, or give a corrupted message.
2875 39. Added isip, isip4, and isip6 to ${if conditions.
2877 40. The ACL variables $acl_xx are now saved with the message, and can be
2878 accessed later in routers, transports, and filters.
2880 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2881 strings in the file are not string-expanded.
2883 42. If a MAIL command specified a SIZE value that was too large to fit into an
2884 int variable, the check against message_size_limit failed. Such values are
2885 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2886 day this will have to be increased, but I don't think I want to be around
2887 when emails are that large.
2894 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2895 iconv() is not standard in FreeBSD.
2897 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2898 IPv6 enabled. The observed symptom was a segmentation fault on return from
2899 the function os_common_find_running_interfaces() in src/os.c.
2901 3. In the check_special_case() function in daemon.c I had used "errno" as an
2902 argument name, which causes warnings on some systems. This was basically a
2903 typo, since it was named "eno" in the comments!
2905 4. The code that waits for the clock to tick (at a resolution of some fraction
2906 of a second) so as to ensure message-id uniqueness was always waiting for
2907 at least one whole tick, when it could have waited for less. [This is
2908 almost certainly not relevant at current processor speeds, where it is
2909 unlikely to ever wait at all. But we try to future-proof.]
2911 5. The function that sleeps for a time interval that includes fractions of a
2912 second contained a race. It did not block SIGALRM between setting the
2913 timer, and suspending (a couple of lines later). If the interval was short
2914 and the sigsuspend() was delayed until after it had expired, the suspension
2915 never ended. On busy systems this could lead to processes getting stuck for
2918 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2919 process, before it forks any delivery processes. The open lookup caching
2920 mechanism meant that the open file or database connection was passed into
2921 the delivery process. The problem was that delivery processes always tidy
2922 up cached lookup data. This could cause a problem for the next delivery
2923 process started by the queue runner, because the external queue runner
2924 process does not know about the closure. So the next delivery process
2925 still has data in the lookup cache. In the case of a file lookup, there was
2926 no problem because closing a file descriptor in a subprocess doesn't affect
2927 the parent. However, if the lookup was caching a connection to a database,
2928 the connection was closed, and the second delivery process was likely to
2929 see errors such as "PGSQL: query failed: server closed the connection
2930 unexpectedly". The problem has been fixed by closing all cached lookups
2931 in a queue runner before running a delivery process.
2933 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2934 seem to have the "const" qualifier which it has on other OS. I've
2937 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2938 *of the same type* (client or server) with the same public name that an
2939 error should be diagnosed.
2941 9. When Exim looked up a host name for an IP address, but failed to find the
2942 original IP address when looking up the host name (a safety check), it
2943 output the message "<ip address> does not match any IP for NULL", which was
2944 confusing, to say the least. The bug was that the host name should have
2945 appeared instead of "NULL".
2947 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2948 user that is built into the binary, and the -C or -D options is used, root
2949 privilege is dropped before the configuration file is read. In addition,
2950 logging is switched to stderr instead of the normal log files. If the
2951 configuration then re-defines the Exim user, the unprivileged environment
2952 is probably not what is expected, so Exim logs a panic warning message (but
2955 However, if deliver_drop_privilege is set, the unprivileged state may well
2956 be exactly what is intended, so the warning has been cut out in that case,
2957 and Exim is allowed to try to write to its normal log files.
2963 1. smtp_return_error_details was not giving details for temporary sender
2964 or receiver verification errors.
2966 2. Diagnose a configuration error if two authenticators have the same public
2969 3. Exim used not to create the message log file for a message until the first
2970 delivery attempt. This could be confusing when incoming messages were held
2971 for policy or load reasons. The message log file is now created at the time
2972 the message is received, and an initial "Received" line is written to it.
2974 4. The automatically generated man page for command line options had a minor
2975 bug that caused no ill effects; however, a more serious problem was that
2976 the procedure for building the man page automatically didn't always
2977 operate. Consequently, release 4.20 contains an out-of-date version. This
2978 shouldn't happen again.
2980 5. When building Exim with embedded Perl support, the script that builds the
2981 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2982 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2984 6. The freeze_tell option was not being used for messages that were frozen on
2985 arrival, either by an ACL or by local_scan().
2987 7. Added the smtp_incomplete_transaction log selector.
2989 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2990 was accepting AUTH without a new EHLO.
2992 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2993 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2996 10. Logging of TCP/IP connections (when configured) now happens in the main
2997 daemon process instead of the child process, so that the TCP/IP connection
2998 count is more accurate (but it can never be perfect).
3000 11. The use of "drop" in a nested ACL was not being handled correctly in the
3001 outer ACL. Now, if condition failure induced by the nested "drop" causes
3002 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3003 or "require"), the connection is dropped.
3005 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3006 that yield "discard" can now be used with an "accept" or a "discard" verb,
3007 but an error is generated for any others (because I can't see a useful way
3008 to define what should happen).
3010 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3011 are now processed in the same way as lines in the Exim configuration file.
3012 In particular, continuation lines are supported.
3014 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3016 15. Added -ti meaning -t -i.
3018 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3019 domains, and warn by logging if others are found.
3021 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3022 returned from GIFCONF when Exim is trying to find the list of interfaces on
3023 a host. The code in os.c has been modified to copy each ifreq to an aligned
3024 structure in all cases.
3026 Also, in some cases, the returned ifreq's were being copied to a 'struct
3027 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3028 means the last couple of bytes of an IPv6 address could be chopped if the
3029 ifreq contained only a normal sockaddr (14 bytes storage).
3031 18. Named domain lists were not supported in the hosts_treat_as_local option.
3032 An entry such as +xxxx was not recognized, and was treated as a literal
3035 19. Ensure that header lines added by a DATA ACL are included in the reject log
3036 if the ACL subsequently rejects the message.
3038 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3039 MD5 (which is deprecated).
3041 21. When testing a filter file using -bf, Exim was writing a message when it
3042 took the sender from a "From " line in the message, but it was not doing so
3043 when it took $return_path from a Return-Path: header line. It now does.
3045 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3046 with a valid header line field name (a series of printing characters
3047 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3049 23. Changed "disc" in the source to "disk" to conform to the documentation and
3050 the book and for uniformity.
3052 24. Ignore Sendmail's -Ooption=value command line item.
3054 25. When execve() failed while trying to run a command in a pipe transport,
3055 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3056 could be confused with a return value of 69 from the command itself. This
3057 has been changed to 127, the value the shell returns if it is asked to run
3058 a non-existent command. The wording for the related log line suggests a
3059 non-existent command as the problem.
3061 26. If received_header_text expands to an empty string, do not add a Received:
3062 header line to the message. (Well, it adds a token one on the spool, but
3063 marks it "old" so that it doesn't get used or transmitted.)
3065 27. Installed eximstats 1.28 (addition of -nt option).
3067 28. There was no check for failure on the call to getsockname() in the daemon
3068 code. This can fail if there is a shortage of resources on the system, with
3069 ENOMEM, for example. A temporary error is now given on failure.
3071 29. Contrary to the C standard, it seems that in some environments, the
3072 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3073 program. Exim now does this explicitly; it affects the formatting of
3074 timestamps using strftime().
3076 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3077 complaints. I've now initialized all the variables, to avoid this.
3079 32. Header lines added by a system filter were not being "seen" during
3080 transport-time rewrites.
3082 33. The info_callback() function passed to OpenSSL is set up with type void
3083 (*)(SSL *, int, int), as described somewhere. However, when calling the
3084 function (actually a macro) that sets it up, the type void(*)() is
3085 expected. I've put in a cast to prevent warnings from picky compilers.
3087 34. If a DNS black list lookup found a CNAME record, but there were no A
3088 records associated with the domain it pointed at, Exim crashed.
3090 35. If a DNS black list lookup returned more than one A record, Exim ignored
3091 all but the first. It now scans all returned addresses if a particular IP
3092 value is being sought. In this situation, the contents of the
3093 $dnslist_value variable are a list of all the addresses, separated by a
3096 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3097 to accept a host name and all its aliases if the forward lookup for any of
3098 them yielded the IP address of the incoming connection. Now it accepts only
3099 those names whose forward lookup yields the correct IP address. Any other
3100 names are discarded. This closes a loophole whereby a rogue DNS
3101 administrator could create reverse DNS records to break through a
3102 wildcarded host restriction in an ACL.
3104 37. If a user filter or a system filter that ran in a subprocess used any of
3105 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3106 the wrong values were passed to the pipe command ($thisaddress had the
3107 value of $0, $0 had the value of $1, etc). This bug was introduced by
3108 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3110 38. Improved the line breaking for long SMTP error messages from ACLs.
3111 Previously, if there was no break point between 40 and 75 characters, Exim
3112 left the rest of the message alone. Two changes have been made: (a) I've
3113 reduced the minimum length to 35 characters; (b) if it can't find a break
3114 point between 35 and 75 characters, it looks ahead and uses the first one
3115 that it finds. This may give the occasional overlong line, but at least the
3116 remaining text gets split now.
3118 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3119 file descriptors might be low, and that setting 1000 would always raise it.
3120 It turns out that in some environments, the limit is already over 1000 and
3121 that lowering it causes trouble. So now Exim takes care not to decrease it.
3123 40. When delivering a message, the value of $return_path is set to $sender_
3124 address at the start of routing (routers may change the value). By an
3125 oversight, this default was not being set up when an address was tested by
3126 -bt or -bv, which affected the outcome if any router or filter referred to
3129 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3130 log only when "message" or "log_message" are set. However, if one of the
3131 conditions was an address verification, or a call to a nested ACL, the
3132 messages generated by the underlying test were being passed through. This
3133 no longer happens. The underlying message is available in $acl_verify_
3134 message for both "message" and "log_message" expansions, so it can be
3135 passed through if needed.
3137 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3138 new expansion $bh_ to give the encoded byte string without charset
3139 translation. Translation happens only if iconv() is available; HAVE_ICONV
3140 indicates this at build time. HEADERS_CHARSET gives the charset to
3141 translate to; headers_charset can change it in the configuration, and
3142 "headers charset" can change it in an individual filter file.
3144 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3145 that creates RFC 2047 encoded "words" labels them as that charset instead
3146 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3147 expansion operator; (ii) when Exim creates a From: line for a local
3148 message; (iii) when a header line is rewritten to include a "phrase" part.
3150 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3151 buggy, causing it to skip the first lines of messages whose message ID
3152 ended in 'D'. This would not have bitten before Exim release 4.14, because
3153 message IDs were unlikely to end in 'D' before then. The effect was to have
3154 incorrect size information for certain domains.
3156 45. #include "config.h" was missing at the start of the crypt16.c module. This
3157 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3160 46. If there was a timeout during a "random" callout check, Exim treated it as
3161 a failure of the random address, and carried on sending RSET and the real
3162 address. If the delay was just some slowness somewhere, the response to the
3163 original RCPT would be taken as a response to RSET and so on, causing
3164 mayhem of various kinds.
3166 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3167 when I implemented it. It didn't allow for the fact that some option values
3168 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3169 the right test for positive values.
3171 48. Domain names in DNS records are case-independent. Exim always looks them up
3172 in lower case. Some resolvers return domain names in exactly the case they
3173 appear in the zone file, that is, they may contain uppercase letters. Not
3174 all resolvers do this - some return always lower case. Exim was treating a
3175 change of case by a resolver as a change of domain, similar to a widening
3176 of a domain abbreviation. This triggered its re-routing code and so it was
3177 trying to route what was effectively the same domain again. This normally
3178 caused routing to fail (because the router wouldn't handle the domain
3179 twice). Now Exim checks for this case specially, and just changes the
3180 casing of the domain that it ultimately uses when it transmits the message
3183 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3186 50. If a filter generated a file delivery with a non-absolute name (possible if
3187 no home directory exists for the router), the forbid_file option was not
3190 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3191 the existing equality matching.
3193 52. Exim was using ints instead of ino_t variables in some places where it was
3194 dealing with inode numbers.
3196 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3197 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3198 called TMPDIR, and if it finds it is different, it changes its value.
3200 54. The smtp_printf() function is now made available to local_scan() so
3201 additional output lines can be written before returning. There is also an
3202 smtp_fflush() function to enable the detection of a dropped connection.
3203 The variables smtp_input and smtp_batched_input are exported to
3206 55. Changed the default runtime configuration: the message "Unknown user"
3207 has been removed from the ACL, and instead placed on the localuser router,
3208 using the cannot_route_message feature. This means that any verification
3209 failures that generate their own messages won't get overridden. Similarly,
3210 the "Unrouteable address" message that was in the ACL for unverifiable
3211 relay addresses has also been removed.
3213 56. Added hosts_avoid_esmtp to the smtp transport.
3215 57. The exicyclog script was not checking for the esoteric option
3216 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3217 will work only if exicyclog is run under the appropriate euid.
3219 58. Following a discussion on the list, the rules by which Exim recognises line
3220 endings on incoming messages have been changed. The -dropcr and drop_cr
3221 options are now no-ops, retained only for backwards compatibility. The
3222 following line terminators are recognized: LF CRLF CR. However, special
3223 processing applies to CR:
3225 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3226 nor a local message in the state where . is a terminator.
3228 (ii) If a bare CR is encountered in a header line, an extra space is added
3229 after the line terminator so as not to end the header. The reasoning
3230 behind this is that bare CRs in header lines are most likely either
3231 to be mistakes, or people trying to play silly games.
3233 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3234 was being incorrectly given as 18 bytes larger than it should have been.
3235 This is a VOB (very old bug).
3237 60. This may never have affected anything current, but just in case it has:
3238 When the local host is found other than at the start of a list of hosts,
3239 the local host, those with the same MX, and any that follow, are discarded.
3240 When the list in question was part of a longer list of hosts, the following
3241 hosts (not currently being processed) were also being discarded. This no
3242 longer happens. I'm not sure if this situation could ever has previously
3245 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3248 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3249 header. This is something that is recommended in a new Internet Draft, and
3250 is something that is documented as being done by Sendmail. There are two
3251 possible values. For messages generated by the autoreply transport, Exim
3254 Auto-Submitted: auto-replied
3256 whereas for all other generated messages (e.g. bounces) it adds
3258 Auto-Submitted: auto-generated
3260 63. The "personal" condition in filters now includes a test for the
3261 Auto-Submitted: header. If it contains the string "auto-" the message it
3262 not considered personal.
3264 64. Added rcpt_include_affixes as a generic transport option.
3266 65. Added queue_only_override (default true).
3268 66. Added the syslog_duplication option.
3270 67. If what should have been the first header line of a message consisted of
3271 a space followed by a colon, Exim was mis-interpreting it as a header line.
3272 It isn't of course - it is syntactically invalid and should therefore be
3273 treated as the start of the message body. The misbehaviour could have
3274 caused a number of strange effects, including loss of data in subsequent
3275 header lines, and spool format errors.
3277 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3278 client host had authenticated. This control can now be exercised by an ACL
3279 for more flexibility.
3281 69. By default, callouts do not happen when testing with -bh. There is now a
3282 variant, -bhc, which does actually run the callout code, including
3283 consulting and updating the callout cache.
3285 70. Added support for saslauthd authentication, courtesy of Alexander
3288 71. If statvfs() failed on the spool or log directories while checking their
3289 size for availability, Exim confusingly gave the error "space shortage".
3290 Furthermore, in debugging mode it crashed with a floating point exception.
3291 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3292 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3293 serious problem, Exim now writes to the main and panic logs when this
3294 happens, with details of the failure. It then refuses to accept the
3295 incoming message, giving the message "spool directory problem" or "log
3296 directory problem" with a 421 code for SMTP messages.
3298 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3299 0, 1, and 2 exist, because some OS complain for execs without them (see
3300 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3301 descriptors. However, the code omitted to check that the open succeeded,
3302 causing mysterious errors if for some reason the permissions on /dev/null
3303 got screwed. Now Exim writes a message to the main and panic logs, and
3304 bombs out if it can't open /dev/null.
3306 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3307 interact so that it is all more flexible. It is supposed to remain
3308 backwards compatible. Also added extra_local_interfaces.
3310 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3311 to bomb out with an assertion failure - to the client this appears as a
3312 connection drop. This problem occurs in the part of the code that was taken
3313 from the Samba project. Fortunately, the assertion is in a very simple
3314 function, so I have fixed this by reproducing the function inline in the
3315 one place where it is called, and arranging for authentication to fail
3316 instead of killing the process with assert().
3318 75. The SPA client code was not working when the server requested OEM rather
3319 than Unicode encoding.
3321 76. Added code to make require_files with a specific uid setting more usable in
3322 the case where statting the file as root fails - usually a non-root-mounted
3323 NFS file system. When this happens and the failure is EACCES, Exim now
3324 forks a subprocess and does the per-uid checking as the relevant uid.
3326 77. Added process_log_path.
3328 78. If log_file_path was not explicitly set, a setting of check_log_space or
3329 check_log_inodes was ignored.
3331 79. If a space check for the spool or log partitions fails, the incident is now
3332 logged. Of course, in the latter case the data may get lost...
3334 80. Added the %p formatting code to string_format() so that it can be used to
3335 print addresses in debug_print(). Adjusted all the address printing in the
3336 debugging in store.c to use %p rather than %d.
3338 81. There was a concern that a line of code in smtp_in.c could overflow a
3339 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3340 initially expressed, the concern was not well-founded, because trailing
3341 spaces are removed early. However, if the trailing spaces were followed by
3342 a NULL, they did not get removed, so the overflow was possible. Two fixes
3345 (a) I re-wrote the offending code in a cleaner fashion.
3346 (b) If an incoming SMTP command contains a NULL character, it is rejected
3349 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3350 runs initgroups(), so that if the Exim user is in any additional groups,
3351 they will be used during message reception.
3357 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob