1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.234 2005/09/16 14:44:11 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added the "success_on_redirect" address verification option. See
10 NewStuff for rationale and an example.
12 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
14 PH/02 Patch to exigrep to allow it to work on syslog lines.
16 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
17 fread() to skip over the body file's header line, because in Cygwin the
18 header line is locked and is inaccessible.
20 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
21 co-exist for some time) to make it clear that it is the Exim ID that is
22 referenced, not the Message-ID: header line.
24 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
25 string_format() function, because snprintf() does not exist on all
28 PH/06 The use of forbid_filter_existstest now also locks out the use of the
29 ${stat: expansion item.
31 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
32 protocol synchronization error", to keep the pedants happy.
34 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
35 well as for IRIX systems, when gcc is being used. See the host.c source
38 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
40 PH/10 Named domain lists were not working if used in a queue_smtp_domains
43 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
44 transport and to the smtp transport in LMTP mode.
46 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
48 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
49 run a filter in a subprocess. This could lead to confusion in subsequent
50 lookups in the parent process. There should also be a search_tidyup() at
51 the end of the subprocess.
53 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
54 only if the host matched helo_try_verify_hosts, which caused the
55 verification to occur when the EHLO/HELO command was issued. The ACL just
56 tested the remembered result. Now, if a previous verification attempt has
57 not happened, "verify = helo" does it there and then.
59 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
61 TK/03 Fix log output including CR from clamd.
63 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
64 exist provoked a memory error which could cause a segfault.
66 PH/15 Installed PCRE 6.2
68 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
70 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
71 of the problem. Specifically, suggested +O2 rather than +O1 for the
74 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
76 PH/20 If a delivery was routed to a non-standard port by means of an SRV
77 record, the port was not correctly logged when the outgoing_port log
78 selector was set (it logged the transort's default port).
80 PH/21 Added support for host-specific ports to manualroute, queryprogram,
81 fallback_hosts, and "hosts" in the smtp transport.
83 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
84 host errors such as "Connection refused".
86 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
87 authentication with radiusclient 0.4.9:
89 - Error returned from rc_read_config was caught wrongly
90 - Username/password not passed on to radius server due to wrong length.
92 The presumption is that some radiusclient API changes for 4.51/PH/17
93 were not taken care of correctly. The code is still untested by me (my
94 Linux distribution still has 0.3.2 of radiusclient), but it was
95 contributed by a Radius user.
97 PH/24 When doing a callout, the value of $domain wasn't set correctly when
98 expanding the "port" option of the smtp transport.
100 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
101 while reading a MIME header. Thanks to Tom Hughes for a patch.
103 PH/24 Include config.h inside local_scan.h so that configuration settings are
106 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
107 that in an ACL for RCPT (for example), you can examine exactly what was
110 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
111 commands, but it was not correctly comparing the address with the actual
112 client host address. Thus, it would show the EHLO address in Received:
113 header lines when this was not necessary.
115 PH/27 Added the % operator to ${eval:}.
117 PH/28 Exim tries to create and chdir to its spool directory when it starts;
118 it should be ignoring failures (because with -C, for example, it has lost
119 privilege). It wasn't ignoring creation failures other than "already
122 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
125 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
126 because an input line was too long, either on its own, or by virtue of
127 too many continuations, the temporary file was not being removed, and the
128 return code was incorrect.
130 PH/31 Missing "BOOL" in function definition in filtertest.c.
132 PH/32 Applied Sieve patches from the maintainer.
134 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
136 PH/33 Added "verify = not_blind".
138 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
139 Local/Makefile (with some defaults set). These are used in built scripts
140 such as exicyclog, but they have never been used in the exim_install
141 script (though there are many overriding facilities there). I have
142 arranged that the exim_install script now takes note of these two
145 PH/35 Installed configuration files for Dragonfly.
147 PH/36 When a locally submitted message by a trusted user did not contain a
148 From: header, and the sender address was obtained from -f or from an SMTP
149 MAIL command, and the trusted user did not use -F to supply a sender
150 name, $originator_name was incorrectly used when constructing a From:
151 header. Furthermore, $originator_name was used for submission mode
152 messages from external hosts without From: headers in a similar way,
153 which is clearly wrong.
155 PH/37 Added control=suppress_local_fixups.
157 PH/38 When log_selector = +received_sender was set, and the addition of the
158 sender made the log line's construction buffer exactly full, or one byte
159 less than full, an overflow happened when the terminating "\n" was
162 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
163 when the result of a list match is failure because a DNS lookup failed.
165 PH/40 RM_COMMAND is now used in the building process.
167 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
168 the "build-* directories that it finds.
170 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
171 address in a domain literal was a prefix of an interface address.
173 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
174 when verifying a sender address, unless rewrite_headers is false.
176 PH/44 Wrote a long comment about why errors_to addresses are verified as
177 recipients, not senders.
179 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
180 the ratelimit ACL was added.
182 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
184 PH/46 Added extra information about PostgreSQL errors to the error string.
186 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
187 avoiding the need to install special zones in a real server. This is
188 backwards compatible; if it can't find the fake resolver, it drops back.
189 Thus, both old and new test suites can be run.
191 TF/02 Added util/ratelimit.pl
193 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
194 clock is set back in time.
196 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
197 Candler <B.Candler@pobox.com>.
199 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
200 OK for addresses that are the result of redirections.
202 PH/48 A number of further additions for the benefit of the new test suite,
203 including a fake gethostbyname() that interfaces to the fake DNS resolver
210 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
212 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
213 command itself ended in error, the underlying message about the transport
214 filter timeout was being overwritten with the pipe command error. Now the
215 underlying error message should be appended to the second error message.
217 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
219 PH/02 Reduce the amount of output that "make" produces by default. Full output
220 can still be requested.
222 PH/03 The warning log line about a condition test deferring for a "warn" verb
223 was being output only once per connection, rather than after each
224 occurrence (because it was using the same function as for successful
225 "warn" verbs). This seems wrong, so I have changed it.
227 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
228 it should not have, which might have caused a crash in the right
229 circumstances, but probably never did.
231 PH/04 Installed a modified version of Tony Finch's patch to make submission
232 mode fix the return path as well as the Sender: header line, and to
233 add a /name= option so that you can make the user's friendly name appear
236 TF/03 Added the control = fakedefer ACL modifier.
238 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
239 Mark Lowes for thorough testing.
241 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
243 TK/03 Merged latest SRS patch from Miles Wilton.
245 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
246 with the definition in sysexits.h (which is #included earlier).
247 Fortunately, Exim does not actually use EX_OK. The code used to try to
248 preserve the sysexits.h value, by assumimg that macro definitions were
249 scanned for macro replacements. I have been disabused of this notion,
250 so now the code just undefines EX_OK before #including unistd.h.
252 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
253 in the smtp transport. When a block could not be written in a single
254 write() function, the timeout was being re-applied to each part-write.
255 This seems wrong - if the receiver was accepting one byte at a time it
256 would take for ever. The timeout is now adjusted when this happens. It
257 doesn't have to be particularly precise.
259 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
260 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
262 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
263 without a domain if the check to <postmaster@domain> fails.
265 SC/01 Eximstats: added -xls and the ability to specify output files
266 (patch written by Frank Heydlauf).
268 SC/02 Eximstats: use FileHandles for outputing results.
270 SC/03 Eximstats: allow any combination of xls, txt, and html output.
272 SC/04 Eximstats: fixed display of large numbers with -nvr option
274 SC/05 Eximstats: fixed merging of reports with empty tables.
276 SC/06 Eximstats: added the -include_original_destination flag
278 SC/07 Eximstats: removed tabs and trailing whitespace.
280 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
282 TK/06 MBOX spool code: Add real "From " MBOX separator line
283 so the .eml file is really in mbox format (even though
284 most programs do not really care). Patch from Alex Miller.
286 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
287 The latter is generated from $received_to and is only set if the
288 message has one envelope recipient. SA can use these headers,
289 obviously out-of-the-box. Patch from Alex Miller.
291 PH/08 The ${def test on a variable was returning false if the variable's
292 value was "0", contrary to what the specification has always said!
293 The result should be true unless the variable is empty.
295 PH/09 The syntax error of a character other than { following "${if
296 def:variable_name" (after optional whitespace) was not being diagnosed.
297 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
298 accidental colon was present, for example, could give incorrect results.
300 PH/10 Tidied the code in a number of places where the st_size field of a stat()
301 result is used (not including appendfile, where other changes are about
304 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
305 This involved changing a lot of size variables from int to off_t. It
306 should work with maildirs and everything.
308 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
309 spamd dying while we are connected to it.
311 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
314 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
315 function for BDB 4.3.
317 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
319 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
320 the sources for PCRE have been re-arranged and more files are now
323 PH/15 The code I had for printing potentially long long variables in PH/11
324 above was not the best (it lost precision). The length of off_t variables
325 is now inspected at build time, and an appropriate printing format (%ld
326 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
327 to be "long long int" or "long int". This is needed for the internal
328 formatting function string_vformat().
330 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
331 the configuration file to be ":syslog", then the script "guesses" where
332 the logs files are, rather than using the compiled in default. In our
333 case the guess is not the same as the compiled default, so the script
334 suddenly stopped working when I started to use syslog. The patch checks
335 to see if log_file_path is "". If so, it attempts to read it from exim
336 with no configuration file to get the compiled in version, before it
337 falls back to the previous guessing code."
339 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
340 implementing BATV in an Exim configuration. See NewStuff for the gory
343 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
344 Makefile that are specific to HP-UX.
346 PH/18 If the "use_postmaster" option was set for a recipient callout together
347 with the "random" option, the postmaster address was used as the MAIL
348 FROM address for the random test, but not for the subsequent recipient
349 test. It is now used for both.
351 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
352 patch removes a few documentation additions to RFC 3028, because the
353 latest draft now contains them. It adds the new en;ascii-case comparator
354 and a new error check for 8bit text in MIME parts. Comparator and
355 require names are now matched exactly. I enabled the subaddress
356 extension, but it is not well tested yet (read: it works for me)."
358 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
359 rework some of the code of TK/09 above to avoid the hardwired use of
360 "%lld" and "long long". Replaced the call to snprintf() with a call to
363 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
364 records point to non-existent hosts", "retry timeout exceeded", and
365 "retry time not reached for any host after a long failure period".
367 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
368 experimental DomainKeys support:
370 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
371 (2) On an error such as an illegally used "control", the wrong name for
372 the control was given.
374 These problems did NOT occur unless DomainKeys support was compiled.
376 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
378 PH/24 Added ${if match_ip condition.
380 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
381 databases so that it will be absolutely obvious if a crash occurs in the
382 DB library. This is a regular occurrence (often caused by mis-matched
385 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
386 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
387 picked up on a user's system that detects such things. There doesn't seem
388 to be a gcc warning option for this - only an attribute that has to be
389 put on the function's prototype. It seems that in Fedora Core 4 they have
390 set this on a number of new functions. No doubt there will be more in due
393 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
394 specify a transport. However, if an address that was verified by such a
395 router was the subject of a callout, Exim crashed because it tried to
396 read the rcpt_include_affixes from the non-existent transport. Now it
397 just assumes that the setting of that option is false. This bug was
398 introduced by 4.51/PH/31.
400 PH/28 Changed -d+all to exclude +memory, because that information is very
401 rarely of interest, but it makes the output a lot bigger. People tend to
402 do -d+all out of habit.
404 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
405 code in os-type was giving problems when libc.so lives in lib64, like on
408 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
409 aren't the modern standard, and it seems that some systems' include files
410 don't always have them. Exim was already checking for some of the newer
411 ones like T_AAAA, and defining it itself. I've added checks for all the
412 record types that Exim uses.
414 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
415 not automatically generating a new one, as it is supposed to. This
416 prevented TLS from working. If the file did exist, but contained invalid
417 data, a new version was generated, as expected. It was only the case of a
418 non-existent file that was broken.
420 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
421 with a change in libdomainkeys > 0.64.
423 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
424 from DNS. If the selector record carries the flag, it now has
425 precedence over the domain-wide flag.
427 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
429 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
430 the use of a port name that isn't defined in /etc/services) occurred, the
431 message was deferred as in a normal delivery, and thus remained on the
432 spool, instead of being failed because of the mua_wrapper setting. This
433 is now fixed, and I tidied up some of the mua_wrapper messages at the
436 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
437 the messages in a hash of arrays rather than using individual hashes.
438 This is a bit cleaner and results in dramatic memory savings, albeit
439 at a slight CPU cost.
441 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
442 as requested by Marc Sherman.
444 SC/10 Eximstats: added histograms for user specified patterns as requested
447 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
449 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
450 fopen() in the content-scanning modules that did not already have it.
456 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
457 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
459 TK/02 Fix ACL "control" statement not being available in MIME ACL.
461 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
463 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
464 to test Sieve filters that use "vacation".
466 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
467 that changes the way the GnuTLS parameters are stored in the cache file.
468 The new format can be generated externally. For backward compatibility,
469 if the data in the cache doesn't make sense, Exim assumes it has read an
470 old-format file, and it generates new data and writes a new file. This
471 means that you can't go back to an older release without removing the
474 PH/03 A redirect router that has both "unseen" and "one_time" set does not
475 work if there are any delivery delays because "one_time" forces the
476 parent to be marked "delivered", so its unseen clone is never tried
477 again. For this reason, Exim now forbids the simultaneous setting of
480 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
481 redirected to themselves ("homonym" addresses). Read the long ChangeLog
482 entry if you want to know the details. The fix, however, neglected to
483 consider the case when local delivery batching is involved. The test for
484 "previously delivered" was not happening when checking to see if an
485 address could be batched with a previous (undelivered) one; under
486 certain circumstances this could lead to multiple deliveries to the same
489 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
490 in its include files, and this causes problems building Exim.
492 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
493 header_syntax) but Exim was just ignoring anything given after a slash.
494 In particular, this caused confusion with an attempt to use "verify =
495 reverse_host_lookup/defer_ok". An error is now given when options are
496 supplied for verify items that do not have them. (Maybe reverse_host_
497 lookup should have a defer_ok option, but that's a different point.)
499 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
500 defined by RFC 821) to 2048, because there were problems with some AUTH
501 commands, and RFC 1869 says the size should be increased for extended
502 SMTP commands that take arguments.
504 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
507 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
508 "unknown" error; now it says that the functionality isn't in the binary.
510 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
511 an address' error message when a string expansion fails (syntax or
512 whatever). Otherwise the password may appear in the log. Following change
513 PH/42 below, there is no longer a chance of it appearing in a bounce
516 PH/11 Installed exipick version 20050225.0 from John Jetmore.
518 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
519 of its addresses was ever tried. (Bugzilla bug #2.)
521 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
522 the result incorrectly in the debug output. (It correctly added a newline
523 to what was transported.)
525 TF/01 Added $received_time.
527 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
528 commented out examples of how to interface to a virus scanner and to
529 SpamAssassin. Also added commented examples of av_scanner and
530 spamd_address settings.
532 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
533 and controls are allowed in which ACLs. There were a couple of minor
534 errors. Some of the entries in the conditions table (which is a table of
535 where they are NOT allowed) were getting very unwieldy; rewrote them as a
536 negation of where the condition IS allowed.
538 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
540 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
541 header file does not have a version number, so I've had to invent a new
542 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
543 API. The code is untested by me (my Linux distribution still has 0.3.2 of
544 radiusclient), but it was contributed by a Radius user.
546 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
547 files or directories, for OpenSSL.
549 PH/19 When an Exim process that is running as root has to create an Exim log
550 file, it does so in a subprocess that runs as exim:exim so as to get the
551 ownership right at creation (otherwise, other Exim processes might see
552 the file with the wrong ownership). There was no test for failure of this
553 fork() call, which would lead to the process getting stuck as it waited
554 for a non-existent subprocess. Forks do occasionally fail when resources
555 run out. I reviewed all the other calls to fork(); they all seem to check
558 PH/20 When checking for unexpected SMTP input at connect time (before writing
559 the banner), Exim was not dealing correctly with a non-positive return
560 from the read() function. If the client had disconnected by this time,
561 the result was a log entry for a synchronization error with an empty
562 string after "input=" when read() returned zero. If read() returned -1
563 (an event I could not check), uninitialized data bytes were printed.
564 There were reports of junk text (parts of files, etc) appearing after
567 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
569 PH/22 Added support for macro redefinition, and (re)definition in between
570 driver and ACL definitions.
572 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
573 forgetting to use the resulting value; it was using the unexpanded value.
575 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
576 hadn't been configured. The fix is from Juergen Kreileder, who
577 understands it better than I do:
579 "Here's what I see happening with three configured cyrus_sasl
580 authenticators configured (plain, login, cram-md5):
582 On startup auth_cyrus_sasl_init() gets called for each of these.
583 This means three calls to sasl_listmech() without a specified mech_list.
584 => SASL tests which mechs of all available mechs actually work
585 => three warnings about OTP not working
586 => the returned list contains: plain, login, cram-md5, digest-md5, ...
588 With the patch, sasl_listmech() also gets called three times. But now
589 SASL's mech_list option is set to the server_mech specified in the the
590 authenticator. Or in other words, the answer from sasl_listmech()
591 gets limited to just the mech you're testing for (which is different
593 => the return list contains just 'plain' or 'login', 'cram-md5' or
594 nothing depending on the value of ob->server_mech.
596 I've just tested the patch: Authentication still works fine,
597 unavailable mechs specified in the exim configuration are still
598 caught, and the auth.log warnings about OTP are gone."
600 PH/25 When debugging is enabled, the contents of the command line are added
601 to the debugging output, even when log_selector=+arguments is not
604 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
605 answer is "GNU", and only if the return is "GNU/something" is the answer
608 PH/27 $acl_verify_message is now set immediately after the failure of a
609 verification in an ACL, and so is available in subsequent modifiers. In
610 particular, the message can be preserved by coding like this:
612 warn !verify = sender
613 set acl_m0 = $acl_verify_message
615 Previously, $acl_verify_message was set only while expanding "message"
616 and "log_message" when a very denied access.
618 PH/28 Modified OS/os.c-Linux with
620 -#ifndef OS_LOAD_AVERAGE
621 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
623 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
624 nomenclature these days.)
626 PH/29 Installed patch from the Sieve maintainer that adds the options
627 sieve_useraddress and sieve_subaddress to the redirect router.
629 PH/30 In these circumstances:
630 . Two addresses routed to the same list of hosts;
631 . First host does not offer TLS;
632 . First host accepts first address;
633 . First host gives temporary error to second address;
634 . Second host offers TLS and a TLS session is established;
635 . Second host accepts second address.
636 Exim incorrectly logged both deliveries with the TLS parameters (cipher
637 and peerdn, if requested) that were in fact used only for the second
640 PH/31 When doing a callout as part of verifying an address, Exim was not paying
641 attention to any local part prefix or suffix that was matched by the
642 router that accepted the address. It now behaves in the same way as it
643 does for delivery: the affixes are removed from the local part unless
644 rcpt_include_affixes is set on the transport.
646 PH/32 Add the sender address, as F=<...>, to the log line when logging a
647 timeout during the DATA phase of an incoming message.
649 PH/33 Sieve envelope tests were broken for match types other than :is. I have
650 applied a patch sanctioned by the Sieve maintainer.
652 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
653 the uid or gid is negative. A case of a negative gid caused this to be
654 noticed. The fix allows for either to be negative.
656 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
657 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
660 PH/36 The change PH/12 above was broken. Fixed it.
662 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
663 the grounds that routing the same address twice would always produce the
664 same answer. This might have been true once, but it is certainly no
665 longer true now. Routing a child address may depend on the previous
666 routing that produced that child. Some complicated redirection strategies
667 went wrong when messages had multiple recipients, and made Exim's
668 behaviour dependent on the order in which the addresses were given.
670 I have moved the duplicate checking until after the routing is complete.
671 Exim scans the addresses that are assigned to local and remote
672 transports, and removes any duplicates. This means that more work will be
673 done, as duplicates will always all be routed, but duplicates are
674 presumably rare, so I don't expect this is of any significance.
676 For deliveries to pipes, files, and autoreplies, the duplicate checking
677 still happens during the routing process, since they are not going to be
680 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
681 It corrects a timeout issue with spamd. This is Ian's comment: "The
682 background is that sometimes spamd either never reads data from a
683 connection it has accepted, or it never writes response data. The exiscan
684 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
685 blindly assumes that writes won't block so it may never time out."
687 PH/39 Allow G after quota size as well as K and M.
689 PH/40 The value set for $authenticated_id in an authenticator may not contain
690 binary zeroes or newlines because the value is written to log lines and
691 to spool files. There was no check on this. Now the value is run through
692 the string_printing() function so that such characters are converted to
693 printable escape sequences.
695 PH/41 $message_linecount is a new variable that contains the total number of
696 lines in the message. Compare $body_linecount, which is the count for the
699 PH/42 Exim no longer gives details of delivery errors for specific addresses in
700 bounce and delay warning messages, except in certain special cases, which
703 (a) An SMTP error message from a remote host;
704 (b) A message specified in a :fail: redirection;
705 (c) A message specified in a "fail" command in a system filter;
706 (d) A message specified in a FAIL return from the queryprogram router;
707 (e) A message specified by the cannot_route_message router option.
709 In these cases only, Exim does include the error details in bounce and
710 warning messages. There are also a few cases where bland messages such
711 as "unrouteable address" or "local delivery error" are given.
713 PH/43 $value is now also set for the "else" part of a ${run expansion.
715 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
716 being worked on, but at least Exim now implements the latest version to
719 PH/45 In a pipe transport, although a timeout while waiting for the pipe
720 process to complete was treated as a delivery failure, a timeout while
721 writing the message to the pipe was logged, but erroneously treated as a
722 successful delivery. Such timeouts include transport filter timeouts. For
723 consistency with the overall process timeout, these timeouts are now
724 treated as errors, giving rise to delivery failures by default. However,
725 there is now a new Boolean option for the pipe transport called
726 timeout_defer, which, if set TRUE, converts the failures into defers for
727 both kinds of timeout. A transport filter timeout is now identified in
730 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
731 systems where "make" and "gmake" are different, calling "gmake" at top
732 level broke things. I've arranged for the value of $(MAKE) to be passed
733 from the Makefile to this script so that it can call the same version of
737 A note about Exim versions 4.44 and 4.50
738 ----------------------------------------
740 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
741 changes of various kinds. As a consequence, a big documentation update was
742 needed. This delayed the release for rather longer than seemed good, especially
743 in the light of a couple of (minor) security issues. Therefore, the changes
744 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
745 release. So 4.44 and 4.50 are in effect two different branches that both start
748 I have left the 4.50 change log unchanged; it contains all the changes since
749 4.43. The change log for 4.44 is below; many of its items are identical to
750 those for 4.50. This seems to be the most sensible way to preserve the
751 historical information.
757 1. Minor wording change to the doc/README.SIEVE file.
759 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
760 computation of the current number of files was incorrect.
762 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
763 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
766 4. Give more explanation in the error message when the command for a transport
767 filter fails to execute.
769 5. There are several places where Exim runs a non-Exim command in a
770 subprocess. The SIGUSR1 signal should be disabled for these processes. This
771 was being done only for the command run by the queryprogram router. It is
772 now done for all such subprocesses. The other cases are: ${run, transport
773 filters, and the commands run by the lmtp and pipe transports.
775 6. Added CONFIGURE_GROUP build-time option.
777 7. Some older OS have a limit of 256 on the maximum number of file
778 descriptors. Exim was using setrlimit() to set 1000 as a large value
779 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
780 systems. I've change it so that if it can't get 1000, it tries for 256.
782 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
783 was an oversight, and furthermore, ever since the addition of extra
784 controls (e.g. 4.43/32), the checks on when to allow different forms of
785 "control" were broken. There should now be diagnostics for all cases when a
786 control that does not make sense is encountered.
788 9. Added the /retain_sender option to "control=submission".
790 10. $recipients is now available in the predata ACL (oversight).
792 11. Tidy the search cache before the fork to do a delivery from a message
793 received from the command line. Otherwise the child will trigger a lookup
794 failure and thereby defer the delivery if it tries to use (for example) a
795 cached ldap connection that the parent has called unbind on.
797 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
798 of $address_data from the recipient verification was clobbered by the
801 13. The value of address_data from a sender verification is now available in
802 $sender_address_data in subsequent conditions in the ACL statement.
804 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
806 15. Added a new option "connect=<time>" to callout options, to set a different
809 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
810 was its contents. (It was OK if the option was not defined at all.)
812 17. A "Completed" log line is now written for messages that are removed from
813 the spool by the -Mrm option.
815 18. New variables $sender_verify_failure and $recipient_verify_failure contain
816 information about exactly what failed.
818 19. Added -dd to debug only the daemon process.
820 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
821 handles timeouts, both on the server side and network timeouts. Renamed the
822 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
824 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
825 It is now set to "smtps".
827 22. $host_address is now set to the target address during the checking of
830 23. When checking ignore_target_hosts for an ipliteral router, no host name was
831 being passed; this would have caused $sender_host_name to have been used if
832 matching the list had actually called for a host name (not very likely,
833 since this list is usually IP addresses). A host name is now passed as
836 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
837 code that specifies a non-restarting handler (typically sigaction() in
838 modern systems) in an attempt to fix a rare and obscure crash bug.
840 25. Narrowed the window for a race in the daemon that could cause it to ignore
841 SIGCHLD signals. This is not a major problem, because they are used only to
842 wake it up if nothing else does.
844 26. A malformed maildirsize file could cause Exim to calculate negative values
845 for the mailbox size or file count. Odd effects could occur as a result.
846 The maildirsize information is now recalculated if the size or filecount
849 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
850 support for a long time. Removed HAVE_SYS_VFS_H.
852 28. Installed the latest version of exipick from John Jetmore.
854 29. In an address list, if the pattern was not a regular expression, an empty
855 subject address (from a bounce message) matched only if the pattern was an
856 empty string. Non-empty patterns were not even tested. This was the wrong
857 because it is perfectly reasonable to use an empty address as part of a
858 database query. An empty address is now tested by patterns that are
859 lookups. However, all the other forms of pattern expect the subject to
860 contain a local part and a domain, and therefore, for them, an empty
861 address still always fails if the pattern is not itself empty.
863 30. Exim went into a mad DNS loop when attempting to do a callout where the
864 host was specified on an smtp transport, and looking it up yielded more
867 31. Re-factored the code for checking spool and log partition space into a
868 function that finds that data and another that does the check. The former
869 is then used to implement four new variables: $spool_space, $log_space,
870 $spool_inodes, and $log_inodes.
872 32. The RFC2047 encoding function was originally intended for short strings
873 such as real names; it was not keeping to the 75-character limit for
874 encoded words that the RFC imposes. It now respects the limit, and
875 generates multiple encoded words if necessary. To be on the safe side, I
876 have increased the buffer size for the ${rfc2047: expansion operator from
879 33. It is now permitted to omit both strings after an "if" condition; if the
880 condition is true, the result is "true". As before, when the second string
881 is omitted, a false condition yields an empty string. This makes it less
882 cumbersome to write custom ACL and router conditions.
884 34. Failure to deliver a bounce message always caused it to be frozen, even if
885 there was an errors_to setting on the router. The errors_to setting is now
888 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
889 canonical form (fully expanded) before being placed in
890 $sender_host_address.
892 36. The table in the code that translates DNS record types into text (T_A to
893 "A" for instance) was missing entries for NS and CNAME. It is just possible
894 that this could have caused confusion if both these types were looked up
895 for the same domain, because the text type is used as part of Exim's
896 per-process caching. But the chance of anyone hitting this buglet seems
899 37. The dnsdb lookup has been extended in a number of ways.
901 (1) There is a new type, "zns", which walks up the domain tree until it
902 finds some nameserver records. It should be used with care.
904 (2) There is a new type, "mxh", which is like "mx" except that it returns
905 just the host names, not the priorities.
907 (3) It is now possible to give a list of domains (or IP addresses) to be
908 looked up. The behaviour when one of the lookups defers can be
909 controlled by a keyword.
911 (4) It is now possible to specify the separator character for use when
912 multiple records are returned.
914 38. The dnslists ACL condition has been extended: it is now possible to supply
915 a list of IP addresses and/or domains to be looked up in a particular DNS
918 39. Added log_selector=+queue_time_overall.
920 40. When running the queue in the test harness, wait just a tad after forking a
921 delivery process, to get repeatability of debugging output.
923 41. Include certificate and key file names in error message when GnuTLS fails
924 to set them up, because the GnuTLS error message doesn't include the name
925 of the failing file when there is a problem reading it.
927 42. Allow both -bf and -bF in the same test run.
929 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
931 44. The "Exiscan patch" is now merged into the mainline Exim source.
933 45. Sometimes the final signoff response after QUIT could fail to get
934 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
935 < 0 before doing a fflush(). This bug looks as though it goes back to the
936 introduction of TLS in release 3.20, but "sometimes" must have been rare
937 because the tests only now provoked it.
939 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
940 (this can affect the format of dates).
942 47. exim_tidydb, when checking for the continued existence of a message for
943 which it has found a message-specific retry record, was not finding
944 messages that were in split spool directories. Consequently, it was
945 deleting retry records that should have stayed in existence.
947 48. Steve fixed some bugs in eximstats.
949 49. The SPA authentication driver was not abandoning authentication and moving
950 on to the next authenticator when an expansion was forced to fail,
951 contradicting the general specification for all authenticators. Instead it
952 was generating a temporary error. It now behaves as specified.
954 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
955 (the order specifies the preference for clients). The order is now AES256,
956 AES128, 3DES, ARCFOUR128.
958 51. Small patch to Sieve code - explicitly set From: when generating an
961 52. Exim crashed if a remote delivery caused a very long error message to be
962 recorded - for instance if somebody sent an entire SpamAssassin report back
963 as a large number of 550 error lines. This bug was coincidentally fixed by
964 increasing the size of one of Exim's internal buffers (big_buffer) that
965 happened as part of the Exiscan merge. However, to be on the safe side, I
966 have made the code more robust (and fixed the comments that describe what
969 53. Now that there can be additional text after "Completed" in log lines (if
970 the queue_time_overall log selector is set), a one-byte patch to exigrep
971 was needed to allow it to recognize "Completed" as not the last thing in
974 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
975 patch that reportedly fixes this has been added. I am not expert enough to
976 create a test for it. This is what the patch creator wrote:
978 "I found a little strange behaviour of ldap code when working with
979 Windows 2003 AD Domain, where users was placed in more than one
980 Organization Units. When I tried to give exim partial DN, the exit code
981 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
982 But simultaneously result of request was absolutely normal ldap result,
983 so I produce this patch..."
985 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
986 REFERENCE, so I have modified the code to exclude the patch when that macro
989 55. Some experimental protocols are using DNS PTR records for new purposes. The
990 keys for these records are domain names, not reversed IP addresses. The
991 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
992 leaves it alone. Component reversal etc. now happens only for IP addresses.
995 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
997 57. Double the size of the debug message buffer (to 2048) so that more of very
998 long debug lines gets shown.
1000 58. The exicyclog utility now does better if the number of log files to keep
1001 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1003 59. Two changes related to the smtp_active_hostname option:
1005 (1) $smtp_active_hostname is now available as a variable.
1006 (2) The default for smtp_banner uses $smtp_active_hostname instead
1007 of $primary_hostname.
1009 60. The host_aton() function is supposed to be passed a string that is known
1010 to be a valid IP address. However, in the case of IPv6 addresses, it was
1011 not checking this. This is a hostage to fortune. Exim now panics and dies
1012 if the condition is not met. A case was found where this could be provoked
1013 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1014 components; fortuitously, this particular loophole had already been fixed
1015 by change 4.50/55 above.
1017 If there are any other similar loopholes, the new check in host_aton()
1018 itself should stop them being exploited. The report I received stated that
1019 data on the command line could provoke the exploit when Exim was running as
1020 exim, but did not say which command line option was involved. All I could
1021 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1022 running as the user.
1025 61. There was a buffer overflow vulnerability in the SPA authentication code
1026 (which came originally from the Samba project). I have added a test to the
1027 spa_base64_to_bits() function which I hope fixes it.
1030 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1031 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1033 63. The daemon start-up calls getloadavg() while still root for those OS that
1034 need the first call to be done as root, but it missed one case: when
1035 deliver_queue_load_max is set with deliver_drop_privilege. This is
1036 necessary for the benefit of the queue runner, because there is no re-exec
1037 when deliver_drop_privilege is set.
1039 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1040 This has been fixed.
1042 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1043 was in use, was not putting the data itself into the right store pool;
1044 consequently, it could be overwritten for a subsequent message in the same
1045 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1048 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1050 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1051 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1052 the function were treating the return as a boolean value, which happened to
1053 work because 0=false and not-0=true, but is not correct code.
1055 68. The host_aton() function was not handling scoped IPv6 addresses (those
1056 with, for example, "%eth0" on the end) correctly.
1058 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1059 negated items (that is, ~something) in unsigned ints. Some compilers
1060 apparently mutter when there is no cast.
1062 70. If an address verification called from an ACL failed, and did not produce a
1063 user-specific message (i.e. there was only a "system" message), nothing was
1064 put in $acl_verify_message. In this situation, it now puts the system
1067 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1068 session; change 4.31/43 added the unwanted input to the log line - except
1069 that it did not do this in the start of session case. It now does.
1071 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1072 This is wrong and can cause the other end to generate a synchronization
1073 error if it is another Exim or anything else that does the synchronization
1074 check. A QUIT command is no longer sent after a timeout.
1076 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1077 during host lookups.
1079 74. The defer_ok option of callout verification was not working if it was used
1080 when verifying addresses in header lines, that is, for this case:
1082 verify = header_sender/callout=defer_ok
1084 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1085 those file descriptors could be used for SMTP connections. If anything
1086 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1087 could be sent to the SMTP client, causing chaos. The daemon now opens
1088 stdin, stdout, and stderr to /dev/null when it puts itself into the
1091 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1092 log by default. The user can override this with suitable Perl magic.
1094 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1095 the log message when discard triggers, was not working for the DATA ACL or
1096 for the non-SMTP ACL.
1098 78. Error message wording change in sieve.c.
1100 79. If smtp_accept_max_per_host was set, the number of connections could be
1101 restricted to fewer than expected, because the daemon was trying to set up
1102 a new connection before checking whether the processes handling previous
1103 connections had finished. The check for completed processes is now done
1104 earlier. On busy systems, this bug wouldn't be noticed because something
1105 else would have woken the daemon, and it would have reaped the completed
1108 80. If a message was submitted locally by a user whose login name contained one
1109 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1110 It caused a spool format error. I have fixed the spool reading code. A
1111 related problem was that the "from" clause in the Received: line became
1112 illegal because of the space(s). It is now covered by ${quote_local_part.
1114 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1117 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1119 83. Patch from Sieve maintainer to fix unterminated string problem in
1120 "vacation" handling.
1122 84. Some minor changes to the Linux configuration files to help with other
1123 OS variants using glibc.
1125 85. One more patch for Sieve to update vacation handling to latest spec.
1128 ----------------------------------------------------
1129 See the note above about the 4.44 and 4.50 releases.
1130 ----------------------------------------------------
1136 1. Change 4.43/35 introduced a bug that caused file counts to be
1137 incorrectly computed when quota_filecount was set in an appendfile
1140 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1141 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1144 3. Give more explanation in the error message when the command for a transport
1145 filter fails to execute.
1147 4. There are several places where Exim runs a non-Exim command in a
1148 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1149 was being done only for the command run by the queryprogram router. It is
1150 now done for all such subprocesses. The other cases are: ${run, transport
1151 filters, and the commands run by the lmtp and pipe transports.
1153 5. Some older OS have a limit of 256 on the maximum number of file
1154 descriptors. Exim was using setrlimit() to set 1000 as a large value
1155 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1156 systems. I've change it so that if it can't get 1000, it tries for 256.
1158 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1159 was an oversight, and furthermore, ever since the addition of extra
1160 controls (e.g. 4.43/32), the checks on when to allow different forms of
1161 "control" were broken. There should now be diagnostics for all cases when a
1162 control that does not make sense is encountered.
1164 7. $recipients is now available in the predata ACL (oversight).
1166 8. Tidy the search cache before the fork to do a delivery from a message
1167 received from the command line. Otherwise the child will trigger a lookup
1168 failure and thereby defer the delivery if it tries to use (for example) a
1169 cached ldap connection that the parent has called unbind on.
1171 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1172 of $address_data from the recipient verification was clobbered by the
1173 sender verification.
1175 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1176 was its contents. (It was OK if the option was not defined at all.)
1178 11. A "Completed" log line is now written for messages that are removed from
1179 the spool by the -Mrm option.
1181 12. $host_address is now set to the target address during the checking of
1182 ignore_target_hosts.
1184 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1185 being passed; this would have caused $sender_host_name to have been used if
1186 matching the list had actually called for a host name (not very likely,
1187 since this list is usually IP addresses). A host name is now passed as
1190 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1191 code that specifies a non-restarting handler (typically sigaction() in
1192 modern systems) in an attempt to fix a rare and obscure crash bug.
1194 15. Narrowed the window for a race in the daemon that could cause it to ignore
1195 SIGCHLD signals. This is not a major problem, because they are used only to
1196 wake it up if nothing else does.
1198 16. A malformed maildirsize file could cause Exim to calculate negative values
1199 for the mailbox size or file count. Odd effects could occur as a result.
1200 The maildirsize information is now recalculated if the size or filecount
1203 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1204 support for a long time. Removed HAVE_SYS_VFS_H.
1206 18. Updated exipick to current release from John Jetmore.
1208 19. Allow an empty sender to be matched against a lookup in an address list.
1209 Previously the only cases considered were a regular expression, or an
1212 20. Exim went into a mad DNS lookup loop when doing a callout where the
1213 host was specified on the transport, if the DNS lookup yielded more than
1216 21. The RFC2047 encoding function was originally intended for short strings
1217 such as real names; it was not keeping to the 75-character limit for
1218 encoded words that the RFC imposes. It now respects the limit, and
1219 generates multiple encoded words if necessary. To be on the safe side, I
1220 have increased the buffer size for the ${rfc2047: expansion operator from
1223 22. Failure to deliver a bounce message always caused it to be frozen, even if
1224 there was an errors_to setting on the router. The errors_to setting is now
1227 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1228 canonical form (fully expanded) before being placed in
1229 $sender_host_address.
1231 24. Updated eximstats to version 1.33
1233 25. Include certificate and key file names in error message when GnuTLS fails
1234 to set them up, because the GnuTLS error message doesn't include the name
1235 of the failing file when there is a problem reading it.
1237 26. Expand error message when OpenSSL has problems setting up cert/key files.
1240 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1241 (this can affect the format of dates).
1243 28. exim_tidydb, when checking for the continued existence of a message for
1244 which it has found a message-specific retry record, was not finding
1245 messages that were in split spool directories. Consequently, it was
1246 deleting retry records that should have stayed in existence.
1248 29. eximstats updated to version 1.35
1249 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1250 1.35 - bugfix such that pie charts by volume are generated correctly
1252 30. The SPA authentication driver was not abandoning authentication and moving
1253 on to the next authenticator when an expansion was forced to fail,
1254 contradicting the general specification for all authenticators. Instead it
1255 was generating a temporary error. It now behaves as specified.
1257 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1258 (the order specifies the preference for clients). The order is now AES256,
1259 AES128, 3DES, ARCFOUR128.
1261 31. Small patch to Sieve code - explicitly set From: when generating an
1264 32. Exim crashed if a remote delivery caused a very long error message to be
1265 recorded - for instance if somebody sent an entire SpamAssassin report back
1266 as a large number of 550 error lines. This bug was coincidentally fixed by
1267 increasing the size of one of Exim's internal buffers (big_buffer) that
1268 happened as part of the Exiscan merge. However, to be on the safe side, I
1269 have made the code more robust (and fixed the comments that describe what
1272 33. Some experimental protocols are using DNS PTR records for new purposes. The
1273 keys for these records are domain names, not reversed IP addresses. The
1274 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1275 leaves it alone. Component reversal etc. now happens only for IP addresses.
1278 34. The host_aton() function is supposed to be passed a string that is known
1279 to be a valid IP address. However, in the case of IPv6 addresses, it was
1280 not checking this. This is a hostage to fortune. Exim now panics and dies
1281 if the condition is not met. A case was found where this could be provoked
1282 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1283 components; fortuitously, this particular loophole had already been fixed
1284 by change 4.50/55 or 4.44/33 above.
1286 If there are any other similar loopholes, the new check in host_aton()
1287 itself should stop them being exploited. The report I received stated that
1288 data on the command line could provoke the exploit when Exim was running as
1289 exim, but did not say which command line option was involved. All I could
1290 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1291 running as the user.
1294 35. There was a buffer overflow vulnerability in the SPA authentication code
1295 (which came originally from the Samba project). I have added a test to the
1296 spa_base64_to_bits() function which I hope fixes it.
1299 36. The daemon start-up calls getloadavg() while still root for those OS that
1300 need the first call to be done as root, but it missed one case: when
1301 deliver_queue_load_max is set with deliver_drop_privilege. This is
1302 necessary for the benefit of the queue runner, because there is no re-exec
1303 when deliver_drop_privilege is set.
1305 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1306 was in use, was not putting the data itself into the right store pool;
1307 consequently, it could be overwritten for a subsequent message in the same
1308 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1311 38. Sometimes the final signoff response after QUIT could fail to get
1312 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1313 < 0 before doing a fflush(). This bug looks as though it goes back to the
1314 introduction of TLS in release 3.20, but "sometimes" must have been rare
1315 because the tests only now provoked it.
1321 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1322 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1323 Now it returns an integer. A number of calls were still expecting a T/F
1324 return. Fortuitously, in all cases, the tests worked in OK situations,
1325 which is the norm. However, things would have gone wrong on any write
1326 failures on the smtp file descriptor. This function is used when sending
1327 messages over SMTP and also when doing verify callouts.
1329 2. When Exim is called to do synchronous delivery of a locally submitted
1330 message (the -odf or -odi options), it no longer closes stderr before doing
1333 3. Implemented the mua_wrapper option.
1335 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1337 5. Implemented the functions header_remove(), header_testname(),
1338 header_add_at_position(), and receive_remove_recipient(), and exported them
1341 6. If an ACL "warn" statement specified the addition of headers, Exim already
1342 inserted X-ACL-Warn: at the start if there was no header name. However, it
1343 was not making this test for the second and subsequent header lines if
1344 there were newlines in the string. This meant that an invalid header could
1345 be inserted if Exim was badly configured.
1347 7. Allow an ACL "warn" statement to add header lines at the start or after all
1348 the Received: headers, as well as at the end.
1350 8. Added the rcpt_4xx retry error code.
1352 9. Added postmaster_mailfrom=xxx to callout verification option.
1354 10. Added mailfrom=xxxx to the callout verification option, for verify=
1357 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1358 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1360 12. Inserted some casts to stop certain compilers warning when using pointer
1361 differences as field lengths or precisions in printf-type calls (mostly
1362 affecting debugging statements).
1364 13. Added optional readline() support for -be (dynamically loaded).
1366 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1367 same clock tick as a message's arrival, so that its received time was the
1368 same as the "first fail" time on the retry record, and that message
1369 remained on the queue past the ultimate address timeout, every queue runner
1370 would try a delivery (because it was past the ultimate address timeout) but
1371 after another failure, the ultimate address timeout, which should have then
1372 bounced the address, did not kick in. This was a "< instead of <=" error;
1373 in most cases the first failure would have been in the next clock tick
1374 after the received time, and all would be well.
1376 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1377 being recognized when the domain list was tested by the match_domain
1378 condition in an expansion string.
1380 16. Added the ${str2b64: operator.
1382 17. Exim was always calling setrlimit() to set a large limit for the number of
1383 processes, without checking whether the existing limit was already
1384 adequate. (It did check for the limit on file descriptors.) Furthermore,
1385 errors from getrlimit() and setrlimit() were being ignored. Now they are
1386 logged to the main and panic logs, but Exim does carry on, to try to do its
1387 job under whatever limits there are.
1389 18. Imported PCRE 5.0.
1391 19. Trivial typo in log message " temporarily refused connection" (the leading
1394 20. If the log selector return_path_on_delivery was set and an address was
1395 redirected to /dev/null, the delivery process crashed because it assumed
1396 that a return path would always be set for a "successful" delivery. In this
1397 case, the whole delivery is bypassed as an optimization, and therefore no
1400 21. Internal re-arrangement: the function for sending a challenge and reading
1401 a response while authentication was assuming a zero-terminated challenge
1402 string. It's now changed to take a pointer and a length, to allow for
1403 binary data in such strings.
1405 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1407 23. Exim was not respecting finduser_retries when seeking the login of the
1408 uid under which it was called; it was always trying 10 times. (The default
1409 setting of finduser_retries is zero.) Also, it was sleeping after the final
1410 failure, which is pointless.
1412 24. Implemented tls_on_connect_ports.
1414 25. Implemented acl_smtp_predata.
1416 26. If the domain in control=submission is set empty, Exim assumes that the
1417 authenticated id is a complete email address when it generates From: or
1418 Sender: header lines.
1420 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1421 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1422 chown and chgrp in /bin and hostname in /usr/bin.
1424 28. Exim was keeping the "process log" file open after each use, just as it
1425 does for the main log. This opens the possibility of it remaining open for
1426 long periods when the USR1 signal hits a daemon. Occasional processlog
1427 errors were reported, that could have been caused by this. Anyway, it seems
1428 much more sensible not to leave this file open at all, so that is what now
1431 29. The long-running daemon process does not normally write to the log once it
1432 has entered its main loop, and it closes the log before doing so. This is
1433 so that log files can straightforwardly be renamed and moved. However,
1434 there are a couple of unusual error situations where the daemon does write
1435 log entries, and I had neglected to close the log afterwards.
1437 30. The text of an SMTP error response that was received during a remote
1438 delivery was being truncated at 512 bytes. This is too short for some of
1439 the long messages that one sometimes sees. I've increased the limit to
1442 31. It is now possible to make retry rules that apply only when a message has a
1443 specific sender, in particular, an empty sender.
1445 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1446 it possible to be selective about when SMTP synchronization is enforced.
1448 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1450 32. Implemented hosts_connection_nolog.
1452 33. Added an ACL for QUIT.
1454 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1457 35. Added mailbox_size and mailbox_filecount to appendfile.
1459 36. Added control = no_multiline_responses to ACLs.
1461 37. There was a bug in the logic of the code that waits for the clock to tick
1462 in the case where the clock went backwards by a substantial amount such
1463 that the microsecond fraction of "now" was more than the microsecond
1464 fraction of "then" (but the whole seconds number was less).
1466 38. Added support for the libradius Radius client library this is found on
1467 FreeBSD (previously only the radiusclient library was supported).
1473 1. When certain lookups returned multiple values in the form name=value, the
1474 quoting of the values was not always being done properly. Specifically:
1475 (a) If the value started with a double quote, but contained no whitespace,
1477 (b) If the value contained whitespace other than a space character (i.e.
1478 tabs or newlines or carriage returns) it was not quoted.
1479 This fix has been applied to the mysql and pgsql lookups by writing a
1480 separate quoting function and calling it from the lookup code. The fix
1481 should probably also be applied to nisplus, ibase and oracle lookups, but
1482 since I cannot test any of those, I have not disturbed their existing code.
1484 2. A hit in the callout cache for a specific address caused a log line with no
1485 reason for rejecting RCPT. Now it says "Previous (cached) callout
1486 verification failure".
1488 3. There was an off-by-one bug in the queryprogram router. An over-long
1489 return line was truncated at 256 instead of 255 characters, thereby
1490 overflowing its buffer with the terminating zero. As well as fixing this, I
1491 have increased the buffer size to 1024 (and made a note to document this).
1493 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1494 when Exim is waiting for an SMTP response from a remote server, Exim
1495 restarts its select() call on the socket, thereby resetting its timeout.
1496 This is not a problem when such interrupts are rare. Somebody set up a cron
1497 job to run exiwhat every 2 minutes, which is less than the normal select()
1498 timeout (5 or 10 minutes). This meant that the select() timeout never
1499 kicked in because it was always reset. I have fixed this by comparing the
1500 time when an interrupt arrives with the time at the start of the first call
1501 to select(). If more time than the timeout has elapsed, the interrupt is
1502 treated as a timeout.
1504 5. Some internal re-factoring in preparation for the addition of Sieve
1505 extensions (by MH). In particular, the "personal" test is moved to a
1506 separate function, and given an option for scanning Cc: and Bcc: (which is
1507 not set for Exim filters).
1509 6. When Exim created an email address using the login of the caller as the
1510 local part (e.g. when creating a From: or Sender: header line), it was not
1511 quoting the local part when it contained special characters such as @.
1513 7. Installed new OpenBSD configuration files.
1515 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1516 try to make them clearer.
1518 9. Callout options, other than the timeout value, were being ignored when
1519 verifying sender addresses in header lines. For example, when using
1521 verify = header_sender/callout=no_cache
1523 the cache was (incorrectly) being used.
1525 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1526 only the Cygwin environment.
1528 11. When return_path_on_delivery was set as a log selector, if different remote
1529 addresses in the same message used different return paths and parallel
1530 remote delivery occurred, the wrong values would sometimes be logged.
1531 (Whenever a remote delivery process finished, the return path value from
1532 the most recently started remote delivery process was logged.)
1534 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1535 lines when AUTH and/or TLS are in use. This is the "received protocol"
1536 field. Exim used to use "asmtp" for authenticated SMTP, without any
1537 indication (in the protocol name) for TLS use. Now it follows the RFC and
1538 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1539 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1540 names appear in log lines as well as in Received: header lines.
1542 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1543 extensions, and comparison tests, and to fix some bugs.
1545 14. Changes to the "personal" filter test:
1547 (1) The test was buggy in that it was just doing the equivalent of
1548 "contains" tests on header lines. For example, if a user's address was
1549 anne@some.where, the "personal" test would incorrectly be true for
1551 To: susanne@some.where
1553 This test is now done by extracting each address from the header in turn,
1554 and checking the entire address. Other tests that are part of "personal"
1555 are now done using regular expressions (for example, to check local parts
1556 of addresses in From: header lines).
1558 (2) The list of non-personal local parts in From: addresses has been
1559 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1560 taken from the Sieve specification recommendations.
1562 (3) If the message contains any header line starting with "List-" it is
1563 treated as non-personal.
1565 (4) The test for "circular" in the Subject: header line has been removed
1566 because it now seems ill-conceived.
1568 15. Minor typos in src/EDITME comments corrected.
1570 16. Installed latest exipick from John Jetmore.
1572 17. If headers_add on a router specified a text string that was too long for
1573 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1574 of string_sprintf() is now avoided.
1576 18. $message_body_size was not set (it was always zero) when running the DATA
1577 ACL and the local_scan() function.
1579 19. For the "mail" command in an Exim filter, no default was being set for
1580 the once_repeat time, causing a random time value to be used if "once" was
1581 specified. (If the value happened to be <= 0, no repeat happened.) The
1582 default is now 0s, meaning "never repeat". The "vacation" command was OK
1583 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1584 (I found it when inspecting the code).
1586 20. There is now an overall timeout for performing a callout verification. It
1587 defaults to 4 times the callout timeout, which applies to individual SMTP
1588 commands during the callout. The overall timeout applies when there is more
1589 than one host that can be tried. The timeout is checked before trying the
1590 next host. This prevents very long delays if there are a large number of
1591 hosts and all are timing out (e.g. when the network connections are timing
1592 out). The value of the overall timeout can be changed by specifying an
1593 additional sub-option for "callout", called "maxwait". For example:
1595 verify = sender/callout=5s,maxwait=20s
1597 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1598 to the end before writing, but this should make it even safer).
1600 22. Exim was forgetting that it had advertised PIPELINING for the second and
1601 subsequent messages on an SMTP connection. It was also not resetting its
1602 memory on STARTTLS and an internal HELO.
1604 23. When Exim logs an SMTP synchronization error within a session, it now
1605 records whether PIPELINING has been advertised or not.
1607 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1608 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1609 rather than long int.
1611 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1613 26. Added the never_mail option to autoreply.
1619 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1620 crash if the getsockname() call failed; this can happen if a connection is
1621 closed very soon after it is established. The problem was simply in the
1622 order in which certain operations were done, causing Exim to try to write
1623 to the SMTP stream before it had set up the file descriptor. The bug has
1624 been fixed by making things happen in the correct order.
1630 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1631 before the connection was closed, thus losing the rejection response.
1633 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1634 some early Solaris releases, but causes trouble in current releases where
1635 socklen_t is defined.
1637 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1640 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1643 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1644 that releasing the top of it at the end releases what was used for sub-
1645 expansions (unless the block got too big). However, discard this block if
1646 the first thing is a variable or header, so that we can use its block when
1647 it is dynamic (useful for very large $message_headers, for example).
1649 6. Lookups now cache *every* query, not just the most recent. A new, separate
1650 store pool is used for this. It can be recovered when all lookup caches are
1651 flushed. Lookups now release memory at the end of their result strings.
1652 This has involved some general refactoring of the lookup sources.
1654 7. Some code has been added to the store_xxx() functions to reduce the amount
1655 of flapping under certain conditions.
1657 8. log_incoming_interface used to affect only the <= reception log lines. Now
1658 it causes the local interface and port to be added to several more SMTP log
1659 lines, for example "SMTP connection from", and rejection lines.
1661 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1663 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1665 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1666 could be overwritten at the end of the current message (or the start of a
1667 new message if it was set in a HELO ACL). The value is now preserved for
1668 the duration of the SMTP connection.
1670 12. If a transport had a headers_rewrite setting, and a matching header line
1671 contained an unqualified address, that address was qualified, even if it
1672 did not match any rewriting rules. The underlying bug was that the values
1673 of the flags that permit the existence of unqualified sender and recipient
1674 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1675 non-local messages, and by -bnq for local messages) were not being
1676 preserved with the message after it was received.
1678 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1679 "next input=" as part of the text comprising the host identity instead of
1680 the correct text. The code was using the same buffer for two different
1681 strings. However, depending on which order the printing function evaluated
1682 its arguments, the bug did not always show up. Under Linux, for example, my
1683 test suite worked just fine.
1685 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1686 doesn't work with some older versions of Perl. It has been changed to "my",
1687 which in any case is probably the better facility to use.
1689 15. A really picky compiler found some instances of statements for creating
1690 error messages that either had too many or two few arguments for the format
1693 16. The size of the buffer for calls to the DNS resolver has been increased
1694 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1695 for addresses that have a lot of PTR records. This alleviates a problem; it
1696 does not fully solve it.
1698 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1699 buffer now truncates the list and logs the incident, which is the same
1700 action as happens when Exim is looking up a host name and its aliases.
1701 Previously in this situation something unpredictable would happen;
1702 sometimes it was "internal error: store_reset failed".
1704 18. If a server dropped the connection unexpectedly when an Exim client was
1705 using GnuTLS and trying to read a response, the client delivery process
1706 crashed while trying to generate an error log message.
1708 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1709 string, for example:
1711 warn message = H1: something\nH2: something
1713 the text was added as a single header line from Exim's point of view
1714 though it ended up OK in the delivered message. However, searching for the
1715 second and subsequent header lines using $h_h2: did not work. This has been
1716 fixed. Similarly, if a system filter added multiple headers in this way,
1717 the routers could not see them.
1719 20. Expanded the error message when iplsearch is called with an invalid key to
1720 suggest using net-iplsearch in a host list.
1722 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1723 ACLs are no longer actually imposed (and a message to that effect is
1726 22. If a "gecos" field in a passwd entry contained escaped characters, in
1727 particular, if it contained a \" sequence, Exim got it wrong when building
1728 a From: or a Sender: header from that name. A second bug also caused
1729 incorrect handling when an unquoted " was present following a character
1730 that needed quoting.
1732 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1733 was not being matched caselessly.
1735 24. Arranged for all hyphens in the exim.8 source to be escaped with
1738 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1739 properly. Recipient callout cache records were still being keyed to include
1740 the sender, even when use_sender was set false. This led to far more
1741 callouts that were necessary. The sender is no longer included in the key
1742 when use_sender is false.
1744 26. Added "control = submission" modifier to ACLs.
1746 27. Added the ${base62d: operator to decode base 62 numbers.
1748 28. dnsdb lookups can now access SRV records.
1750 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1751 the configuration file.
1753 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1754 (-v) mode. This makes the output for a verbose queue run more intelligible.
1756 31. Added a use_postmaster feature to recipient callouts.
1758 32. Added the $body_zerocount variable, containing the number of binary zero
1759 bytes in the message body.
1761 33. The time of last modification of the "new" subdirectory is now used as the
1762 "mailbox time last read" when there is a quota error for a maildir
1765 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1767 35. Added +ignore_unknown as a special item in host lists.
1769 36. Code for decoding IPv6 addresses in host lists is now included, even if
1770 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1771 address was recognized as an IP address, but was then not correctly decoded
1772 into binary, causing unexpected and incorrect effects when compared with
1779 1. Very minor rewording of debugging text in manualroute to say "list of
1780 hosts" instead of "hostlist".
1782 2. If verify=header_syntax was set, and a header line with an unqualified
1783 address (no domain) and a large number of spaces between the end of the
1784 name and the colon was received, the reception process suffered a buffer
1785 overflow, and (when I tested it) crashed. This was caused by some obsolete
1786 code that should have been removed. The fix is to remove it!
1788 3. When running in the test harness, delay a bit after writing a bounce
1789 message to get a bit more predictability in the log output.
1791 4. Added a call to search_tidyup() just before forking a reception process. In
1792 theory, someone could use a lookup in the expansion of smtp_accept_max_
1793 per_host which, without the tidyup, could leave open a database connection.
1795 5. Added the variables $recipient_data and $sender_data which get set from a
1796 lookup success in an ACL "recipients" or "senders" condition, or a router
1797 "senders" option, similar to $domain_data and $local_part_data.
1799 6. Moved the writing of debug_print from before to after the "senders" test
1802 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1803 problems for message scanning, either using a data ACL, or using
1804 local_scan() because the Received: header was not generated till after they
1805 were called (in order to set the time as the time of reception completion).
1806 I have revised the way this works. The header is now generated after the
1807 body is received, but before the ACL or local_scan() are called. After they
1808 are run, the timestamp in the header is updated.
1814 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1815 before starting a queue runner without re-exec. This happened only when
1816 deliver_drop_privilege was set or when the Exim user was set to root. The
1817 effect of the bug was that timeouts during subsequent deliveries caused
1818 crashes instead of being properly handled. The handler is now left at its
1819 default (and expected) setting.
1821 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1822 message, again when deliver_drop_privilege is set or Exim is run as root.
1823 The bug described in (1) was not present in this case, but the tidying up
1824 of the other signals was missing. I have made the two cases consistent.
1826 3. The ignore_target_hosts setting on a manualroute router was being ignored
1827 for hosts that were looked up using the /MX notation.
1829 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1832 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1833 operated on the sender address. After changing the $sender_address to <>
1834 for the sender address verify, Exim was re-instated it as the original
1835 (before rewriting) address, but remembering that it had rewritten it, so it
1836 wasn't rewriting it again. This bug also had the effect of breaking the
1837 sender address verification caching when the sender address was rewritten.
1839 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1840 This has been changed so that if the ip literal address matches
1841 ignore_target_hosts, the router declines.
1843 7. Added expansion conditions match_domain, match_address, and match_local_
1844 part (NOT match_host).
1846 8. The placeholder for the Received: header didn't have a length field set.
1848 9. Added code to Exim itself and to exim_lock to test for a specific race
1849 condition that could lead to file corruption when using MBX delivery. The
1850 issue is with the lockfile that is created in /tmp. If this file is removed
1851 after a process has opened it but before that process has acquired a lock,
1852 there is the potential for a second process to recreate the file and also
1853 acquire a lock. This could lead to two Exim processes writing to the file
1854 at the same time. The added code performs the same test as UW imapd; it
1855 checks after acquiring the lock that its file descriptor still refers to
1856 the same named file.
1858 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1859 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1862 11. Added the smtp_active_hostname option. If used, this will typically be made
1863 to depend on the incoming interface address. Because $interface_address is
1864 not set up until the daemon has forked a reception process, error responses
1865 that can happen earlier (such as "too many connections") no longer contain
1868 12. If an expansion in a condition on a "warn" statement fails because a lookup
1869 defers, the "warn" statement is abandoned, and the next ACL statement is
1870 processed. Previously this caused the whole ACL to be aborted.
1872 13. Added the iplsearch lookup type.
1874 14. Added ident_timeout as a log selector.
1876 15. Added tls_certificate_verified as a log selector.
1878 16. Added a global option tls_require_ciphers (compare the smtp transport
1879 option of the same name). This controls incoming TLS connections.
1881 17. I finally figured out how to make tls_require_ciphers do a similar thing
1882 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1883 before starting the TLS session.
1885 18. Tabs are now shown as \t in -bP output.
1887 19. If the log selector return_path_on_delivery was set, Exim crashed when
1888 bouncing a message because it had too many Received: header lines.
1890 20. If two routers both had headers_remove settings, and the first one included
1891 a superfluous trailing colon, the final name in the first list and the
1892 first name in the second list were incorrectly joined into one item (with a
1893 colon in the middle).
1899 1. Added -C and -D options to the exinext utility, mainly to make it easier
1900 to include in the automated testing, but these could be helpful when
1901 multiple configurations are in use.
1903 2. The exinext utility was not formatting the output nicely when there was
1904 an alternate port involved in the retry record key, nor when there was a
1905 message id as well (for retries that were specific to a specific message
1906 and a specific host). It was also confused by IPv6 addresses, because of
1907 the additional colons they contain. I have fixed the IPv4 problem, and
1908 patched it up to do a reasonable job for IPv6.
1910 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1911 delivery, the log line now contains "pipelined" if PIPELINING was used.
1913 4. An SMTP transport process used to panic and die if the bind() call to set
1914 an explicit outgoing interface failed. This has been changed; it is now
1915 treated in the same way as a connect() failure.
1917 5. A reference to $sender_host_name in the part of a conditional expansion
1918 that was being skipped was still causing a DNS lookup. This no longer
1921 6. The def: expansion condition was not recognizing references to header lines
1922 that used bh_ and bheader_.
1924 7. Added the _cache feature to named lists.
1926 8. The code for checking quota_filecount in the appendfile transport was
1927 allowing one more file than it should have been.
1929 9. For compatibility with Sendmail, the command line option
1937 and sets the incoming protocol and host name (for trusted callers). The
1938 host name and its colon can be omitted when only the protocol is to be set.
1939 Note the Exim already has two private options, -pd and -ps, that refer to
1940 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1941 "s", but I don't think that's a major issue.
1943 10. A number of refactoring changes to the code, none of which should affect
1946 (a) The number of logging options was getting close to filling up the
1947 32-bit word that was used as a bit map. I have split them into two classes:
1948 those that are passed in the argument to log_write(), and those that are
1949 only ever tested independently outside of that function. These are now in
1950 separate 32-bit words, so there is plenty of room for expansion again.
1951 There is no change in the user interface or the logging behaviour.
1953 (b) When building, for example, log lines, the code previously used a
1954 macro that called string_cat() twice, in order to add two strings. This is
1955 not really sufficiently general. Furthermore, there was one instance where
1956 it was actually wrong because one of the argument was used twice, and in
1957 one call a function was used. (As it happened, calling the function twice
1958 did not affect the overall behaviour.) The macro has been replaced by a
1959 function that can join an arbitrary number of extra strings onto a growing
1962 (c) The code for expansion conditions now uses a table and a binary chop
1963 instead of a serial search (which was left over from when there were very
1964 few conditions). Also, it now recognizes conditions like "pam" even when
1965 the relevant support is not compiled in: a suitably worded error message is
1966 given if an attempt is made to use such a condition.
1968 11. Added ${time_interval:xxxxx}.
1970 12. A bug was causing one of the ddress fields not to be passed back correctly
1971 from remote delivery subprocesses. The field in question was not being
1972 subsequently used, so this caused to problems in practice.
1974 13. Added new log selectors queue_time and deliver_time.
1976 14. Might have fixed a bug in maildirsizefile handling that threw up
1977 "unexpected character" debug warnings, and recalculated the data
1978 unnecessarily. In any case, I expanded the warning message to give more
1981 15. Added the message "Restricted characters in address" to the statements in
1982 the default ACL that block characters like @ and % in local parts.
1984 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1985 Three changes have been made:
1987 (a) There was a serious bug; a negative response to MAIL caused the whole
1988 recipient domain to be cached as invalid, thereby blocking all messages
1989 to all local parts at the same domain, from all senders. This bug has
1990 been fixed. The domain is no longer cached after a negative response to
1991 MAIL if the sender used is not empty.
1993 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1996 (c) A new callout option, "use_sender" has been added for people who want
1997 the modified behaviour.
2003 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2006 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2007 indeed breaks things for older releases.
2009 3. Added additional logging to the case where there is a problem reading data
2010 from a filter that is running in a subprocess using a pipe, in order to
2011 try to track down a specific problem.
2013 4. Testing facility fudge: when running in the test harness and attempting
2014 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2015 getting "No route to host". Convert this to a timeout.
2017 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2020 6. Some OS don't have socklen_t but use size_t instead. This affects the
2021 fifth argument of getsockopt() amongst other things. This is now
2022 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2023 can be set for individual OS. I have set it for SunOS5, OSF1, and
2024 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2025 some earlier ones do not.
2027 7. Change 4.30/15 was not doing the test caselessly.
2029 8. The standard form for an IPv6 address literal was being rejected by address
2030 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2031 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2032 this, as well as the form without the "IPv6" on the front (but only when
2033 address literals are enabled, of course).
2035 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2037 10. Exim crashed if a message with an empty sender address specified by -f
2038 encountered a router with an errors_to setting. This could be provoked only
2039 by a command such as
2043 where an empty string was supplied; "<>" did not hit this bug.
2045 11. Installed PCRE release 4.5.
2047 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2048 remained set. It is now erased.
2050 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2051 times from message ids (which are base 36 rather than the normal 62).
2053 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2054 were being counted as actual protocol errors, and logged if the log
2055 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2056 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2057 and DATA following a set of rejected RCPTs do not count as protocol errors.
2058 In other words, Exim assumes they were pipelined, though this may not
2059 actually be the case. Of course, in all cases the client gets an
2060 appropriate error code.
2062 15. If a lookup fails in an ACL condition, a message about the failure may
2063 be available; it is used if testing the ACL cannot continue, because most
2064 such messages specify what the cause of the deferral is. However, some
2065 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2066 that caused an old message to be retained and used if a later statement
2067 caused a defer, replacing the real cause of the deferral.
2069 16. If an IP address had so many PTR records that the DNS lookup buffer
2070 was not large enough to hold them, Exim could crash while trying to process
2071 the truncated data. It now detects and logs this case.
2073 17. Further to 4.21/58, another change has been made: if (and only if) the
2074 first line of a message (the first header line) ends with CRLF, a bare LF
2075 in a subsequent header line has a space inserted after it, so as not to
2076 terminate the header.
2078 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2079 unnecessarily, used atoi() instead of strtol(), and didn't check the
2080 termination when getting file sizes from file names by regex.
2082 19. Completely re-implemented the support for maildirsize files, in the light
2083 of a number of problems with the previous contributed implementation
2084 (4.30/29). In particular:
2086 . If the quota is zero, the maildirsize file is maintained, but no quota is
2089 . If the maildir directory does not exist, it is created before any attempt
2090 to write a maildirsize file.
2092 . The quota value in the file is just a cache; if the quota is changed in
2093 the transport, the new value overrides.
2095 . A regular expression is available for excluding directories from the
2098 20. The autoreply transport checks the characters in options that define the
2099 message's headers; it allows continued headers, but it was checking with
2100 isspace() after an embedded newline instead of explicitly looking for a
2103 21. If all the "regular" hosts to which an address was routed had passed their
2104 expiry times, and had not reached their retry times, the address was
2105 bounced, even if fallback hosts were defined. Now Exim should go on to try
2108 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2109 equivalent code in the SMTP transport. Some hosts send humungous responses
2110 to HELO/EHLO, more than 1024 it seems.
2112 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2113 gives compiler warnings in some environments. I've now done it "properly",
2116 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2117 (because of problems with the built-in one) was declared to return uschar *
2118 instead of char *, causing compiler failure.
2120 25. Fixed a file descriptor leak when processing alias/forward files.
2122 26. Fixed a minor format string issue in dbfn.c.
2124 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2126 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2127 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2130 29. When a sender address is verified, it is cached, to save repeating the test
2131 when there is more than one recipient in a message. However, when the
2132 verification involves a callout, it is possible for different callout
2133 options to be set for different recipients. It is too complicated to keep
2134 track of this in the cache, so now Exim always runs a verification when a
2135 callout is required, relying on the callout cache for the optimization.
2136 The overhead is duplication of the address routing, but this should not be
2139 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2140 to be verified with callout=postmaster, and the main callout worked but the
2141 postmaster check failed, the verification correctly failed. However, if a
2142 subsequent RCPT command asked for sender verification *without* the
2143 postmaster check, incorrect caching caused this verification also to fail,
2146 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2147 it was not caching the DNS options (qualify_single, search_parents) that
2148 were used when the lookup failed. A subsequent lookup with different
2149 options therefore always gave the same answer, though there were cases
2150 where it should not have. (Example: a "domains = !$mx_any" option on a
2151 dnslookup router: the "domains" option is always processed without any
2152 widening, but the router might have qualify_single set.) Now Exim uses the
2153 cached value only when the same options are set.
2155 32. Added John Jetmore's "exipick" utility to the distribution.
2157 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2158 than a timeout (e.g. a certificate is required, and is not provided), an
2159 Exim server now closes the connection immediately. Previously it waited for
2160 the client to close - but if the client is SSL, it seems that they each
2161 wait for each other, leading to a delay before one of them times out.
2163 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2164 maintained 0.8.x compatibility because I don't think many are using it, and
2165 it is clearly obsolete.
2167 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2170 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2171 client certificate was expired. A simple patch fixes this, though I don't
2172 understand the full logic of why the verify callback is called multiple
2175 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2176 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2177 which causes problems with some clients (such as the Certicom SSL Plus
2178 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2179 disables the coutermeasure allowing Eudora to connect."
2181 38. Exim was not checking that a write() to a log file succeeded. This could
2182 lead to Bad Things if a log got too big, in particular if it hit a file
2183 size limit. Exim now panics and dies if it cannot write to a log file, just
2184 as it does if it cannot open a log file.
2186 39. Modified OS/Makefile-Linux so that it now contains
2188 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2190 The two -D definitions ensure that Exim is compiled with large file
2191 support, which makes it possible to handle log files that are bigger than
2194 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2195 instance) a domain was checked against a named list that involved a lookup,
2196 causing $domain_data to be set, then another domain was checked against the
2197 same list, then the first domain was re-checked, the value of $domain_data
2198 after the final check could be wrong. In particular, if the second check
2199 failed, it could be set empty. This bug probably also applied to
2202 41. The strip_trailing_dot option was not being applied to the address given
2203 with the -f command-line option.
2205 42. The code for reading a message's header from the spool was incrementing
2206 $received_count, but never initializing it. This meant that the value was
2207 incorrect (doubled) while delivering a message in the same process in which
2208 it was received. In the most common configuration of Exim, this never
2209 happens - a fresh exec is done - but it can happen when
2210 deliver_drop_privilege is set.
2212 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2213 it now includes up to 150 characters of the unexpected data in the log
2216 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2217 and building data strings. The size of both of these buffers was 10 000
2218 bytes - far larger than anybody would *ever* want, thought I. Needless to
2219 say, somebody hit the limit. I have increased the maximum line length to
2220 20 000 and the maximum data length of concatenated lines to 100 000. I have
2221 also fixed two bugs, because there was no checking on these buffers. Tsk,
2222 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2223 buffer is too small.
2225 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2226 lsearch lookups. Now it does.
2228 46. When parsing a route_list item in a manualroute router, a fixed-length
2229 buffer was used for the list of hosts. I made this 1024 bytes long,
2230 thinking that nobody would ever have a list of hosts that long. Wrong.
2231 Somebody had a whole pile of complicated expansion conditions, and the
2232 string was silently truncated, leading to an expansion error. It turns out
2233 that it is easier to change to an unlimited length (owing to other changes
2234 that have happened since this code was originally written) than to build
2235 structure for giving a limitation error. The length of the item that
2236 expands into the list of hosts is now unlimited.
2238 47. The lsearch lookup could not handle data where the length of text line was
2239 more than 4095 characters. Such lines were truncated, leading to shortened
2240 data being returned. It should now handle lines of any length.
2242 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2243 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2246 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2247 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2248 no longer happens in comment lines. A list of replacements is now placed
2249 at the head of all of the source files, except those whose only change is
2250 to replace PERL_COMMAND in the very first #! line.
2252 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2253 messages on the queue, with a bottom-up merge sort, using code contributed
2254 by Michael Haardt. This should make operations like -bp somewhat faster on
2255 large queues. It won't affect queue runners, except when queue_run_in_order
2258 51. Installed eximstats 1.31 in the distribution.
2260 52. Added support for SRV lookups to the dnslookup router.
2262 53. If an ACL referred to $message_body or $message_body_end, the value was not
2263 reset for any messages that followed in the same SMTP session.
2265 54. The store-handling optimization for building very long strings was not
2266 differentiating between the different store pools. I don't think this
2267 actually made any difference in practice, but I've tidied it.
2269 55. While running the routers to verify a sender address, $sender_address
2270 was still set to the sender address. This is wrong, because when routing to
2271 send a bounce to the sender, it would be empty. Therefore, I have changed
2272 it so that, while verifying a sender address, $sender_address is set to <>.
2273 (There is no change to what happens when verifying a recipient address.)
2275 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2276 target A or AAAA records (if not already returned) without resetting the
2277 qualify_single or search_parents options of the DNS resolver. These are
2278 inappropriate in this case because the targets of MX and SRV records must
2279 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2280 target that, when qualified, matched something in the local domain. These
2281 two options are now turned off when doing these lookups.
2283 57. It seems that at least some releases of Reiserfs (which does not have the
2284 concept of a fixed number of inodes) returns zero and not -1 for the
2285 number of available inodes. This interacted badly with check_spool_inodes,
2286 which assumed that -1 was the "no such thing" setting. What I have done is
2287 to check that the total number of inodes is greater than zero before doing
2288 the test of how many are available.
2290 58. When a "warn" ACL statement has a log_message modifier, the message is
2291 remembered, and not repeated. This is to avoid a lot of repetition when a
2292 message has many recipients that cause the same warning to be written.
2293 Howewer, Exim was preserving the list of already written lines for an
2294 entire SMTP session, which doesn't seem right. The memory is now reset if a
2295 new message is started.
2297 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2298 debugging output unless log_rewrite was also set.
2300 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2301 of (int)(handle) when we know that handle contains (void *)(-1).
2303 61. The Exim daemon panic-logs an error return when it closes the incoming
2304 connection. However "connection reset by peer" seems to be common, and
2305 isn't really an error worthy of noting specially, so that particular error
2308 62. When Exim is trying to find all the local interfaces, it used to panic and
2309 die if the ioctl to get the interface flags failed. However, it seems that
2310 on at least one OS (Solaris 9) it is possible to have an interface that is
2311 included in the list of interfaces, but for which you get a failure error
2312 for this call. This happens when the interface is not "plumbed" into a
2313 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2314 failure of the "get flags" call assumes that the interface is down.
2316 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2317 makes life easier for people who are doing arithmetic on fields extracted
2318 from dates, where you often get leading zeros that should not be
2319 interpreted as octal.
2321 64. Added qualify_domain to the redirect router, to override the global
2324 65. If a pathologically long header line contained very many addresses (the
2325 report of this problem mentioned 10 000) and each of them was rewritten,
2326 Exim could use up a very large amount of memory. (It kept on making new
2327 copies of the header line as it rewrote, and never released the old ones.)
2328 At the expense of a bit more processing, the header rewriting function has
2329 been changed so that it no longer eats memory in this way.
2331 66. The generation of the Received: header has been moved from the time that a
2332 message starts to be received, to the time that it finishes. The timestamp
2333 in the Received: header should now be very close to that of the <= log
2334 line. There are two side-effects of this change:
2336 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2337 logged header lines no longer include the local Received: line, because
2338 it has not yet been created. The same applies to a copy of the message
2339 that is returned to a non-SMTP sender when a message is rejected.
2341 (b) When a filter file is tested using -bf, no additional Received: header
2342 is added to the test message. After some thought, I decided that this
2345 This change does not affect the value of $received_for. It is still set
2346 after address rewriting, but before local_scan() is called.
2348 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2350 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2351 gave an unhelpful panic error message, and a defer error. I have managed to
2352 change this behaviour so that it now rejects any supplied certificate,
2353 which seems right, as the list of acceptable certificates is empty.
2355 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2356 gave an unhelpful defer error. I have not managed to make this reject any
2357 supplied certificates, but the error message it gives is "no certificate
2358 supplied", which is not helpful.
2360 70. exigrep's output now also includes lines that are not associated with any
2361 message, but which match the given pattern. Implemented by a patch from
2362 Martin Sluka, which also tidied up the Perl a bit.
2364 71. Recipient callout verification, like sender verification, was using <> in
2365 the MAIL FROM command. This isn't really the right thing, since the actual
2366 sender may affect whether the remote host accepts the recipient or not. I
2367 have changed it to use the actual sender in the callout; this means that
2368 the cache record is now keyed on a recipient/sender pair, not just the
2369 recipient address. There doesn't seem to be a real danger of callout loops,
2370 since a callout by the remote host to check the sender would use <>.
2371 [SEE ABOVE: changed after hitting problems.]
2373 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2374 temporary errors. However, in the case of such a code being given after
2375 the end of a data transmission (i.e. after ".") Exim was failing to write
2376 a retry record for the message. (Yes, there was some broken host that was
2377 actually sending 8xx at this point.)
2379 73. An unknown lookup type in a host list could cause Exim to panic-die when
2380 the list was checked. (An example that provoked this was putting <; in the
2381 middle of a list instead of at the start.) If this happened during a DATA
2382 ACL check, a -D file could be left lying around. This kind of configuration
2383 error no longer causes Exim to die; instead it causes a defer errror. The
2384 incident is still logged to the main and panic logs.
2386 74. Buglet left over from Exim 3 conversion. The message "too many messages
2387 in one connection" was written to the rejectlog but not the mainlog, except
2388 when address rewriting (yes!) was being logged.
2390 75. Added write_rejectlog option.
2392 76. When a system filter was run not as root (that is, when system_filter_user
2393 was set), the values of the $n variables were not being returned to the
2394 main process; thus, they were not subsequently available in the $sn
2397 77. Added +return_path_on_delivery log selector.
2399 78. A connection timeout was being treated differently from recipients deferred
2400 when testing hosts_max_try with a message that was older than the host's
2401 retry timeout. (The host should not be counted, thus allowing all hosts to
2402 be tried at least once before bouncing.) This may have been the cause of an
2403 occasionally reported bug whereby a message would remain on the queue
2404 longer than the retry timeout, but would be bounced if a delivery was
2405 forced. I say "may" because I never totally pinned down the problem;
2406 setting up timeout/retry tests is difficult. See also the next item.
2408 79. The ultimate address timeout was not being applied to errors that involved
2409 a combination of host plus message (for example, a timeout on a MAIL
2410 command). When an address resolved to a number of possible hosts, and they
2411 were not all tried for each delivery (e.g. because of hosts_max_try), a
2412 message could remain on the queue longer than the retry timeout.
2414 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2417 81. Fixed an obscure SMTP outgoing bug which required at least the following
2418 conditions: (a) there was another message waiting for the same server;
2419 (b) the server returned 5xx to all RCPT commands in the first message so
2420 that the message was not completed; (c) the server dropped the connection
2421 or gave a negative response to the RSET that Exim sends to abort the
2422 transaction. The observed case was a dropped connection after DATA that had
2423 been sent in pipelining mode. That is, the server had advertised PIPELINING
2424 but was not implementing it correctly. The effect of the bug was incorrect
2425 behaviour, such as trying another host, and this could lead to a crash.
2431 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2432 and daemon.c were passed as pointers to ints; they should have been
2433 pointers to socklen_t variables (which are typically unsigned ints).
2435 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2438 3. Fixed a really odd bug that affected only the testing scheme; patching a
2439 certain fixed string in the binary changed the value of another string that
2440 happened to be identical to the end of the original first string.
2442 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2443 name", it returns that address as the IP address. On some operating
2444 systems (e.g. Solaris), it also passes back the IP address string as the
2445 "host name". However, on others (e.g. Linux), it passes back an empty
2446 string. Exim wasn't checking for this, and was changing the host name to an
2447 empty string, assuming it had been canonicized.
2449 5. Although rare, it is permitted to have more than one PTR record for a given
2450 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2451 all the names associated with an address, because they do in Solaris.
2452 However, it seems that they do not in Linux for data that comes from the
2453 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2454 I found this out when I moved to a new Linux workstation and tried to run
2455 the Exim test suite.
2457 To get round this problem I have changed the code so that it now does its
2458 own call to the DNS to look up PTR records when searching for a host name.
2459 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2460 addresses that are only in /etc/hosts are still found.
2462 This behaviour is, however, controlled by an option called host_lookup_
2463 order, which defaults to "bydns:byaddr". If people want to use the other
2464 order, or indeed, just use one or the other means of lookup, they can
2465 specify it in this variable.
2467 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2468 some operating systems, this comes back from gethostbyaddr() as an empty
2469 string, and this is what Exim used to test for. However, it seems that in
2470 other systems, "." is yielded. Exim now tests for this case too.
2472 7. The values of check_spool_space and check_log_space are now held internally
2473 as a number of kilobytes instead of an absolute number of bytes. If a
2474 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2475 kilobyte. This means that much larger values can be stored.
2477 8. Exim monitor: an attempt to get the action menu when not actually pointing
2478 at a message produces an empty menu entitled "No message selected". This
2479 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2480 no entries in it ("Shell widget menu has zero width and/or height"). So I
2481 have added a single, blank menu entry in this case.
2483 9. Added ${quote_local_part.
2485 10. MIME decoding is now applied to the contents of Subject: header lines when
2488 11. Now that a reference to $sender_host_address automatically causes a reverse
2489 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2490 host lookup before query-style lookups in lists that might use this
2491 variable. This has therefore been abolished, and the "net-" prefix is no
2492 longer necessary for query-style lookups.
2494 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2495 have been a typo for LFLAGS, so it has been changed.
2497 13. The install script calls Exim with "-C /dev/null" in order to find the
2498 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2499 to be output. Howeve, since Exim outputs its version number before the
2500 error, it didn't break the script. It just looked ugly. I fixed this by
2501 always allowing "-C /dev/null" if the caller is root.
2503 14. Ignore overlarge ACL variable number when reading spool file - insurance
2504 against a later release with more variables having written the file.
2506 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2507 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2508 this, as well as the form without the "IPv6" on the front.
2510 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2511 OS/Makefile-Darwin file.
2513 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2514 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2516 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2517 variable. However, some OS use ints for inodes. I've added cast to long int
2518 to get rid of the compiler warning.
2520 19. I had forgotten to lock out "/../" in configuration file names when
2521 ALT_CONFIG_PREFIX was set.
2523 20. Routers used for verification do not need to specify transports. However,
2524 if such a router generated a host list, and callout was configured, Exim
2525 crashed, because it could not find a port number from the (non-existent)
2526 transport. It now assumes port 25 in this circumstance.
2528 21. Added the -t option to exigrep.
2530 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2531 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2532 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2533 them is set, LOOKUP_LSEARCH is forced.
2535 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2537 24. Added sender and host information to the "rejected by local_scan()" log
2538 line; previously there was no indication of these.
2540 25. Added .include_if_exists.
2542 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2543 Linux. It turns out that not all file systems support this. Apparently some
2544 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2545 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2546 not supported on the file descriptor, is now ignored when Exim is trying to
2547 sync a directory. This applies only to Linux.
2549 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2551 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2552 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2555 29. Added support for maildirsize files from supplied patch (modified a bit).
2557 30. The use of :fail: followed by an empty string could lead Exim to respond to
2558 sender verification failures with (e.g.):
2560 550 Verification failed for <xxx>
2561 550 Sender verify failed
2563 where the first response line was missing the '-' that indicates it is not
2564 the final line of the response.
2566 31. The loop for finding the name of the user that called Exim had a hardwired
2567 limit of 10; it now uses the value of finduser_retries, which is used for
2568 all other user lookups.
2570 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2573 33. Exim was neglecting to zero errno before one call of strtol() when
2574 expanding a string and expecting an integer value. On some systems this
2575 resulted in spurious "integer overflow" errors. Also, it was casting the
2576 result into an int without checking.
2578 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2579 did not work. The code looks as if it has *never* worked, though it appears
2580 to have been documented since at least releast 1.62. I have made it work.
2582 35. The "timeout_DNS" error in retry rules, also documented since at least
2583 1.62, also never worked. As it isn't clear exactly what this means, and
2584 clearly it isn't a major issue, I have abolished the feature by treating it
2585 as "timeout", and writing a warning to the main and panic logs.
2587 36. The display of retry rules for -brt wasn't always showing the error code
2590 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2591 timeout_connect_A, timeout_connect_MX.
2593 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2594 to the empty sender.
2596 39. The daemon was not analysing the content of -oX till after it had closed
2597 stderr and disconnected from the controlling terminal. This meant that any
2598 syntax errors were only noted on the panic log, and the return code from
2599 the command was 0. By re-arranging the code a little, I've made the
2600 decoding happen first, so such errors now appear on stderr, and the return
2601 code is 1. However, the actual setting up of the sockets still happens in
2602 the disconnected process, so errors there are still only recorded on the
2605 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2606 connections (as happens on some IP stacks) was logged at start up time as
2607 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2608 it from "IPv6 and IPv4", which means that two separate sockets are being
2611 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2612 says whether AF_INET or AF_INET6 was passed as an argument.
2614 42. Exiwhat output was messed up when time zones were included in log
2617 43. Exiwhat now gives more information about the daemon's listening ports,
2618 and whether -tls-on-connect was used.
2620 44. The "port" option of the smtp transport is now expanded.
2622 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2623 silently ignored. Now an error message is written to the main and panic
2626 46. There's a new ACL modifier called "logwrite" which writes to a log file
2627 as soon as it is encountered.
2629 47. Added $local_user_uid and $local_user_gid at routing time.
2631 48. Exim crashed when trying to verify a sender address that was being
2634 49. Exim was recognizing only a space character after ".include". It now also
2635 recognizes a tab character.
2637 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2638 extracting the relevant information from the specification. The man page no
2639 longer contains scrambled data for the -d option, and I've added a section
2640 at the front about calling Exim under different names.
2642 51. Added "extra_headers" argument to the "mail" command in filter files.
2644 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2647 53. Installed eximstats 1.29.
2649 54. Added transport_filter_timeout as a generic transport option.
2651 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2652 Cc: header lines. This was required by RFC 822, but it not required by RFC
2655 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2656 incoming messages that did not have them. Now it does so only if the
2657 message originates locally, that is, if there is no associated remote host
2658 address. When Resent- header lines are present, this applies to the Resent-
2659 lines rather than the non-Resent- lines.
2661 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2662 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2664 58. Messages for configuration errors now include the name of the main
2665 configuration file - useful now that there may be more than one file in a
2666 list (.included file names were always shown).
2668 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2669 for those rare installations that do not start the daemon as root or run it
2670 setuid root. I've cut out the call to initgroups() if the daemon is not
2673 60. The Exim user and group can now be bound into the binary as text strings
2674 that are looked up at the start of Exim's processing.
2676 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2678 62. Added $mailstore_basename variable.
2680 63. Installed patch to sieve.c from Michael Haardt.
2682 64. When Exim failed to open the panic log after failing to open the main log,
2683 the original message it was trying to log was written to stderr and debug
2684 output, but if they were not available (the usual case in production), it
2685 was lost. Now it is written to syslog before the two lines that record the
2686 failures to open the logs.
2688 65. Users' Exim filters run in subprocesses under the user's uid. It is
2689 possible for a "deliver" command or an alias in a "personal" command to
2690 provoke an address rewrite. If logging of address rewriting is configured,
2691 this fails because the process is not running as root or exim. There may be
2692 a better way of dealing with this, but for the moment (because 4.30 needs
2693 to be released), I have disabled address rewrite logging when running a
2694 filter in a non-root, non-exim process.
2700 1. The buildconfig auxiliary program wasn't quoting the value set for
2701 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2702 not defined. This bug was masked in 4.22 by the effect that was fixed in
2705 2. Some messages that were rejected after a message id was allocated were
2706 shown as "incomplete" by exigrep. It no longer does this for messages that
2707 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2709 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2710 have allow_domain_literals set, the ID did not get logged in the <= line.
2711 Domain literals are now always recognized in Message-ID: header lines.
2713 4. The first argument for a ${extract expansion item is the key name or field
2714 number. Leading and trailing spaces in this item were not being ignored,
2715 causing some misleading effects.
2717 5. When deliver_drop_privilege was set, single queue runner processes started
2718 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2719 same command in the process it spins off) were not dropping privilege.
2721 6. When the daemon running as "exim" started a queue runner, it always
2722 re-executed Exim in the spun-off process. This is a waste of effort when
2723 deliver_drop_privilege is set. The new process now just calls the
2724 queue-runner function directly.
2730 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2733 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2734 ignored. Though the use of -oP was forcing the writing of a pid file, it
2735 was always written to the default place.
2737 3. If the message "no IP address found for host xxxx" is generated during
2738 incoming verification, it is now followed by identification of the incoming
2739 connection (so you can more easily find what provoked it).
2741 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2743 5. Added some features to "harden" Exim a bit more against certain attacks:
2745 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2746 be put in Local/Makefile. This is like the never_users runtime option,
2747 but it cannot be overridden. The default setting is "root".
2749 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2750 prefix string with which any file named in a -C command line option
2753 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2754 is retained for -C and -D only if the caller of Exim is root. Without
2755 it, the exim user may also use -C and -D and retain privilege.
2757 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2758 command line option is disabled.
2760 6. Macro names set by the -D option must start with an upper case letter, just
2761 like macro names defined in the configuration file.
2763 7. Added "dereference=" facility to LDAP.
2765 8. Two instances of the typo "uknown" in the source files are fixed.
2767 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2768 the Configure-Makefile script screwed up while processing it.
2770 10. Incorporated PCRE 4.4.
2772 11. The SMTP synchronization check was not operating right at the start of an
2773 SMTP session. For example, it could not catch a HELO sent before the client
2774 waited for the greeting. There is now a check for outstanding input at the
2775 point when the greeting is written. Because of the duplex, asynchronous
2776 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2777 way, but not yet received, when the check is performed.
2779 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2780 on TCP/IP sockets, because this apparently causes some broken clients to
2783 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2784 unchanged) from the Cygwin maintainer.
2786 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2787 was supported instead of testing for mbx. Effectively a typo.
2789 15. The spa authenticator server code was not checking that the input it
2790 received was valid base64.
2792 16. The debug output line for the "set" modifier in ACLs was not showing the
2793 name of the variable that was being set.
2795 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2797 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2798 lookup on its own. Something else was needed to trigger the lookup. For
2799 example, a match in host_lookup or the need for a host name in a host list.
2800 Now, if $sender_host_name is referenced and the host name has not yet been
2801 looked up, a lookup is performed. If the lookup fails, the variable remains
2802 empty, and $host_lookup_failed is set to "1".
2804 19. Added "eqi" as a case-independent comparison operator.
2806 20. The saslauthd authentication condition could segfault if neither service
2807 nor realm was specified.
2809 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2810 error message that was logged was misleading, and incoming SMTP
2811 connections were dropped. The message is now more accurate, and temporary
2812 errors are given to SMTP connections.
2814 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2815 (except RSET) with a 421 error, until QUIT is received. However, it was
2816 failing to send a response to QUIT.
2818 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2819 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2820 ACL is now run after the helo_try_verify_hosts code.
2822 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2823 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2824 case-independent, so other case variants are also recognized). Apparently
2825 some systems use these upper case variants.
2827 25. If more than two messages were waiting for the same host, and a transport
2828 filter was specified for the transport, Exim sent two messages over the
2829 same TCP/IP connection, and then failed with "socket operation on non-
2830 socket" when it tried to send the third.
2832 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2834 27. The extern definition of crypt16() in expand.c was not being excluded when
2835 the OS had its own crypt16() function.
2837 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2838 as a preferred synonym for return_size_limit, both as an option and as an
2841 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2843 30. Changed the default configuration ACL to relax the local part checking rule
2844 for addresses that are not in any local domains. For these addresses,
2845 slashes and pipe symbols are allowed within local parts, but the sequence
2846 /../ is explicitly forbidden.
2848 31. SPA server authentication was not clearing the challenge buffer before
2851 32. log_message in a "warn" ACL statement was writing to the reject log as
2852 well as to the main log, which contradicts the documentation and doesn't
2853 seem right (because no rejection is happening). So I have stopped it.
2855 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2856 However, I am unable to do any testing of this.
2858 34. Fixed an infelicity in the appendfile transport. When checking directories
2859 for a mailbox, to see if any needed to be created, it was accidentally
2860 using path names with one or more superfluous leading slashes; tracing
2861 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2863 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2864 contents are added to the log line that is written for every discarded
2865 recipient. (Previously a log_message setting was ignored.)
2867 36. The ${quote: operator now quotes the string if it is empty.
2869 37. The install script runs exim in order to find its version number. If for
2870 some reason other than non-existence or emptiness, which it checks, it
2871 could not run './exim', it was installing it with an empty version number,
2872 i.e. as "exim-". This error state is now caught, and the installation is
2875 38. An argument was missing from the function that creates an error message
2876 when Exim fails to connect to the socket for saslauthd authentication.
2877 This could cause Exim to crash, or give a corrupted message.
2879 39. Added isip, isip4, and isip6 to ${if conditions.
2881 40. The ACL variables $acl_xx are now saved with the message, and can be
2882 accessed later in routers, transports, and filters.
2884 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2885 strings in the file are not string-expanded.
2887 42. If a MAIL command specified a SIZE value that was too large to fit into an
2888 int variable, the check against message_size_limit failed. Such values are
2889 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2890 day this will have to be increased, but I don't think I want to be around
2891 when emails are that large.
2898 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2899 iconv() is not standard in FreeBSD.
2901 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2902 IPv6 enabled. The observed symptom was a segmentation fault on return from
2903 the function os_common_find_running_interfaces() in src/os.c.
2905 3. In the check_special_case() function in daemon.c I had used "errno" as an
2906 argument name, which causes warnings on some systems. This was basically a
2907 typo, since it was named "eno" in the comments!
2909 4. The code that waits for the clock to tick (at a resolution of some fraction
2910 of a second) so as to ensure message-id uniqueness was always waiting for
2911 at least one whole tick, when it could have waited for less. [This is
2912 almost certainly not relevant at current processor speeds, where it is
2913 unlikely to ever wait at all. But we try to future-proof.]
2915 5. The function that sleeps for a time interval that includes fractions of a
2916 second contained a race. It did not block SIGALRM between setting the
2917 timer, and suspending (a couple of lines later). If the interval was short
2918 and the sigsuspend() was delayed until after it had expired, the suspension
2919 never ended. On busy systems this could lead to processes getting stuck for
2922 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2923 process, before it forks any delivery processes. The open lookup caching
2924 mechanism meant that the open file or database connection was passed into
2925 the delivery process. The problem was that delivery processes always tidy
2926 up cached lookup data. This could cause a problem for the next delivery
2927 process started by the queue runner, because the external queue runner
2928 process does not know about the closure. So the next delivery process
2929 still has data in the lookup cache. In the case of a file lookup, there was
2930 no problem because closing a file descriptor in a subprocess doesn't affect
2931 the parent. However, if the lookup was caching a connection to a database,
2932 the connection was closed, and the second delivery process was likely to
2933 see errors such as "PGSQL: query failed: server closed the connection
2934 unexpectedly". The problem has been fixed by closing all cached lookups
2935 in a queue runner before running a delivery process.
2937 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2938 seem to have the "const" qualifier which it has on other OS. I've
2941 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2942 *of the same type* (client or server) with the same public name that an
2943 error should be diagnosed.
2945 9. When Exim looked up a host name for an IP address, but failed to find the
2946 original IP address when looking up the host name (a safety check), it
2947 output the message "<ip address> does not match any IP for NULL", which was
2948 confusing, to say the least. The bug was that the host name should have
2949 appeared instead of "NULL".
2951 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2952 user that is built into the binary, and the -C or -D options is used, root
2953 privilege is dropped before the configuration file is read. In addition,
2954 logging is switched to stderr instead of the normal log files. If the
2955 configuration then re-defines the Exim user, the unprivileged environment
2956 is probably not what is expected, so Exim logs a panic warning message (but
2959 However, if deliver_drop_privilege is set, the unprivileged state may well
2960 be exactly what is intended, so the warning has been cut out in that case,
2961 and Exim is allowed to try to write to its normal log files.
2967 1. smtp_return_error_details was not giving details for temporary sender
2968 or receiver verification errors.
2970 2. Diagnose a configuration error if two authenticators have the same public
2973 3. Exim used not to create the message log file for a message until the first
2974 delivery attempt. This could be confusing when incoming messages were held
2975 for policy or load reasons. The message log file is now created at the time
2976 the message is received, and an initial "Received" line is written to it.
2978 4. The automatically generated man page for command line options had a minor
2979 bug that caused no ill effects; however, a more serious problem was that
2980 the procedure for building the man page automatically didn't always
2981 operate. Consequently, release 4.20 contains an out-of-date version. This
2982 shouldn't happen again.
2984 5. When building Exim with embedded Perl support, the script that builds the
2985 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2986 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2988 6. The freeze_tell option was not being used for messages that were frozen on
2989 arrival, either by an ACL or by local_scan().
2991 7. Added the smtp_incomplete_transaction log selector.
2993 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2994 was accepting AUTH without a new EHLO.
2996 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2997 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3000 10. Logging of TCP/IP connections (when configured) now happens in the main
3001 daemon process instead of the child process, so that the TCP/IP connection
3002 count is more accurate (but it can never be perfect).
3004 11. The use of "drop" in a nested ACL was not being handled correctly in the
3005 outer ACL. Now, if condition failure induced by the nested "drop" causes
3006 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3007 or "require"), the connection is dropped.
3009 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3010 that yield "discard" can now be used with an "accept" or a "discard" verb,
3011 but an error is generated for any others (because I can't see a useful way
3012 to define what should happen).
3014 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3015 are now processed in the same way as lines in the Exim configuration file.
3016 In particular, continuation lines are supported.
3018 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3020 15. Added -ti meaning -t -i.
3022 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3023 domains, and warn by logging if others are found.
3025 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3026 returned from GIFCONF when Exim is trying to find the list of interfaces on
3027 a host. The code in os.c has been modified to copy each ifreq to an aligned
3028 structure in all cases.
3030 Also, in some cases, the returned ifreq's were being copied to a 'struct
3031 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3032 means the last couple of bytes of an IPv6 address could be chopped if the
3033 ifreq contained only a normal sockaddr (14 bytes storage).
3035 18. Named domain lists were not supported in the hosts_treat_as_local option.
3036 An entry such as +xxxx was not recognized, and was treated as a literal
3039 19. Ensure that header lines added by a DATA ACL are included in the reject log
3040 if the ACL subsequently rejects the message.
3042 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3043 MD5 (which is deprecated).
3045 21. When testing a filter file using -bf, Exim was writing a message when it
3046 took the sender from a "From " line in the message, but it was not doing so
3047 when it took $return_path from a Return-Path: header line. It now does.
3049 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3050 with a valid header line field name (a series of printing characters
3051 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3053 23. Changed "disc" in the source to "disk" to conform to the documentation and
3054 the book and for uniformity.
3056 24. Ignore Sendmail's -Ooption=value command line item.
3058 25. When execve() failed while trying to run a command in a pipe transport,
3059 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3060 could be confused with a return value of 69 from the command itself. This
3061 has been changed to 127, the value the shell returns if it is asked to run
3062 a non-existent command. The wording for the related log line suggests a
3063 non-existent command as the problem.
3065 26. If received_header_text expands to an empty string, do not add a Received:
3066 header line to the message. (Well, it adds a token one on the spool, but
3067 marks it "old" so that it doesn't get used or transmitted.)
3069 27. Installed eximstats 1.28 (addition of -nt option).
3071 28. There was no check for failure on the call to getsockname() in the daemon
3072 code. This can fail if there is a shortage of resources on the system, with
3073 ENOMEM, for example. A temporary error is now given on failure.
3075 29. Contrary to the C standard, it seems that in some environments, the
3076 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3077 program. Exim now does this explicitly; it affects the formatting of
3078 timestamps using strftime().
3080 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3081 complaints. I've now initialized all the variables, to avoid this.
3083 32. Header lines added by a system filter were not being "seen" during
3084 transport-time rewrites.
3086 33. The info_callback() function passed to OpenSSL is set up with type void
3087 (*)(SSL *, int, int), as described somewhere. However, when calling the
3088 function (actually a macro) that sets it up, the type void(*)() is
3089 expected. I've put in a cast to prevent warnings from picky compilers.
3091 34. If a DNS black list lookup found a CNAME record, but there were no A
3092 records associated with the domain it pointed at, Exim crashed.
3094 35. If a DNS black list lookup returned more than one A record, Exim ignored
3095 all but the first. It now scans all returned addresses if a particular IP
3096 value is being sought. In this situation, the contents of the
3097 $dnslist_value variable are a list of all the addresses, separated by a
3100 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3101 to accept a host name and all its aliases if the forward lookup for any of
3102 them yielded the IP address of the incoming connection. Now it accepts only
3103 those names whose forward lookup yields the correct IP address. Any other
3104 names are discarded. This closes a loophole whereby a rogue DNS
3105 administrator could create reverse DNS records to break through a
3106 wildcarded host restriction in an ACL.
3108 37. If a user filter or a system filter that ran in a subprocess used any of
3109 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3110 the wrong values were passed to the pipe command ($thisaddress had the
3111 value of $0, $0 had the value of $1, etc). This bug was introduced by
3112 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3114 38. Improved the line breaking for long SMTP error messages from ACLs.
3115 Previously, if there was no break point between 40 and 75 characters, Exim
3116 left the rest of the message alone. Two changes have been made: (a) I've
3117 reduced the minimum length to 35 characters; (b) if it can't find a break
3118 point between 35 and 75 characters, it looks ahead and uses the first one
3119 that it finds. This may give the occasional overlong line, but at least the
3120 remaining text gets split now.
3122 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3123 file descriptors might be low, and that setting 1000 would always raise it.
3124 It turns out that in some environments, the limit is already over 1000 and
3125 that lowering it causes trouble. So now Exim takes care not to decrease it.
3127 40. When delivering a message, the value of $return_path is set to $sender_
3128 address at the start of routing (routers may change the value). By an
3129 oversight, this default was not being set up when an address was tested by
3130 -bt or -bv, which affected the outcome if any router or filter referred to
3133 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3134 log only when "message" or "log_message" are set. However, if one of the
3135 conditions was an address verification, or a call to a nested ACL, the
3136 messages generated by the underlying test were being passed through. This
3137 no longer happens. The underlying message is available in $acl_verify_
3138 message for both "message" and "log_message" expansions, so it can be
3139 passed through if needed.
3141 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3142 new expansion $bh_ to give the encoded byte string without charset
3143 translation. Translation happens only if iconv() is available; HAVE_ICONV
3144 indicates this at build time. HEADERS_CHARSET gives the charset to
3145 translate to; headers_charset can change it in the configuration, and
3146 "headers charset" can change it in an individual filter file.
3148 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3149 that creates RFC 2047 encoded "words" labels them as that charset instead
3150 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3151 expansion operator; (ii) when Exim creates a From: line for a local
3152 message; (iii) when a header line is rewritten to include a "phrase" part.
3154 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3155 buggy, causing it to skip the first lines of messages whose message ID
3156 ended in 'D'. This would not have bitten before Exim release 4.14, because
3157 message IDs were unlikely to end in 'D' before then. The effect was to have
3158 incorrect size information for certain domains.
3160 45. #include "config.h" was missing at the start of the crypt16.c module. This
3161 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3164 46. If there was a timeout during a "random" callout check, Exim treated it as
3165 a failure of the random address, and carried on sending RSET and the real
3166 address. If the delay was just some slowness somewhere, the response to the
3167 original RCPT would be taken as a response to RSET and so on, causing
3168 mayhem of various kinds.
3170 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3171 when I implemented it. It didn't allow for the fact that some option values
3172 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3173 the right test for positive values.
3175 48. Domain names in DNS records are case-independent. Exim always looks them up
3176 in lower case. Some resolvers return domain names in exactly the case they
3177 appear in the zone file, that is, they may contain uppercase letters. Not
3178 all resolvers do this - some return always lower case. Exim was treating a
3179 change of case by a resolver as a change of domain, similar to a widening
3180 of a domain abbreviation. This triggered its re-routing code and so it was
3181 trying to route what was effectively the same domain again. This normally
3182 caused routing to fail (because the router wouldn't handle the domain
3183 twice). Now Exim checks for this case specially, and just changes the
3184 casing of the domain that it ultimately uses when it transmits the message
3187 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3190 50. If a filter generated a file delivery with a non-absolute name (possible if
3191 no home directory exists for the router), the forbid_file option was not
3194 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3195 the existing equality matching.
3197 52. Exim was using ints instead of ino_t variables in some places where it was
3198 dealing with inode numbers.
3200 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3201 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3202 called TMPDIR, and if it finds it is different, it changes its value.
3204 54. The smtp_printf() function is now made available to local_scan() so
3205 additional output lines can be written before returning. There is also an
3206 smtp_fflush() function to enable the detection of a dropped connection.
3207 The variables smtp_input and smtp_batched_input are exported to
3210 55. Changed the default runtime configuration: the message "Unknown user"
3211 has been removed from the ACL, and instead placed on the localuser router,
3212 using the cannot_route_message feature. This means that any verification
3213 failures that generate their own messages won't get overridden. Similarly,
3214 the "Unrouteable address" message that was in the ACL for unverifiable
3215 relay addresses has also been removed.
3217 56. Added hosts_avoid_esmtp to the smtp transport.
3219 57. The exicyclog script was not checking for the esoteric option
3220 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3221 will work only if exicyclog is run under the appropriate euid.
3223 58. Following a discussion on the list, the rules by which Exim recognises line
3224 endings on incoming messages have been changed. The -dropcr and drop_cr
3225 options are now no-ops, retained only for backwards compatibility. The
3226 following line terminators are recognized: LF CRLF CR. However, special
3227 processing applies to CR:
3229 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3230 nor a local message in the state where . is a terminator.
3232 (ii) If a bare CR is encountered in a header line, an extra space is added
3233 after the line terminator so as not to end the header. The reasoning
3234 behind this is that bare CRs in header lines are most likely either
3235 to be mistakes, or people trying to play silly games.
3237 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3238 was being incorrectly given as 18 bytes larger than it should have been.
3239 This is a VOB (very old bug).
3241 60. This may never have affected anything current, but just in case it has:
3242 When the local host is found other than at the start of a list of hosts,
3243 the local host, those with the same MX, and any that follow, are discarded.
3244 When the list in question was part of a longer list of hosts, the following
3245 hosts (not currently being processed) were also being discarded. This no
3246 longer happens. I'm not sure if this situation could ever has previously
3249 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3252 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3253 header. This is something that is recommended in a new Internet Draft, and
3254 is something that is documented as being done by Sendmail. There are two
3255 possible values. For messages generated by the autoreply transport, Exim
3258 Auto-Submitted: auto-replied
3260 whereas for all other generated messages (e.g. bounces) it adds
3262 Auto-Submitted: auto-generated
3264 63. The "personal" condition in filters now includes a test for the
3265 Auto-Submitted: header. If it contains the string "auto-" the message it
3266 not considered personal.
3268 64. Added rcpt_include_affixes as a generic transport option.
3270 65. Added queue_only_override (default true).
3272 66. Added the syslog_duplication option.
3274 67. If what should have been the first header line of a message consisted of
3275 a space followed by a colon, Exim was mis-interpreting it as a header line.
3276 It isn't of course - it is syntactically invalid and should therefore be
3277 treated as the start of the message body. The misbehaviour could have
3278 caused a number of strange effects, including loss of data in subsequent
3279 header lines, and spool format errors.
3281 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3282 client host had authenticated. This control can now be exercised by an ACL
3283 for more flexibility.
3285 69. By default, callouts do not happen when testing with -bh. There is now a
3286 variant, -bhc, which does actually run the callout code, including
3287 consulting and updating the callout cache.
3289 70. Added support for saslauthd authentication, courtesy of Alexander
3292 71. If statvfs() failed on the spool or log directories while checking their
3293 size for availability, Exim confusingly gave the error "space shortage".
3294 Furthermore, in debugging mode it crashed with a floating point exception.
3295 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3296 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3297 serious problem, Exim now writes to the main and panic logs when this
3298 happens, with details of the failure. It then refuses to accept the
3299 incoming message, giving the message "spool directory problem" or "log
3300 directory problem" with a 421 code for SMTP messages.
3302 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3303 0, 1, and 2 exist, because some OS complain for execs without them (see
3304 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3305 descriptors. However, the code omitted to check that the open succeeded,
3306 causing mysterious errors if for some reason the permissions on /dev/null
3307 got screwed. Now Exim writes a message to the main and panic logs, and
3308 bombs out if it can't open /dev/null.
3310 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3311 interact so that it is all more flexible. It is supposed to remain
3312 backwards compatible. Also added extra_local_interfaces.
3314 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3315 to bomb out with an assertion failure - to the client this appears as a
3316 connection drop. This problem occurs in the part of the code that was taken
3317 from the Samba project. Fortunately, the assertion is in a very simple
3318 function, so I have fixed this by reproducing the function inline in the
3319 one place where it is called, and arranging for authentication to fail
3320 instead of killing the process with assert().
3322 75. The SPA client code was not working when the server requested OEM rather
3323 than Unicode encoding.
3325 76. Added code to make require_files with a specific uid setting more usable in
3326 the case where statting the file as root fails - usually a non-root-mounted
3327 NFS file system. When this happens and the failure is EACCES, Exim now
3328 forks a subprocess and does the per-uid checking as the relevant uid.
3330 77. Added process_log_path.
3332 78. If log_file_path was not explicitly set, a setting of check_log_space or
3333 check_log_inodes was ignored.
3335 79. If a space check for the spool or log partitions fails, the incident is now
3336 logged. Of course, in the latter case the data may get lost...
3338 80. Added the %p formatting code to string_format() so that it can be used to
3339 print addresses in debug_print(). Adjusted all the address printing in the
3340 debugging in store.c to use %p rather than %d.
3342 81. There was a concern that a line of code in smtp_in.c could overflow a
3343 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3344 initially expressed, the concern was not well-founded, because trailing
3345 spaces are removed early. However, if the trailing spaces were followed by
3346 a NULL, they did not get removed, so the overflow was possible. Two fixes
3349 (a) I re-wrote the offending code in a cleaner fashion.
3350 (b) If an incoming SMTP command contains a NULL character, it is rejected
3353 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3354 runs initgroups(), so that if the Exim user is in any additional groups,
3355 they will be used during message reception.
3361 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob