1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.225 2005/09/13 09:54:14 fanf2 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added the "success_on_redirect" address verification option. See
10 NewStuff for rationale and an example.
12 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
14 PH/02 Patch to exigrep to allow it to work on syslog lines.
16 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
17 fread() to skip over the body file's header line, because in Cygwin the
18 header line is locked and is inaccessible.
20 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
21 co-exist for some time) to make it clear that it is the Exim ID that is
22 referenced, not the Message-ID: header line.
24 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
25 string_format() function, because snprintf() does not exist on all
28 PH/06 The use of forbid_filter_existstest now also locks out the use of the
29 ${stat: expansion item.
31 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
32 protocol synchronization error", to keep the pedants happy.
34 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
35 well as for IRIX systems, when gcc is being used. See the host.c source
38 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
40 PH/10 Named domain lists were not working if used in a queue_smtp_domains
43 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
44 transport and to the smtp transport in LMTP mode.
46 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
48 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
49 run a filter in a subprocess. This could lead to confusion in subsequent
50 lookups in the parent process. There should also be a search_tidyup() at
51 the end of the subprocess.
53 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
54 only if the host matched helo_try_verify_hosts, which caused the
55 verification to occur when the EHLO/HELO command was issued. The ACL just
56 tested the remembered result. Now, if a previous verification attempt has
57 not happened, "verify = helo" does it there and then.
59 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
61 TK/03 Fix log output including CR from clamd.
63 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
64 exist provoked a memory error which could cause a segfault.
66 PH/15 Installed PCRE 6.2
68 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
70 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
71 of the problem. Specifically, suggested +O2 rather than +O1 for the
74 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
76 PH/20 If a delivery was routed to a non-standard port by means of an SRV
77 record, the port was not correctly logged when the outgoing_port log
78 selector was set (it logged the transort's default port).
80 PH/21 Added support for host-specific ports to manualroute, queryprogram,
81 fallback_hosts, and "hosts" in the smtp transport.
83 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
84 host errors such as "Connection refused".
86 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
87 authentication with radiusclient 0.4.9:
89 - Error returned from rc_read_config was caught wrongly
90 - Username/password not passed on to radius server due to wrong length.
92 The presumption is that some radiusclient API changes for 4.51/PH/17
93 were not taken care of correctly. The code is still untested by me (my
94 Linux distribution still has 0.3.2 of radiusclient), but it was
95 contributed by a Radius user.
97 PH/24 When doing a callout, the value of $domain wasn't set correctly when
98 expanding the "port" option of the smtp transport.
100 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
101 while reading a MIME header. Thanks to Tom Hughes for a patch.
103 PH/24 Include config.h inside local_scan.h so that configuration settings are
106 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
107 that in an ACL for RCPT (for example), you can examine exactly what was
110 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
111 commands, but it was not correctly comparing the address with the actual
112 client host address. Thus, it would show the EHLO address in Received:
113 header lines when this was not necessary.
115 PH/27 Added the % operator to ${eval:}.
117 PH/28 Exim tries to create and chdir to its spool directory when it starts;
118 it should be ignoring failures (because with -C, for example, it has lost
119 privilege). It wasn't ignoring creation failures other than "already
122 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
125 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
126 because an input line was too long, either on its own, or by virtue of
127 too many continuations, the temporary file was not being removed, and the
128 return code was incorrect.
130 PH/31 Missing "BOOL" in function definition in filtertest.c.
132 PH/32 Applied Sieve patches from the maintainer.
134 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
136 PH/33 Added "verify = not_blind".
138 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
139 Local/Makefile (with some defaults set). These are used in built scripts
140 such as exicyclog, but they have never been used in the exim_install
141 script (though there are many overriding facilities there). I have
142 arranged that the exim_install script now takes note of these two
145 PH/35 Installed configuration files for Dragonfly.
147 PH/36 When a locally submitted message by a trusted user did not contain a
148 From: header, and the sender address was obtained from -f or from an SMTP
149 MAIL command, and the trusted user did not use -F to supply a sender
150 name, $originator_name was incorrectly used when constructing a From:
151 header. Furthermore, $originator_name was used for submission mode
152 messages from external hosts without From: headers in a similar way,
153 which is clearly wrong.
155 PH/37 Added control=suppress_local_fixups.
157 PH/38 When log_selector = +received_sender was set, and the addition of the
158 sender made the log line's construction buffer exactly full, or one byte
159 less than full, an overflow happened when the terminating "\n" was
162 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
163 when the result of a list match is failure because a DNS lookup failed.
165 PH/40 RM_COMMAND is now used in the building process.
167 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
168 the "build-* directories that it finds.
170 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
171 address in a domain literal was a prefix of an interface address.
173 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
174 when verifying a sender address, unless rewrite_headers is false.
176 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
177 the ratelimit ACL was added.
183 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
185 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
186 command itself ended in error, the underlying message about the transport
187 filter timeout was being overwritten with the pipe command error. Now the
188 underlying error message should be appended to the second error message.
190 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
192 PH/02 Reduce the amount of output that "make" produces by default. Full output
193 can still be requested.
195 PH/03 The warning log line about a condition test deferring for a "warn" verb
196 was being output only once per connection, rather than after each
197 occurrence (because it was using the same function as for successful
198 "warn" verbs). This seems wrong, so I have changed it.
200 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
201 it should not have, which might have caused a crash in the right
202 circumstances, but probably never did.
204 PH/04 Installed a modified version of Tony Finch's patch to make submission
205 mode fix the return path as well as the Sender: header line, and to
206 add a /name= option so that you can make the user's friendly name appear
209 TF/03 Added the control = fakedefer ACL modifier.
211 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
212 Mark Lowes for thorough testing.
214 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
216 TK/03 Merged latest SRS patch from Miles Wilton.
218 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
219 with the definition in sysexits.h (which is #included earlier).
220 Fortunately, Exim does not actually use EX_OK. The code used to try to
221 preserve the sysexits.h value, by assumimg that macro definitions were
222 scanned for macro replacements. I have been disabused of this notion,
223 so now the code just undefines EX_OK before #including unistd.h.
225 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
226 in the smtp transport. When a block could not be written in a single
227 write() function, the timeout was being re-applied to each part-write.
228 This seems wrong - if the receiver was accepting one byte at a time it
229 would take for ever. The timeout is now adjusted when this happens. It
230 doesn't have to be particularly precise.
232 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
233 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
235 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
236 without a domain if the check to <postmaster@domain> fails.
238 SC/01 Eximstats: added -xls and the ability to specify output files
239 (patch written by Frank Heydlauf).
241 SC/02 Eximstats: use FileHandles for outputing results.
243 SC/03 Eximstats: allow any combination of xls, txt, and html output.
245 SC/04 Eximstats: fixed display of large numbers with -nvr option
247 SC/05 Eximstats: fixed merging of reports with empty tables.
249 SC/06 Eximstats: added the -include_original_destination flag
251 SC/07 Eximstats: removed tabs and trailing whitespace.
253 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
255 TK/06 MBOX spool code: Add real "From " MBOX separator line
256 so the .eml file is really in mbox format (even though
257 most programs do not really care). Patch from Alex Miller.
259 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
260 The latter is generated from $received_to and is only set if the
261 message has one envelope recipient. SA can use these headers,
262 obviously out-of-the-box. Patch from Alex Miller.
264 PH/08 The ${def test on a variable was returning false if the variable's
265 value was "0", contrary to what the specification has always said!
266 The result should be true unless the variable is empty.
268 PH/09 The syntax error of a character other than { following "${if
269 def:variable_name" (after optional whitespace) was not being diagnosed.
270 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
271 accidental colon was present, for example, could give incorrect results.
273 PH/10 Tidied the code in a number of places where the st_size field of a stat()
274 result is used (not including appendfile, where other changes are about
277 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
278 This involved changing a lot of size variables from int to off_t. It
279 should work with maildirs and everything.
281 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
282 spamd dying while we are connected to it.
284 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
287 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
288 function for BDB 4.3.
290 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
292 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
293 the sources for PCRE have been re-arranged and more files are now
296 PH/15 The code I had for printing potentially long long variables in PH/11
297 above was not the best (it lost precision). The length of off_t variables
298 is now inspected at build time, and an appropriate printing format (%ld
299 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
300 to be "long long int" or "long int". This is needed for the internal
301 formatting function string_vformat().
303 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
304 the configuration file to be ":syslog", then the script "guesses" where
305 the logs files are, rather than using the compiled in default. In our
306 case the guess is not the same as the compiled default, so the script
307 suddenly stopped working when I started to use syslog. The patch checks
308 to see if log_file_path is "". If so, it attempts to read it from exim
309 with no configuration file to get the compiled in version, before it
310 falls back to the previous guessing code."
312 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
313 implementing BATV in an Exim configuration. See NewStuff for the gory
316 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
317 Makefile that are specific to HP-UX.
319 PH/18 If the "use_postmaster" option was set for a recipient callout together
320 with the "random" option, the postmaster address was used as the MAIL
321 FROM address for the random test, but not for the subsequent recipient
322 test. It is now used for both.
324 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
325 patch removes a few documentation additions to RFC 3028, because the
326 latest draft now contains them. It adds the new en;ascii-case comparator
327 and a new error check for 8bit text in MIME parts. Comparator and
328 require names are now matched exactly. I enabled the subaddress
329 extension, but it is not well tested yet (read: it works for me)."
331 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
332 rework some of the code of TK/09 above to avoid the hardwired use of
333 "%lld" and "long long". Replaced the call to snprintf() with a call to
336 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
337 records point to non-existent hosts", "retry timeout exceeded", and
338 "retry time not reached for any host after a long failure period".
340 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
341 experimental DomainKeys support:
343 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
344 (2) On an error such as an illegally used "control", the wrong name for
345 the control was given.
347 These problems did NOT occur unless DomainKeys support was compiled.
349 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
351 PH/24 Added ${if match_ip condition.
353 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
354 databases so that it will be absolutely obvious if a crash occurs in the
355 DB library. This is a regular occurrence (often caused by mis-matched
358 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
359 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
360 picked up on a user's system that detects such things. There doesn't seem
361 to be a gcc warning option for this - only an attribute that has to be
362 put on the function's prototype. It seems that in Fedora Core 4 they have
363 set this on a number of new functions. No doubt there will be more in due
366 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
367 specify a transport. However, if an address that was verified by such a
368 router was the subject of a callout, Exim crashed because it tried to
369 read the rcpt_include_affixes from the non-existent transport. Now it
370 just assumes that the setting of that option is false. This bug was
371 introduced by 4.51/PH/31.
373 PH/28 Changed -d+all to exclude +memory, because that information is very
374 rarely of interest, but it makes the output a lot bigger. People tend to
375 do -d+all out of habit.
377 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
378 code in os-type was giving problems when libc.so lives in lib64, like on
381 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
382 aren't the modern standard, and it seems that some systems' include files
383 don't always have them. Exim was already checking for some of the newer
384 ones like T_AAAA, and defining it itself. I've added checks for all the
385 record types that Exim uses.
387 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
388 not automatically generating a new one, as it is supposed to. This
389 prevented TLS from working. If the file did exist, but contained invalid
390 data, a new version was generated, as expected. It was only the case of a
391 non-existent file that was broken.
393 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
394 with a change in libdomainkeys > 0.64.
396 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
397 from DNS. If the selector record carries the flag, it now has
398 precedence over the domain-wide flag.
400 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
402 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
403 the use of a port name that isn't defined in /etc/services) occurred, the
404 message was deferred as in a normal delivery, and thus remained on the
405 spool, instead of being failed because of the mua_wrapper setting. This
406 is now fixed, and I tidied up some of the mua_wrapper messages at the
409 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
410 the messages in a hash of arrays rather than using individual hashes.
411 This is a bit cleaner and results in dramatic memory savings, albeit
412 at a slight CPU cost.
414 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
415 as requested by Marc Sherman.
417 SC/10 Eximstats: added histograms for user specified patterns as requested
420 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
422 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
423 fopen() in the content-scanning modules that did not already have it.
429 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
430 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
432 TK/02 Fix ACL "control" statement not being available in MIME ACL.
434 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
436 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
437 to test Sieve filters that use "vacation".
439 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
440 that changes the way the GnuTLS parameters are stored in the cache file.
441 The new format can be generated externally. For backward compatibility,
442 if the data in the cache doesn't make sense, Exim assumes it has read an
443 old-format file, and it generates new data and writes a new file. This
444 means that you can't go back to an older release without removing the
447 PH/03 A redirect router that has both "unseen" and "one_time" set does not
448 work if there are any delivery delays because "one_time" forces the
449 parent to be marked "delivered", so its unseen clone is never tried
450 again. For this reason, Exim now forbids the simultaneous setting of
453 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
454 redirected to themselves ("homonym" addresses). Read the long ChangeLog
455 entry if you want to know the details. The fix, however, neglected to
456 consider the case when local delivery batching is involved. The test for
457 "previously delivered" was not happening when checking to see if an
458 address could be batched with a previous (undelivered) one; under
459 certain circumstances this could lead to multiple deliveries to the same
462 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
463 in its include files, and this causes problems building Exim.
465 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
466 header_syntax) but Exim was just ignoring anything given after a slash.
467 In particular, this caused confusion with an attempt to use "verify =
468 reverse_host_lookup/defer_ok". An error is now given when options are
469 supplied for verify items that do not have them. (Maybe reverse_host_
470 lookup should have a defer_ok option, but that's a different point.)
472 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
473 defined by RFC 821) to 2048, because there were problems with some AUTH
474 commands, and RFC 1869 says the size should be increased for extended
475 SMTP commands that take arguments.
477 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
480 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
481 "unknown" error; now it says that the functionality isn't in the binary.
483 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
484 an address' error message when a string expansion fails (syntax or
485 whatever). Otherwise the password may appear in the log. Following change
486 PH/42 below, there is no longer a chance of it appearing in a bounce
489 PH/11 Installed exipick version 20050225.0 from John Jetmore.
491 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
492 of its addresses was ever tried. (Bugzilla bug #2.)
494 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
495 the result incorrectly in the debug output. (It correctly added a newline
496 to what was transported.)
498 TF/01 Added $received_time.
500 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
501 commented out examples of how to interface to a virus scanner and to
502 SpamAssassin. Also added commented examples of av_scanner and
503 spamd_address settings.
505 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
506 and controls are allowed in which ACLs. There were a couple of minor
507 errors. Some of the entries in the conditions table (which is a table of
508 where they are NOT allowed) were getting very unwieldy; rewrote them as a
509 negation of where the condition IS allowed.
511 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
513 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
514 header file does not have a version number, so I've had to invent a new
515 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
516 API. The code is untested by me (my Linux distribution still has 0.3.2 of
517 radiusclient), but it was contributed by a Radius user.
519 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
520 files or directories, for OpenSSL.
522 PH/19 When an Exim process that is running as root has to create an Exim log
523 file, it does so in a subprocess that runs as exim:exim so as to get the
524 ownership right at creation (otherwise, other Exim processes might see
525 the file with the wrong ownership). There was no test for failure of this
526 fork() call, which would lead to the process getting stuck as it waited
527 for a non-existent subprocess. Forks do occasionally fail when resources
528 run out. I reviewed all the other calls to fork(); they all seem to check
531 PH/20 When checking for unexpected SMTP input at connect time (before writing
532 the banner), Exim was not dealing correctly with a non-positive return
533 from the read() function. If the client had disconnected by this time,
534 the result was a log entry for a synchronization error with an empty
535 string after "input=" when read() returned zero. If read() returned -1
536 (an event I could not check), uninitialized data bytes were printed.
537 There were reports of junk text (parts of files, etc) appearing after
540 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
542 PH/22 Added support for macro redefinition, and (re)definition in between
543 driver and ACL definitions.
545 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
546 forgetting to use the resulting value; it was using the unexpanded value.
548 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
549 hadn't been configured. The fix is from Juergen Kreileder, who
550 understands it better than I do:
552 "Here's what I see happening with three configured cyrus_sasl
553 authenticators configured (plain, login, cram-md5):
555 On startup auth_cyrus_sasl_init() gets called for each of these.
556 This means three calls to sasl_listmech() without a specified mech_list.
557 => SASL tests which mechs of all available mechs actually work
558 => three warnings about OTP not working
559 => the returned list contains: plain, login, cram-md5, digest-md5, ...
561 With the patch, sasl_listmech() also gets called three times. But now
562 SASL's mech_list option is set to the server_mech specified in the the
563 authenticator. Or in other words, the answer from sasl_listmech()
564 gets limited to just the mech you're testing for (which is different
566 => the return list contains just 'plain' or 'login', 'cram-md5' or
567 nothing depending on the value of ob->server_mech.
569 I've just tested the patch: Authentication still works fine,
570 unavailable mechs specified in the exim configuration are still
571 caught, and the auth.log warnings about OTP are gone."
573 PH/25 When debugging is enabled, the contents of the command line are added
574 to the debugging output, even when log_selector=+arguments is not
577 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
578 answer is "GNU", and only if the return is "GNU/something" is the answer
581 PH/27 $acl_verify_message is now set immediately after the failure of a
582 verification in an ACL, and so is available in subsequent modifiers. In
583 particular, the message can be preserved by coding like this:
585 warn !verify = sender
586 set acl_m0 = $acl_verify_message
588 Previously, $acl_verify_message was set only while expanding "message"
589 and "log_message" when a very denied access.
591 PH/28 Modified OS/os.c-Linux with
593 -#ifndef OS_LOAD_AVERAGE
594 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
596 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
597 nomenclature these days.)
599 PH/29 Installed patch from the Sieve maintainer that adds the options
600 sieve_useraddress and sieve_subaddress to the redirect router.
602 PH/30 In these circumstances:
603 . Two addresses routed to the same list of hosts;
604 . First host does not offer TLS;
605 . First host accepts first address;
606 . First host gives temporary error to second address;
607 . Second host offers TLS and a TLS session is established;
608 . Second host accepts second address.
609 Exim incorrectly logged both deliveries with the TLS parameters (cipher
610 and peerdn, if requested) that were in fact used only for the second
613 PH/31 When doing a callout as part of verifying an address, Exim was not paying
614 attention to any local part prefix or suffix that was matched by the
615 router that accepted the address. It now behaves in the same way as it
616 does for delivery: the affixes are removed from the local part unless
617 rcpt_include_affixes is set on the transport.
619 PH/32 Add the sender address, as F=<...>, to the log line when logging a
620 timeout during the DATA phase of an incoming message.
622 PH/33 Sieve envelope tests were broken for match types other than :is. I have
623 applied a patch sanctioned by the Sieve maintainer.
625 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
626 the uid or gid is negative. A case of a negative gid caused this to be
627 noticed. The fix allows for either to be negative.
629 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
630 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
633 PH/36 The change PH/12 above was broken. Fixed it.
635 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
636 the grounds that routing the same address twice would always produce the
637 same answer. This might have been true once, but it is certainly no
638 longer true now. Routing a child address may depend on the previous
639 routing that produced that child. Some complicated redirection strategies
640 went wrong when messages had multiple recipients, and made Exim's
641 behaviour dependent on the order in which the addresses were given.
643 I have moved the duplicate checking until after the routing is complete.
644 Exim scans the addresses that are assigned to local and remote
645 transports, and removes any duplicates. This means that more work will be
646 done, as duplicates will always all be routed, but duplicates are
647 presumably rare, so I don't expect this is of any significance.
649 For deliveries to pipes, files, and autoreplies, the duplicate checking
650 still happens during the routing process, since they are not going to be
653 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
654 It corrects a timeout issue with spamd. This is Ian's comment: "The
655 background is that sometimes spamd either never reads data from a
656 connection it has accepted, or it never writes response data. The exiscan
657 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
658 blindly assumes that writes won't block so it may never time out."
660 PH/39 Allow G after quota size as well as K and M.
662 PH/40 The value set for $authenticated_id in an authenticator may not contain
663 binary zeroes or newlines because the value is written to log lines and
664 to spool files. There was no check on this. Now the value is run through
665 the string_printing() function so that such characters are converted to
666 printable escape sequences.
668 PH/41 $message_linecount is a new variable that contains the total number of
669 lines in the message. Compare $body_linecount, which is the count for the
672 PH/42 Exim no longer gives details of delivery errors for specific addresses in
673 bounce and delay warning messages, except in certain special cases, which
676 (a) An SMTP error message from a remote host;
677 (b) A message specified in a :fail: redirection;
678 (c) A message specified in a "fail" command in a system filter;
679 (d) A message specified in a FAIL return from the queryprogram router;
680 (e) A message specified by the cannot_route_message router option.
682 In these cases only, Exim does include the error details in bounce and
683 warning messages. There are also a few cases where bland messages such
684 as "unrouteable address" or "local delivery error" are given.
686 PH/43 $value is now also set for the "else" part of a ${run expansion.
688 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
689 being worked on, but at least Exim now implements the latest version to
692 PH/45 In a pipe transport, although a timeout while waiting for the pipe
693 process to complete was treated as a delivery failure, a timeout while
694 writing the message to the pipe was logged, but erroneously treated as a
695 successful delivery. Such timeouts include transport filter timeouts. For
696 consistency with the overall process timeout, these timeouts are now
697 treated as errors, giving rise to delivery failures by default. However,
698 there is now a new Boolean option for the pipe transport called
699 timeout_defer, which, if set TRUE, converts the failures into defers for
700 both kinds of timeout. A transport filter timeout is now identified in
703 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
704 systems where "make" and "gmake" are different, calling "gmake" at top
705 level broke things. I've arranged for the value of $(MAKE) to be passed
706 from the Makefile to this script so that it can call the same version of
710 A note about Exim versions 4.44 and 4.50
711 ----------------------------------------
713 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
714 changes of various kinds. As a consequence, a big documentation update was
715 needed. This delayed the release for rather longer than seemed good, especially
716 in the light of a couple of (minor) security issues. Therefore, the changes
717 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
718 release. So 4.44 and 4.50 are in effect two different branches that both start
721 I have left the 4.50 change log unchanged; it contains all the changes since
722 4.43. The change log for 4.44 is below; many of its items are identical to
723 those for 4.50. This seems to be the most sensible way to preserve the
724 historical information.
730 1. Minor wording change to the doc/README.SIEVE file.
732 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
733 computation of the current number of files was incorrect.
735 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
736 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
739 4. Give more explanation in the error message when the command for a transport
740 filter fails to execute.
742 5. There are several places where Exim runs a non-Exim command in a
743 subprocess. The SIGUSR1 signal should be disabled for these processes. This
744 was being done only for the command run by the queryprogram router. It is
745 now done for all such subprocesses. The other cases are: ${run, transport
746 filters, and the commands run by the lmtp and pipe transports.
748 6. Added CONFIGURE_GROUP build-time option.
750 7. Some older OS have a limit of 256 on the maximum number of file
751 descriptors. Exim was using setrlimit() to set 1000 as a large value
752 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
753 systems. I've change it so that if it can't get 1000, it tries for 256.
755 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
756 was an oversight, and furthermore, ever since the addition of extra
757 controls (e.g. 4.43/32), the checks on when to allow different forms of
758 "control" were broken. There should now be diagnostics for all cases when a
759 control that does not make sense is encountered.
761 9. Added the /retain_sender option to "control=submission".
763 10. $recipients is now available in the predata ACL (oversight).
765 11. Tidy the search cache before the fork to do a delivery from a message
766 received from the command line. Otherwise the child will trigger a lookup
767 failure and thereby defer the delivery if it tries to use (for example) a
768 cached ldap connection that the parent has called unbind on.
770 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
771 of $address_data from the recipient verification was clobbered by the
774 13. The value of address_data from a sender verification is now available in
775 $sender_address_data in subsequent conditions in the ACL statement.
777 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
779 15. Added a new option "connect=<time>" to callout options, to set a different
782 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
783 was its contents. (It was OK if the option was not defined at all.)
785 17. A "Completed" log line is now written for messages that are removed from
786 the spool by the -Mrm option.
788 18. New variables $sender_verify_failure and $recipient_verify_failure contain
789 information about exactly what failed.
791 19. Added -dd to debug only the daemon process.
793 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
794 handles timeouts, both on the server side and network timeouts. Renamed the
795 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
797 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
798 It is now set to "smtps".
800 22. $host_address is now set to the target address during the checking of
803 23. When checking ignore_target_hosts for an ipliteral router, no host name was
804 being passed; this would have caused $sender_host_name to have been used if
805 matching the list had actually called for a host name (not very likely,
806 since this list is usually IP addresses). A host name is now passed as
809 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
810 code that specifies a non-restarting handler (typically sigaction() in
811 modern systems) in an attempt to fix a rare and obscure crash bug.
813 25. Narrowed the window for a race in the daemon that could cause it to ignore
814 SIGCHLD signals. This is not a major problem, because they are used only to
815 wake it up if nothing else does.
817 26. A malformed maildirsize file could cause Exim to calculate negative values
818 for the mailbox size or file count. Odd effects could occur as a result.
819 The maildirsize information is now recalculated if the size or filecount
822 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
823 support for a long time. Removed HAVE_SYS_VFS_H.
825 28. Installed the latest version of exipick from John Jetmore.
827 29. In an address list, if the pattern was not a regular expression, an empty
828 subject address (from a bounce message) matched only if the pattern was an
829 empty string. Non-empty patterns were not even tested. This was the wrong
830 because it is perfectly reasonable to use an empty address as part of a
831 database query. An empty address is now tested by patterns that are
832 lookups. However, all the other forms of pattern expect the subject to
833 contain a local part and a domain, and therefore, for them, an empty
834 address still always fails if the pattern is not itself empty.
836 30. Exim went into a mad DNS loop when attempting to do a callout where the
837 host was specified on an smtp transport, and looking it up yielded more
840 31. Re-factored the code for checking spool and log partition space into a
841 function that finds that data and another that does the check. The former
842 is then used to implement four new variables: $spool_space, $log_space,
843 $spool_inodes, and $log_inodes.
845 32. The RFC2047 encoding function was originally intended for short strings
846 such as real names; it was not keeping to the 75-character limit for
847 encoded words that the RFC imposes. It now respects the limit, and
848 generates multiple encoded words if necessary. To be on the safe side, I
849 have increased the buffer size for the ${rfc2047: expansion operator from
852 33. It is now permitted to omit both strings after an "if" condition; if the
853 condition is true, the result is "true". As before, when the second string
854 is omitted, a false condition yields an empty string. This makes it less
855 cumbersome to write custom ACL and router conditions.
857 34. Failure to deliver a bounce message always caused it to be frozen, even if
858 there was an errors_to setting on the router. The errors_to setting is now
861 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
862 canonical form (fully expanded) before being placed in
863 $sender_host_address.
865 36. The table in the code that translates DNS record types into text (T_A to
866 "A" for instance) was missing entries for NS and CNAME. It is just possible
867 that this could have caused confusion if both these types were looked up
868 for the same domain, because the text type is used as part of Exim's
869 per-process caching. But the chance of anyone hitting this buglet seems
872 37. The dnsdb lookup has been extended in a number of ways.
874 (1) There is a new type, "zns", which walks up the domain tree until it
875 finds some nameserver records. It should be used with care.
877 (2) There is a new type, "mxh", which is like "mx" except that it returns
878 just the host names, not the priorities.
880 (3) It is now possible to give a list of domains (or IP addresses) to be
881 looked up. The behaviour when one of the lookups defers can be
882 controlled by a keyword.
884 (4) It is now possible to specify the separator character for use when
885 multiple records are returned.
887 38. The dnslists ACL condition has been extended: it is now possible to supply
888 a list of IP addresses and/or domains to be looked up in a particular DNS
891 39. Added log_selector=+queue_time_overall.
893 40. When running the queue in the test harness, wait just a tad after forking a
894 delivery process, to get repeatability of debugging output.
896 41. Include certificate and key file names in error message when GnuTLS fails
897 to set them up, because the GnuTLS error message doesn't include the name
898 of the failing file when there is a problem reading it.
900 42. Allow both -bf and -bF in the same test run.
902 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
904 44. The "Exiscan patch" is now merged into the mainline Exim source.
906 45. Sometimes the final signoff response after QUIT could fail to get
907 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
908 < 0 before doing a fflush(). This bug looks as though it goes back to the
909 introduction of TLS in release 3.20, but "sometimes" must have been rare
910 because the tests only now provoked it.
912 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
913 (this can affect the format of dates).
915 47. exim_tidydb, when checking for the continued existence of a message for
916 which it has found a message-specific retry record, was not finding
917 messages that were in split spool directories. Consequently, it was
918 deleting retry records that should have stayed in existence.
920 48. Steve fixed some bugs in eximstats.
922 49. The SPA authentication driver was not abandoning authentication and moving
923 on to the next authenticator when an expansion was forced to fail,
924 contradicting the general specification for all authenticators. Instead it
925 was generating a temporary error. It now behaves as specified.
927 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
928 (the order specifies the preference for clients). The order is now AES256,
929 AES128, 3DES, ARCFOUR128.
931 51. Small patch to Sieve code - explicitly set From: when generating an
934 52. Exim crashed if a remote delivery caused a very long error message to be
935 recorded - for instance if somebody sent an entire SpamAssassin report back
936 as a large number of 550 error lines. This bug was coincidentally fixed by
937 increasing the size of one of Exim's internal buffers (big_buffer) that
938 happened as part of the Exiscan merge. However, to be on the safe side, I
939 have made the code more robust (and fixed the comments that describe what
942 53. Now that there can be additional text after "Completed" in log lines (if
943 the queue_time_overall log selector is set), a one-byte patch to exigrep
944 was needed to allow it to recognize "Completed" as not the last thing in
947 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
948 patch that reportedly fixes this has been added. I am not expert enough to
949 create a test for it. This is what the patch creator wrote:
951 "I found a little strange behaviour of ldap code when working with
952 Windows 2003 AD Domain, where users was placed in more than one
953 Organization Units. When I tried to give exim partial DN, the exit code
954 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
955 But simultaneously result of request was absolutely normal ldap result,
956 so I produce this patch..."
958 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
959 REFERENCE, so I have modified the code to exclude the patch when that macro
962 55. Some experimental protocols are using DNS PTR records for new purposes. The
963 keys for these records are domain names, not reversed IP addresses. The
964 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
965 leaves it alone. Component reversal etc. now happens only for IP addresses.
968 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
970 57. Double the size of the debug message buffer (to 2048) so that more of very
971 long debug lines gets shown.
973 58. The exicyclog utility now does better if the number of log files to keep
974 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
976 59. Two changes related to the smtp_active_hostname option:
978 (1) $smtp_active_hostname is now available as a variable.
979 (2) The default for smtp_banner uses $smtp_active_hostname instead
980 of $primary_hostname.
982 60. The host_aton() function is supposed to be passed a string that is known
983 to be a valid IP address. However, in the case of IPv6 addresses, it was
984 not checking this. This is a hostage to fortune. Exim now panics and dies
985 if the condition is not met. A case was found where this could be provoked
986 from a dnsdb PTR lookup with an IPv6 address that had more than 8
987 components; fortuitously, this particular loophole had already been fixed
988 by change 4.50/55 above.
990 If there are any other similar loopholes, the new check in host_aton()
991 itself should stop them being exploited. The report I received stated that
992 data on the command line could provoke the exploit when Exim was running as
993 exim, but did not say which command line option was involved. All I could
994 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
998 61. There was a buffer overflow vulnerability in the SPA authentication code
999 (which came originally from the Samba project). I have added a test to the
1000 spa_base64_to_bits() function which I hope fixes it.
1003 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1004 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1006 63. The daemon start-up calls getloadavg() while still root for those OS that
1007 need the first call to be done as root, but it missed one case: when
1008 deliver_queue_load_max is set with deliver_drop_privilege. This is
1009 necessary for the benefit of the queue runner, because there is no re-exec
1010 when deliver_drop_privilege is set.
1012 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1013 This has been fixed.
1015 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1016 was in use, was not putting the data itself into the right store pool;
1017 consequently, it could be overwritten for a subsequent message in the same
1018 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1021 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1023 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1024 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1025 the function were treating the return as a boolean value, which happened to
1026 work because 0=false and not-0=true, but is not correct code.
1028 68. The host_aton() function was not handling scoped IPv6 addresses (those
1029 with, for example, "%eth0" on the end) correctly.
1031 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1032 negated items (that is, ~something) in unsigned ints. Some compilers
1033 apparently mutter when there is no cast.
1035 70. If an address verification called from an ACL failed, and did not produce a
1036 user-specific message (i.e. there was only a "system" message), nothing was
1037 put in $acl_verify_message. In this situation, it now puts the system
1040 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1041 session; change 4.31/43 added the unwanted input to the log line - except
1042 that it did not do this in the start of session case. It now does.
1044 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1045 This is wrong and can cause the other end to generate a synchronization
1046 error if it is another Exim or anything else that does the synchronization
1047 check. A QUIT command is no longer sent after a timeout.
1049 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1050 during host lookups.
1052 74. The defer_ok option of callout verification was not working if it was used
1053 when verifying addresses in header lines, that is, for this case:
1055 verify = header_sender/callout=defer_ok
1057 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1058 those file descriptors could be used for SMTP connections. If anything
1059 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1060 could be sent to the SMTP client, causing chaos. The daemon now opens
1061 stdin, stdout, and stderr to /dev/null when it puts itself into the
1064 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1065 log by default. The user can override this with suitable Perl magic.
1067 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1068 the log message when discard triggers, was not working for the DATA ACL or
1069 for the non-SMTP ACL.
1071 78. Error message wording change in sieve.c.
1073 79. If smtp_accept_max_per_host was set, the number of connections could be
1074 restricted to fewer than expected, because the daemon was trying to set up
1075 a new connection before checking whether the processes handling previous
1076 connections had finished. The check for completed processes is now done
1077 earlier. On busy systems, this bug wouldn't be noticed because something
1078 else would have woken the daemon, and it would have reaped the completed
1081 80. If a message was submitted locally by a user whose login name contained one
1082 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1083 It caused a spool format error. I have fixed the spool reading code. A
1084 related problem was that the "from" clause in the Received: line became
1085 illegal because of the space(s). It is now covered by ${quote_local_part.
1087 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1090 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1092 83. Patch from Sieve maintainer to fix unterminated string problem in
1093 "vacation" handling.
1095 84. Some minor changes to the Linux configuration files to help with other
1096 OS variants using glibc.
1098 85. One more patch for Sieve to update vacation handling to latest spec.
1101 ----------------------------------------------------
1102 See the note above about the 4.44 and 4.50 releases.
1103 ----------------------------------------------------
1109 1. Change 4.43/35 introduced a bug that caused file counts to be
1110 incorrectly computed when quota_filecount was set in an appendfile
1113 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1114 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1117 3. Give more explanation in the error message when the command for a transport
1118 filter fails to execute.
1120 4. There are several places where Exim runs a non-Exim command in a
1121 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1122 was being done only for the command run by the queryprogram router. It is
1123 now done for all such subprocesses. The other cases are: ${run, transport
1124 filters, and the commands run by the lmtp and pipe transports.
1126 5. Some older OS have a limit of 256 on the maximum number of file
1127 descriptors. Exim was using setrlimit() to set 1000 as a large value
1128 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1129 systems. I've change it so that if it can't get 1000, it tries for 256.
1131 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1132 was an oversight, and furthermore, ever since the addition of extra
1133 controls (e.g. 4.43/32), the checks on when to allow different forms of
1134 "control" were broken. There should now be diagnostics for all cases when a
1135 control that does not make sense is encountered.
1137 7. $recipients is now available in the predata ACL (oversight).
1139 8. Tidy the search cache before the fork to do a delivery from a message
1140 received from the command line. Otherwise the child will trigger a lookup
1141 failure and thereby defer the delivery if it tries to use (for example) a
1142 cached ldap connection that the parent has called unbind on.
1144 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1145 of $address_data from the recipient verification was clobbered by the
1146 sender verification.
1148 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1149 was its contents. (It was OK if the option was not defined at all.)
1151 11. A "Completed" log line is now written for messages that are removed from
1152 the spool by the -Mrm option.
1154 12. $host_address is now set to the target address during the checking of
1155 ignore_target_hosts.
1157 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1158 being passed; this would have caused $sender_host_name to have been used if
1159 matching the list had actually called for a host name (not very likely,
1160 since this list is usually IP addresses). A host name is now passed as
1163 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1164 code that specifies a non-restarting handler (typically sigaction() in
1165 modern systems) in an attempt to fix a rare and obscure crash bug.
1167 15. Narrowed the window for a race in the daemon that could cause it to ignore
1168 SIGCHLD signals. This is not a major problem, because they are used only to
1169 wake it up if nothing else does.
1171 16. A malformed maildirsize file could cause Exim to calculate negative values
1172 for the mailbox size or file count. Odd effects could occur as a result.
1173 The maildirsize information is now recalculated if the size or filecount
1176 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1177 support for a long time. Removed HAVE_SYS_VFS_H.
1179 18. Updated exipick to current release from John Jetmore.
1181 19. Allow an empty sender to be matched against a lookup in an address list.
1182 Previously the only cases considered were a regular expression, or an
1185 20. Exim went into a mad DNS lookup loop when doing a callout where the
1186 host was specified on the transport, if the DNS lookup yielded more than
1189 21. The RFC2047 encoding function was originally intended for short strings
1190 such as real names; it was not keeping to the 75-character limit for
1191 encoded words that the RFC imposes. It now respects the limit, and
1192 generates multiple encoded words if necessary. To be on the safe side, I
1193 have increased the buffer size for the ${rfc2047: expansion operator from
1196 22. Failure to deliver a bounce message always caused it to be frozen, even if
1197 there was an errors_to setting on the router. The errors_to setting is now
1200 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1201 canonical form (fully expanded) before being placed in
1202 $sender_host_address.
1204 24. Updated eximstats to version 1.33
1206 25. Include certificate and key file names in error message when GnuTLS fails
1207 to set them up, because the GnuTLS error message doesn't include the name
1208 of the failing file when there is a problem reading it.
1210 26. Expand error message when OpenSSL has problems setting up cert/key files.
1213 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1214 (this can affect the format of dates).
1216 28. exim_tidydb, when checking for the continued existence of a message for
1217 which it has found a message-specific retry record, was not finding
1218 messages that were in split spool directories. Consequently, it was
1219 deleting retry records that should have stayed in existence.
1221 29. eximstats updated to version 1.35
1222 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1223 1.35 - bugfix such that pie charts by volume are generated correctly
1225 30. The SPA authentication driver was not abandoning authentication and moving
1226 on to the next authenticator when an expansion was forced to fail,
1227 contradicting the general specification for all authenticators. Instead it
1228 was generating a temporary error. It now behaves as specified.
1230 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1231 (the order specifies the preference for clients). The order is now AES256,
1232 AES128, 3DES, ARCFOUR128.
1234 31. Small patch to Sieve code - explicitly set From: when generating an
1237 32. Exim crashed if a remote delivery caused a very long error message to be
1238 recorded - for instance if somebody sent an entire SpamAssassin report back
1239 as a large number of 550 error lines. This bug was coincidentally fixed by
1240 increasing the size of one of Exim's internal buffers (big_buffer) that
1241 happened as part of the Exiscan merge. However, to be on the safe side, I
1242 have made the code more robust (and fixed the comments that describe what
1245 33. Some experimental protocols are using DNS PTR records for new purposes. The
1246 keys for these records are domain names, not reversed IP addresses. The
1247 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1248 leaves it alone. Component reversal etc. now happens only for IP addresses.
1251 34. The host_aton() function is supposed to be passed a string that is known
1252 to be a valid IP address. However, in the case of IPv6 addresses, it was
1253 not checking this. This is a hostage to fortune. Exim now panics and dies
1254 if the condition is not met. A case was found where this could be provoked
1255 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1256 components; fortuitously, this particular loophole had already been fixed
1257 by change 4.50/55 or 4.44/33 above.
1259 If there are any other similar loopholes, the new check in host_aton()
1260 itself should stop them being exploited. The report I received stated that
1261 data on the command line could provoke the exploit when Exim was running as
1262 exim, but did not say which command line option was involved. All I could
1263 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1264 running as the user.
1267 35. There was a buffer overflow vulnerability in the SPA authentication code
1268 (which came originally from the Samba project). I have added a test to the
1269 spa_base64_to_bits() function which I hope fixes it.
1272 36. The daemon start-up calls getloadavg() while still root for those OS that
1273 need the first call to be done as root, but it missed one case: when
1274 deliver_queue_load_max is set with deliver_drop_privilege. This is
1275 necessary for the benefit of the queue runner, because there is no re-exec
1276 when deliver_drop_privilege is set.
1278 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1279 was in use, was not putting the data itself into the right store pool;
1280 consequently, it could be overwritten for a subsequent message in the same
1281 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1284 38. Sometimes the final signoff response after QUIT could fail to get
1285 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1286 < 0 before doing a fflush(). This bug looks as though it goes back to the
1287 introduction of TLS in release 3.20, but "sometimes" must have been rare
1288 because the tests only now provoked it.
1294 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1295 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1296 Now it returns an integer. A number of calls were still expecting a T/F
1297 return. Fortuitously, in all cases, the tests worked in OK situations,
1298 which is the norm. However, things would have gone wrong on any write
1299 failures on the smtp file descriptor. This function is used when sending
1300 messages over SMTP and also when doing verify callouts.
1302 2. When Exim is called to do synchronous delivery of a locally submitted
1303 message (the -odf or -odi options), it no longer closes stderr before doing
1306 3. Implemented the mua_wrapper option.
1308 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1310 5. Implemented the functions header_remove(), header_testname(),
1311 header_add_at_position(), and receive_remove_recipient(), and exported them
1314 6. If an ACL "warn" statement specified the addition of headers, Exim already
1315 inserted X-ACL-Warn: at the start if there was no header name. However, it
1316 was not making this test for the second and subsequent header lines if
1317 there were newlines in the string. This meant that an invalid header could
1318 be inserted if Exim was badly configured.
1320 7. Allow an ACL "warn" statement to add header lines at the start or after all
1321 the Received: headers, as well as at the end.
1323 8. Added the rcpt_4xx retry error code.
1325 9. Added postmaster_mailfrom=xxx to callout verification option.
1327 10. Added mailfrom=xxxx to the callout verification option, for verify=
1330 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1331 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1333 12. Inserted some casts to stop certain compilers warning when using pointer
1334 differences as field lengths or precisions in printf-type calls (mostly
1335 affecting debugging statements).
1337 13. Added optional readline() support for -be (dynamically loaded).
1339 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1340 same clock tick as a message's arrival, so that its received time was the
1341 same as the "first fail" time on the retry record, and that message
1342 remained on the queue past the ultimate address timeout, every queue runner
1343 would try a delivery (because it was past the ultimate address timeout) but
1344 after another failure, the ultimate address timeout, which should have then
1345 bounced the address, did not kick in. This was a "< instead of <=" error;
1346 in most cases the first failure would have been in the next clock tick
1347 after the received time, and all would be well.
1349 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1350 being recognized when the domain list was tested by the match_domain
1351 condition in an expansion string.
1353 16. Added the ${str2b64: operator.
1355 17. Exim was always calling setrlimit() to set a large limit for the number of
1356 processes, without checking whether the existing limit was already
1357 adequate. (It did check for the limit on file descriptors.) Furthermore,
1358 errors from getrlimit() and setrlimit() were being ignored. Now they are
1359 logged to the main and panic logs, but Exim does carry on, to try to do its
1360 job under whatever limits there are.
1362 18. Imported PCRE 5.0.
1364 19. Trivial typo in log message " temporarily refused connection" (the leading
1367 20. If the log selector return_path_on_delivery was set and an address was
1368 redirected to /dev/null, the delivery process crashed because it assumed
1369 that a return path would always be set for a "successful" delivery. In this
1370 case, the whole delivery is bypassed as an optimization, and therefore no
1373 21. Internal re-arrangement: the function for sending a challenge and reading
1374 a response while authentication was assuming a zero-terminated challenge
1375 string. It's now changed to take a pointer and a length, to allow for
1376 binary data in such strings.
1378 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1380 23. Exim was not respecting finduser_retries when seeking the login of the
1381 uid under which it was called; it was always trying 10 times. (The default
1382 setting of finduser_retries is zero.) Also, it was sleeping after the final
1383 failure, which is pointless.
1385 24. Implemented tls_on_connect_ports.
1387 25. Implemented acl_smtp_predata.
1389 26. If the domain in control=submission is set empty, Exim assumes that the
1390 authenticated id is a complete email address when it generates From: or
1391 Sender: header lines.
1393 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1394 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1395 chown and chgrp in /bin and hostname in /usr/bin.
1397 28. Exim was keeping the "process log" file open after each use, just as it
1398 does for the main log. This opens the possibility of it remaining open for
1399 long periods when the USR1 signal hits a daemon. Occasional processlog
1400 errors were reported, that could have been caused by this. Anyway, it seems
1401 much more sensible not to leave this file open at all, so that is what now
1404 29. The long-running daemon process does not normally write to the log once it
1405 has entered its main loop, and it closes the log before doing so. This is
1406 so that log files can straightforwardly be renamed and moved. However,
1407 there are a couple of unusual error situations where the daemon does write
1408 log entries, and I had neglected to close the log afterwards.
1410 30. The text of an SMTP error response that was received during a remote
1411 delivery was being truncated at 512 bytes. This is too short for some of
1412 the long messages that one sometimes sees. I've increased the limit to
1415 31. It is now possible to make retry rules that apply only when a message has a
1416 specific sender, in particular, an empty sender.
1418 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1419 it possible to be selective about when SMTP synchronization is enforced.
1421 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1423 32. Implemented hosts_connection_nolog.
1425 33. Added an ACL for QUIT.
1427 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1430 35. Added mailbox_size and mailbox_filecount to appendfile.
1432 36. Added control = no_multiline_responses to ACLs.
1434 37. There was a bug in the logic of the code that waits for the clock to tick
1435 in the case where the clock went backwards by a substantial amount such
1436 that the microsecond fraction of "now" was more than the microsecond
1437 fraction of "then" (but the whole seconds number was less).
1439 38. Added support for the libradius Radius client library this is found on
1440 FreeBSD (previously only the radiusclient library was supported).
1446 1. When certain lookups returned multiple values in the form name=value, the
1447 quoting of the values was not always being done properly. Specifically:
1448 (a) If the value started with a double quote, but contained no whitespace,
1450 (b) If the value contained whitespace other than a space character (i.e.
1451 tabs or newlines or carriage returns) it was not quoted.
1452 This fix has been applied to the mysql and pgsql lookups by writing a
1453 separate quoting function and calling it from the lookup code. The fix
1454 should probably also be applied to nisplus, ibase and oracle lookups, but
1455 since I cannot test any of those, I have not disturbed their existing code.
1457 2. A hit in the callout cache for a specific address caused a log line with no
1458 reason for rejecting RCPT. Now it says "Previous (cached) callout
1459 verification failure".
1461 3. There was an off-by-one bug in the queryprogram router. An over-long
1462 return line was truncated at 256 instead of 255 characters, thereby
1463 overflowing its buffer with the terminating zero. As well as fixing this, I
1464 have increased the buffer size to 1024 (and made a note to document this).
1466 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1467 when Exim is waiting for an SMTP response from a remote server, Exim
1468 restarts its select() call on the socket, thereby resetting its timeout.
1469 This is not a problem when such interrupts are rare. Somebody set up a cron
1470 job to run exiwhat every 2 minutes, which is less than the normal select()
1471 timeout (5 or 10 minutes). This meant that the select() timeout never
1472 kicked in because it was always reset. I have fixed this by comparing the
1473 time when an interrupt arrives with the time at the start of the first call
1474 to select(). If more time than the timeout has elapsed, the interrupt is
1475 treated as a timeout.
1477 5. Some internal re-factoring in preparation for the addition of Sieve
1478 extensions (by MH). In particular, the "personal" test is moved to a
1479 separate function, and given an option for scanning Cc: and Bcc: (which is
1480 not set for Exim filters).
1482 6. When Exim created an email address using the login of the caller as the
1483 local part (e.g. when creating a From: or Sender: header line), it was not
1484 quoting the local part when it contained special characters such as @.
1486 7. Installed new OpenBSD configuration files.
1488 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1489 try to make them clearer.
1491 9. Callout options, other than the timeout value, were being ignored when
1492 verifying sender addresses in header lines. For example, when using
1494 verify = header_sender/callout=no_cache
1496 the cache was (incorrectly) being used.
1498 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1499 only the Cygwin environment.
1501 11. When return_path_on_delivery was set as a log selector, if different remote
1502 addresses in the same message used different return paths and parallel
1503 remote delivery occurred, the wrong values would sometimes be logged.
1504 (Whenever a remote delivery process finished, the return path value from
1505 the most recently started remote delivery process was logged.)
1507 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1508 lines when AUTH and/or TLS are in use. This is the "received protocol"
1509 field. Exim used to use "asmtp" for authenticated SMTP, without any
1510 indication (in the protocol name) for TLS use. Now it follows the RFC and
1511 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1512 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1513 names appear in log lines as well as in Received: header lines.
1515 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1516 extensions, and comparison tests, and to fix some bugs.
1518 14. Changes to the "personal" filter test:
1520 (1) The test was buggy in that it was just doing the equivalent of
1521 "contains" tests on header lines. For example, if a user's address was
1522 anne@some.where, the "personal" test would incorrectly be true for
1524 To: susanne@some.where
1526 This test is now done by extracting each address from the header in turn,
1527 and checking the entire address. Other tests that are part of "personal"
1528 are now done using regular expressions (for example, to check local parts
1529 of addresses in From: header lines).
1531 (2) The list of non-personal local parts in From: addresses has been
1532 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1533 taken from the Sieve specification recommendations.
1535 (3) If the message contains any header line starting with "List-" it is
1536 treated as non-personal.
1538 (4) The test for "circular" in the Subject: header line has been removed
1539 because it now seems ill-conceived.
1541 15. Minor typos in src/EDITME comments corrected.
1543 16. Installed latest exipick from John Jetmore.
1545 17. If headers_add on a router specified a text string that was too long for
1546 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1547 of string_sprintf() is now avoided.
1549 18. $message_body_size was not set (it was always zero) when running the DATA
1550 ACL and the local_scan() function.
1552 19. For the "mail" command in an Exim filter, no default was being set for
1553 the once_repeat time, causing a random time value to be used if "once" was
1554 specified. (If the value happened to be <= 0, no repeat happened.) The
1555 default is now 0s, meaning "never repeat". The "vacation" command was OK
1556 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1557 (I found it when inspecting the code).
1559 20. There is now an overall timeout for performing a callout verification. It
1560 defaults to 4 times the callout timeout, which applies to individual SMTP
1561 commands during the callout. The overall timeout applies when there is more
1562 than one host that can be tried. The timeout is checked before trying the
1563 next host. This prevents very long delays if there are a large number of
1564 hosts and all are timing out (e.g. when the network connections are timing
1565 out). The value of the overall timeout can be changed by specifying an
1566 additional sub-option for "callout", called "maxwait". For example:
1568 verify = sender/callout=5s,maxwait=20s
1570 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1571 to the end before writing, but this should make it even safer).
1573 22. Exim was forgetting that it had advertised PIPELINING for the second and
1574 subsequent messages on an SMTP connection. It was also not resetting its
1575 memory on STARTTLS and an internal HELO.
1577 23. When Exim logs an SMTP synchronization error within a session, it now
1578 records whether PIPELINING has been advertised or not.
1580 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1581 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1582 rather than long int.
1584 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1586 26. Added the never_mail option to autoreply.
1592 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1593 crash if the getsockname() call failed; this can happen if a connection is
1594 closed very soon after it is established. The problem was simply in the
1595 order in which certain operations were done, causing Exim to try to write
1596 to the SMTP stream before it had set up the file descriptor. The bug has
1597 been fixed by making things happen in the correct order.
1603 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1604 before the connection was closed, thus losing the rejection response.
1606 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1607 some early Solaris releases, but causes trouble in current releases where
1608 socklen_t is defined.
1610 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1613 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1616 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1617 that releasing the top of it at the end releases what was used for sub-
1618 expansions (unless the block got too big). However, discard this block if
1619 the first thing is a variable or header, so that we can use its block when
1620 it is dynamic (useful for very large $message_headers, for example).
1622 6. Lookups now cache *every* query, not just the most recent. A new, separate
1623 store pool is used for this. It can be recovered when all lookup caches are
1624 flushed. Lookups now release memory at the end of their result strings.
1625 This has involved some general refactoring of the lookup sources.
1627 7. Some code has been added to the store_xxx() functions to reduce the amount
1628 of flapping under certain conditions.
1630 8. log_incoming_interface used to affect only the <= reception log lines. Now
1631 it causes the local interface and port to be added to several more SMTP log
1632 lines, for example "SMTP connection from", and rejection lines.
1634 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1636 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1638 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1639 could be overwritten at the end of the current message (or the start of a
1640 new message if it was set in a HELO ACL). The value is now preserved for
1641 the duration of the SMTP connection.
1643 12. If a transport had a headers_rewrite setting, and a matching header line
1644 contained an unqualified address, that address was qualified, even if it
1645 did not match any rewriting rules. The underlying bug was that the values
1646 of the flags that permit the existence of unqualified sender and recipient
1647 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1648 non-local messages, and by -bnq for local messages) were not being
1649 preserved with the message after it was received.
1651 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1652 "next input=" as part of the text comprising the host identity instead of
1653 the correct text. The code was using the same buffer for two different
1654 strings. However, depending on which order the printing function evaluated
1655 its arguments, the bug did not always show up. Under Linux, for example, my
1656 test suite worked just fine.
1658 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1659 doesn't work with some older versions of Perl. It has been changed to "my",
1660 which in any case is probably the better facility to use.
1662 15. A really picky compiler found some instances of statements for creating
1663 error messages that either had too many or two few arguments for the format
1666 16. The size of the buffer for calls to the DNS resolver has been increased
1667 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1668 for addresses that have a lot of PTR records. This alleviates a problem; it
1669 does not fully solve it.
1671 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1672 buffer now truncates the list and logs the incident, which is the same
1673 action as happens when Exim is looking up a host name and its aliases.
1674 Previously in this situation something unpredictable would happen;
1675 sometimes it was "internal error: store_reset failed".
1677 18. If a server dropped the connection unexpectedly when an Exim client was
1678 using GnuTLS and trying to read a response, the client delivery process
1679 crashed while trying to generate an error log message.
1681 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1682 string, for example:
1684 warn message = H1: something\nH2: something
1686 the text was added as a single header line from Exim's point of view
1687 though it ended up OK in the delivered message. However, searching for the
1688 second and subsequent header lines using $h_h2: did not work. This has been
1689 fixed. Similarly, if a system filter added multiple headers in this way,
1690 the routers could not see them.
1692 20. Expanded the error message when iplsearch is called with an invalid key to
1693 suggest using net-iplsearch in a host list.
1695 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1696 ACLs are no longer actually imposed (and a message to that effect is
1699 22. If a "gecos" field in a passwd entry contained escaped characters, in
1700 particular, if it contained a \" sequence, Exim got it wrong when building
1701 a From: or a Sender: header from that name. A second bug also caused
1702 incorrect handling when an unquoted " was present following a character
1703 that needed quoting.
1705 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1706 was not being matched caselessly.
1708 24. Arranged for all hyphens in the exim.8 source to be escaped with
1711 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1712 properly. Recipient callout cache records were still being keyed to include
1713 the sender, even when use_sender was set false. This led to far more
1714 callouts that were necessary. The sender is no longer included in the key
1715 when use_sender is false.
1717 26. Added "control = submission" modifier to ACLs.
1719 27. Added the ${base62d: operator to decode base 62 numbers.
1721 28. dnsdb lookups can now access SRV records.
1723 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1724 the configuration file.
1726 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1727 (-v) mode. This makes the output for a verbose queue run more intelligible.
1729 31. Added a use_postmaster feature to recipient callouts.
1731 32. Added the $body_zerocount variable, containing the number of binary zero
1732 bytes in the message body.
1734 33. The time of last modification of the "new" subdirectory is now used as the
1735 "mailbox time last read" when there is a quota error for a maildir
1738 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1740 35. Added +ignore_unknown as a special item in host lists.
1742 36. Code for decoding IPv6 addresses in host lists is now included, even if
1743 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1744 address was recognized as an IP address, but was then not correctly decoded
1745 into binary, causing unexpected and incorrect effects when compared with
1752 1. Very minor rewording of debugging text in manualroute to say "list of
1753 hosts" instead of "hostlist".
1755 2. If verify=header_syntax was set, and a header line with an unqualified
1756 address (no domain) and a large number of spaces between the end of the
1757 name and the colon was received, the reception process suffered a buffer
1758 overflow, and (when I tested it) crashed. This was caused by some obsolete
1759 code that should have been removed. The fix is to remove it!
1761 3. When running in the test harness, delay a bit after writing a bounce
1762 message to get a bit more predictability in the log output.
1764 4. Added a call to search_tidyup() just before forking a reception process. In
1765 theory, someone could use a lookup in the expansion of smtp_accept_max_
1766 per_host which, without the tidyup, could leave open a database connection.
1768 5. Added the variables $recipient_data and $sender_data which get set from a
1769 lookup success in an ACL "recipients" or "senders" condition, or a router
1770 "senders" option, similar to $domain_data and $local_part_data.
1772 6. Moved the writing of debug_print from before to after the "senders" test
1775 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1776 problems for message scanning, either using a data ACL, or using
1777 local_scan() because the Received: header was not generated till after they
1778 were called (in order to set the time as the time of reception completion).
1779 I have revised the way this works. The header is now generated after the
1780 body is received, but before the ACL or local_scan() are called. After they
1781 are run, the timestamp in the header is updated.
1787 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1788 before starting a queue runner without re-exec. This happened only when
1789 deliver_drop_privilege was set or when the Exim user was set to root. The
1790 effect of the bug was that timeouts during subsequent deliveries caused
1791 crashes instead of being properly handled. The handler is now left at its
1792 default (and expected) setting.
1794 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1795 message, again when deliver_drop_privilege is set or Exim is run as root.
1796 The bug described in (1) was not present in this case, but the tidying up
1797 of the other signals was missing. I have made the two cases consistent.
1799 3. The ignore_target_hosts setting on a manualroute router was being ignored
1800 for hosts that were looked up using the /MX notation.
1802 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1805 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1806 operated on the sender address. After changing the $sender_address to <>
1807 for the sender address verify, Exim was re-instated it as the original
1808 (before rewriting) address, but remembering that it had rewritten it, so it
1809 wasn't rewriting it again. This bug also had the effect of breaking the
1810 sender address verification caching when the sender address was rewritten.
1812 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1813 This has been changed so that if the ip literal address matches
1814 ignore_target_hosts, the router declines.
1816 7. Added expansion conditions match_domain, match_address, and match_local_
1817 part (NOT match_host).
1819 8. The placeholder for the Received: header didn't have a length field set.
1821 9. Added code to Exim itself and to exim_lock to test for a specific race
1822 condition that could lead to file corruption when using MBX delivery. The
1823 issue is with the lockfile that is created in /tmp. If this file is removed
1824 after a process has opened it but before that process has acquired a lock,
1825 there is the potential for a second process to recreate the file and also
1826 acquire a lock. This could lead to two Exim processes writing to the file
1827 at the same time. The added code performs the same test as UW imapd; it
1828 checks after acquiring the lock that its file descriptor still refers to
1829 the same named file.
1831 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1832 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1835 11. Added the smtp_active_hostname option. If used, this will typically be made
1836 to depend on the incoming interface address. Because $interface_address is
1837 not set up until the daemon has forked a reception process, error responses
1838 that can happen earlier (such as "too many connections") no longer contain
1841 12. If an expansion in a condition on a "warn" statement fails because a lookup
1842 defers, the "warn" statement is abandoned, and the next ACL statement is
1843 processed. Previously this caused the whole ACL to be aborted.
1845 13. Added the iplsearch lookup type.
1847 14. Added ident_timeout as a log selector.
1849 15. Added tls_certificate_verified as a log selector.
1851 16. Added a global option tls_require_ciphers (compare the smtp transport
1852 option of the same name). This controls incoming TLS connections.
1854 17. I finally figured out how to make tls_require_ciphers do a similar thing
1855 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1856 before starting the TLS session.
1858 18. Tabs are now shown as \t in -bP output.
1860 19. If the log selector return_path_on_delivery was set, Exim crashed when
1861 bouncing a message because it had too many Received: header lines.
1863 20. If two routers both had headers_remove settings, and the first one included
1864 a superfluous trailing colon, the final name in the first list and the
1865 first name in the second list were incorrectly joined into one item (with a
1866 colon in the middle).
1872 1. Added -C and -D options to the exinext utility, mainly to make it easier
1873 to include in the automated testing, but these could be helpful when
1874 multiple configurations are in use.
1876 2. The exinext utility was not formatting the output nicely when there was
1877 an alternate port involved in the retry record key, nor when there was a
1878 message id as well (for retries that were specific to a specific message
1879 and a specific host). It was also confused by IPv6 addresses, because of
1880 the additional colons they contain. I have fixed the IPv4 problem, and
1881 patched it up to do a reasonable job for IPv6.
1883 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1884 delivery, the log line now contains "pipelined" if PIPELINING was used.
1886 4. An SMTP transport process used to panic and die if the bind() call to set
1887 an explicit outgoing interface failed. This has been changed; it is now
1888 treated in the same way as a connect() failure.
1890 5. A reference to $sender_host_name in the part of a conditional expansion
1891 that was being skipped was still causing a DNS lookup. This no longer
1894 6. The def: expansion condition was not recognizing references to header lines
1895 that used bh_ and bheader_.
1897 7. Added the _cache feature to named lists.
1899 8. The code for checking quota_filecount in the appendfile transport was
1900 allowing one more file than it should have been.
1902 9. For compatibility with Sendmail, the command line option
1910 and sets the incoming protocol and host name (for trusted callers). The
1911 host name and its colon can be omitted when only the protocol is to be set.
1912 Note the Exim already has two private options, -pd and -ps, that refer to
1913 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1914 "s", but I don't think that's a major issue.
1916 10. A number of refactoring changes to the code, none of which should affect
1919 (a) The number of logging options was getting close to filling up the
1920 32-bit word that was used as a bit map. I have split them into two classes:
1921 those that are passed in the argument to log_write(), and those that are
1922 only ever tested independently outside of that function. These are now in
1923 separate 32-bit words, so there is plenty of room for expansion again.
1924 There is no change in the user interface or the logging behaviour.
1926 (b) When building, for example, log lines, the code previously used a
1927 macro that called string_cat() twice, in order to add two strings. This is
1928 not really sufficiently general. Furthermore, there was one instance where
1929 it was actually wrong because one of the argument was used twice, and in
1930 one call a function was used. (As it happened, calling the function twice
1931 did not affect the overall behaviour.) The macro has been replaced by a
1932 function that can join an arbitrary number of extra strings onto a growing
1935 (c) The code for expansion conditions now uses a table and a binary chop
1936 instead of a serial search (which was left over from when there were very
1937 few conditions). Also, it now recognizes conditions like "pam" even when
1938 the relevant support is not compiled in: a suitably worded error message is
1939 given if an attempt is made to use such a condition.
1941 11. Added ${time_interval:xxxxx}.
1943 12. A bug was causing one of the ddress fields not to be passed back correctly
1944 from remote delivery subprocesses. The field in question was not being
1945 subsequently used, so this caused to problems in practice.
1947 13. Added new log selectors queue_time and deliver_time.
1949 14. Might have fixed a bug in maildirsizefile handling that threw up
1950 "unexpected character" debug warnings, and recalculated the data
1951 unnecessarily. In any case, I expanded the warning message to give more
1954 15. Added the message "Restricted characters in address" to the statements in
1955 the default ACL that block characters like @ and % in local parts.
1957 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1958 Three changes have been made:
1960 (a) There was a serious bug; a negative response to MAIL caused the whole
1961 recipient domain to be cached as invalid, thereby blocking all messages
1962 to all local parts at the same domain, from all senders. This bug has
1963 been fixed. The domain is no longer cached after a negative response to
1964 MAIL if the sender used is not empty.
1966 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1969 (c) A new callout option, "use_sender" has been added for people who want
1970 the modified behaviour.
1976 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1979 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1980 indeed breaks things for older releases.
1982 3. Added additional logging to the case where there is a problem reading data
1983 from a filter that is running in a subprocess using a pipe, in order to
1984 try to track down a specific problem.
1986 4. Testing facility fudge: when running in the test harness and attempting
1987 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1988 getting "No route to host". Convert this to a timeout.
1990 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
1993 6. Some OS don't have socklen_t but use size_t instead. This affects the
1994 fifth argument of getsockopt() amongst other things. This is now
1995 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
1996 can be set for individual OS. I have set it for SunOS5, OSF1, and
1997 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
1998 some earlier ones do not.
2000 7. Change 4.30/15 was not doing the test caselessly.
2002 8. The standard form for an IPv6 address literal was being rejected by address
2003 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2004 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2005 this, as well as the form without the "IPv6" on the front (but only when
2006 address literals are enabled, of course).
2008 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2010 10. Exim crashed if a message with an empty sender address specified by -f
2011 encountered a router with an errors_to setting. This could be provoked only
2012 by a command such as
2016 where an empty string was supplied; "<>" did not hit this bug.
2018 11. Installed PCRE release 4.5.
2020 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2021 remained set. It is now erased.
2023 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2024 times from message ids (which are base 36 rather than the normal 62).
2026 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2027 were being counted as actual protocol errors, and logged if the log
2028 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2029 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2030 and DATA following a set of rejected RCPTs do not count as protocol errors.
2031 In other words, Exim assumes they were pipelined, though this may not
2032 actually be the case. Of course, in all cases the client gets an
2033 appropriate error code.
2035 15. If a lookup fails in an ACL condition, a message about the failure may
2036 be available; it is used if testing the ACL cannot continue, because most
2037 such messages specify what the cause of the deferral is. However, some
2038 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2039 that caused an old message to be retained and used if a later statement
2040 caused a defer, replacing the real cause of the deferral.
2042 16. If an IP address had so many PTR records that the DNS lookup buffer
2043 was not large enough to hold them, Exim could crash while trying to process
2044 the truncated data. It now detects and logs this case.
2046 17. Further to 4.21/58, another change has been made: if (and only if) the
2047 first line of a message (the first header line) ends with CRLF, a bare LF
2048 in a subsequent header line has a space inserted after it, so as not to
2049 terminate the header.
2051 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2052 unnecessarily, used atoi() instead of strtol(), and didn't check the
2053 termination when getting file sizes from file names by regex.
2055 19. Completely re-implemented the support for maildirsize files, in the light
2056 of a number of problems with the previous contributed implementation
2057 (4.30/29). In particular:
2059 . If the quota is zero, the maildirsize file is maintained, but no quota is
2062 . If the maildir directory does not exist, it is created before any attempt
2063 to write a maildirsize file.
2065 . The quota value in the file is just a cache; if the quota is changed in
2066 the transport, the new value overrides.
2068 . A regular expression is available for excluding directories from the
2071 20. The autoreply transport checks the characters in options that define the
2072 message's headers; it allows continued headers, but it was checking with
2073 isspace() after an embedded newline instead of explicitly looking for a
2076 21. If all the "regular" hosts to which an address was routed had passed their
2077 expiry times, and had not reached their retry times, the address was
2078 bounced, even if fallback hosts were defined. Now Exim should go on to try
2081 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2082 equivalent code in the SMTP transport. Some hosts send humungous responses
2083 to HELO/EHLO, more than 1024 it seems.
2085 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2086 gives compiler warnings in some environments. I've now done it "properly",
2089 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2090 (because of problems with the built-in one) was declared to return uschar *
2091 instead of char *, causing compiler failure.
2093 25. Fixed a file descriptor leak when processing alias/forward files.
2095 26. Fixed a minor format string issue in dbfn.c.
2097 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2099 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2100 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2103 29. When a sender address is verified, it is cached, to save repeating the test
2104 when there is more than one recipient in a message. However, when the
2105 verification involves a callout, it is possible for different callout
2106 options to be set for different recipients. It is too complicated to keep
2107 track of this in the cache, so now Exim always runs a verification when a
2108 callout is required, relying on the callout cache for the optimization.
2109 The overhead is duplication of the address routing, but this should not be
2112 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2113 to be verified with callout=postmaster, and the main callout worked but the
2114 postmaster check failed, the verification correctly failed. However, if a
2115 subsequent RCPT command asked for sender verification *without* the
2116 postmaster check, incorrect caching caused this verification also to fail,
2119 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2120 it was not caching the DNS options (qualify_single, search_parents) that
2121 were used when the lookup failed. A subsequent lookup with different
2122 options therefore always gave the same answer, though there were cases
2123 where it should not have. (Example: a "domains = !$mx_any" option on a
2124 dnslookup router: the "domains" option is always processed without any
2125 widening, but the router might have qualify_single set.) Now Exim uses the
2126 cached value only when the same options are set.
2128 32. Added John Jetmore's "exipick" utility to the distribution.
2130 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2131 than a timeout (e.g. a certificate is required, and is not provided), an
2132 Exim server now closes the connection immediately. Previously it waited for
2133 the client to close - but if the client is SSL, it seems that they each
2134 wait for each other, leading to a delay before one of them times out.
2136 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2137 maintained 0.8.x compatibility because I don't think many are using it, and
2138 it is clearly obsolete.
2140 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2143 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2144 client certificate was expired. A simple patch fixes this, though I don't
2145 understand the full logic of why the verify callback is called multiple
2148 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2149 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2150 which causes problems with some clients (such as the Certicom SSL Plus
2151 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2152 disables the coutermeasure allowing Eudora to connect."
2154 38. Exim was not checking that a write() to a log file succeeded. This could
2155 lead to Bad Things if a log got too big, in particular if it hit a file
2156 size limit. Exim now panics and dies if it cannot write to a log file, just
2157 as it does if it cannot open a log file.
2159 39. Modified OS/Makefile-Linux so that it now contains
2161 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2163 The two -D definitions ensure that Exim is compiled with large file
2164 support, which makes it possible to handle log files that are bigger than
2167 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2168 instance) a domain was checked against a named list that involved a lookup,
2169 causing $domain_data to be set, then another domain was checked against the
2170 same list, then the first domain was re-checked, the value of $domain_data
2171 after the final check could be wrong. In particular, if the second check
2172 failed, it could be set empty. This bug probably also applied to
2175 41. The strip_trailing_dot option was not being applied to the address given
2176 with the -f command-line option.
2178 42. The code for reading a message's header from the spool was incrementing
2179 $received_count, but never initializing it. This meant that the value was
2180 incorrect (doubled) while delivering a message in the same process in which
2181 it was received. In the most common configuration of Exim, this never
2182 happens - a fresh exec is done - but it can happen when
2183 deliver_drop_privilege is set.
2185 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2186 it now includes up to 150 characters of the unexpected data in the log
2189 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2190 and building data strings. The size of both of these buffers was 10 000
2191 bytes - far larger than anybody would *ever* want, thought I. Needless to
2192 say, somebody hit the limit. I have increased the maximum line length to
2193 20 000 and the maximum data length of concatenated lines to 100 000. I have
2194 also fixed two bugs, because there was no checking on these buffers. Tsk,
2195 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2196 buffer is too small.
2198 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2199 lsearch lookups. Now it does.
2201 46. When parsing a route_list item in a manualroute router, a fixed-length
2202 buffer was used for the list of hosts. I made this 1024 bytes long,
2203 thinking that nobody would ever have a list of hosts that long. Wrong.
2204 Somebody had a whole pile of complicated expansion conditions, and the
2205 string was silently truncated, leading to an expansion error. It turns out
2206 that it is easier to change to an unlimited length (owing to other changes
2207 that have happened since this code was originally written) than to build
2208 structure for giving a limitation error. The length of the item that
2209 expands into the list of hosts is now unlimited.
2211 47. The lsearch lookup could not handle data where the length of text line was
2212 more than 4095 characters. Such lines were truncated, leading to shortened
2213 data being returned. It should now handle lines of any length.
2215 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2216 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2219 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2220 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2221 no longer happens in comment lines. A list of replacements is now placed
2222 at the head of all of the source files, except those whose only change is
2223 to replace PERL_COMMAND in the very first #! line.
2225 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2226 messages on the queue, with a bottom-up merge sort, using code contributed
2227 by Michael Haardt. This should make operations like -bp somewhat faster on
2228 large queues. It won't affect queue runners, except when queue_run_in_order
2231 51. Installed eximstats 1.31 in the distribution.
2233 52. Added support for SRV lookups to the dnslookup router.
2235 53. If an ACL referred to $message_body or $message_body_end, the value was not
2236 reset for any messages that followed in the same SMTP session.
2238 54. The store-handling optimization for building very long strings was not
2239 differentiating between the different store pools. I don't think this
2240 actually made any difference in practice, but I've tidied it.
2242 55. While running the routers to verify a sender address, $sender_address
2243 was still set to the sender address. This is wrong, because when routing to
2244 send a bounce to the sender, it would be empty. Therefore, I have changed
2245 it so that, while verifying a sender address, $sender_address is set to <>.
2246 (There is no change to what happens when verifying a recipient address.)
2248 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2249 target A or AAAA records (if not already returned) without resetting the
2250 qualify_single or search_parents options of the DNS resolver. These are
2251 inappropriate in this case because the targets of MX and SRV records must
2252 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2253 target that, when qualified, matched something in the local domain. These
2254 two options are now turned off when doing these lookups.
2256 57. It seems that at least some releases of Reiserfs (which does not have the
2257 concept of a fixed number of inodes) returns zero and not -1 for the
2258 number of available inodes. This interacted badly with check_spool_inodes,
2259 which assumed that -1 was the "no such thing" setting. What I have done is
2260 to check that the total number of inodes is greater than zero before doing
2261 the test of how many are available.
2263 58. When a "warn" ACL statement has a log_message modifier, the message is
2264 remembered, and not repeated. This is to avoid a lot of repetition when a
2265 message has many recipients that cause the same warning to be written.
2266 Howewer, Exim was preserving the list of already written lines for an
2267 entire SMTP session, which doesn't seem right. The memory is now reset if a
2268 new message is started.
2270 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2271 debugging output unless log_rewrite was also set.
2273 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2274 of (int)(handle) when we know that handle contains (void *)(-1).
2276 61. The Exim daemon panic-logs an error return when it closes the incoming
2277 connection. However "connection reset by peer" seems to be common, and
2278 isn't really an error worthy of noting specially, so that particular error
2281 62. When Exim is trying to find all the local interfaces, it used to panic and
2282 die if the ioctl to get the interface flags failed. However, it seems that
2283 on at least one OS (Solaris 9) it is possible to have an interface that is
2284 included in the list of interfaces, but for which you get a failure error
2285 for this call. This happens when the interface is not "plumbed" into a
2286 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2287 failure of the "get flags" call assumes that the interface is down.
2289 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2290 makes life easier for people who are doing arithmetic on fields extracted
2291 from dates, where you often get leading zeros that should not be
2292 interpreted as octal.
2294 64. Added qualify_domain to the redirect router, to override the global
2297 65. If a pathologically long header line contained very many addresses (the
2298 report of this problem mentioned 10 000) and each of them was rewritten,
2299 Exim could use up a very large amount of memory. (It kept on making new
2300 copies of the header line as it rewrote, and never released the old ones.)
2301 At the expense of a bit more processing, the header rewriting function has
2302 been changed so that it no longer eats memory in this way.
2304 66. The generation of the Received: header has been moved from the time that a
2305 message starts to be received, to the time that it finishes. The timestamp
2306 in the Received: header should now be very close to that of the <= log
2307 line. There are two side-effects of this change:
2309 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2310 logged header lines no longer include the local Received: line, because
2311 it has not yet been created. The same applies to a copy of the message
2312 that is returned to a non-SMTP sender when a message is rejected.
2314 (b) When a filter file is tested using -bf, no additional Received: header
2315 is added to the test message. After some thought, I decided that this
2318 This change does not affect the value of $received_for. It is still set
2319 after address rewriting, but before local_scan() is called.
2321 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2323 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2324 gave an unhelpful panic error message, and a defer error. I have managed to
2325 change this behaviour so that it now rejects any supplied certificate,
2326 which seems right, as the list of acceptable certificates is empty.
2328 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2329 gave an unhelpful defer error. I have not managed to make this reject any
2330 supplied certificates, but the error message it gives is "no certificate
2331 supplied", which is not helpful.
2333 70. exigrep's output now also includes lines that are not associated with any
2334 message, but which match the given pattern. Implemented by a patch from
2335 Martin Sluka, which also tidied up the Perl a bit.
2337 71. Recipient callout verification, like sender verification, was using <> in
2338 the MAIL FROM command. This isn't really the right thing, since the actual
2339 sender may affect whether the remote host accepts the recipient or not. I
2340 have changed it to use the actual sender in the callout; this means that
2341 the cache record is now keyed on a recipient/sender pair, not just the
2342 recipient address. There doesn't seem to be a real danger of callout loops,
2343 since a callout by the remote host to check the sender would use <>.
2344 [SEE ABOVE: changed after hitting problems.]
2346 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2347 temporary errors. However, in the case of such a code being given after
2348 the end of a data transmission (i.e. after ".") Exim was failing to write
2349 a retry record for the message. (Yes, there was some broken host that was
2350 actually sending 8xx at this point.)
2352 73. An unknown lookup type in a host list could cause Exim to panic-die when
2353 the list was checked. (An example that provoked this was putting <; in the
2354 middle of a list instead of at the start.) If this happened during a DATA
2355 ACL check, a -D file could be left lying around. This kind of configuration
2356 error no longer causes Exim to die; instead it causes a defer errror. The
2357 incident is still logged to the main and panic logs.
2359 74. Buglet left over from Exim 3 conversion. The message "too many messages
2360 in one connection" was written to the rejectlog but not the mainlog, except
2361 when address rewriting (yes!) was being logged.
2363 75. Added write_rejectlog option.
2365 76. When a system filter was run not as root (that is, when system_filter_user
2366 was set), the values of the $n variables were not being returned to the
2367 main process; thus, they were not subsequently available in the $sn
2370 77. Added +return_path_on_delivery log selector.
2372 78. A connection timeout was being treated differently from recipients deferred
2373 when testing hosts_max_try with a message that was older than the host's
2374 retry timeout. (The host should not be counted, thus allowing all hosts to
2375 be tried at least once before bouncing.) This may have been the cause of an
2376 occasionally reported bug whereby a message would remain on the queue
2377 longer than the retry timeout, but would be bounced if a delivery was
2378 forced. I say "may" because I never totally pinned down the problem;
2379 setting up timeout/retry tests is difficult. See also the next item.
2381 79. The ultimate address timeout was not being applied to errors that involved
2382 a combination of host plus message (for example, a timeout on a MAIL
2383 command). When an address resolved to a number of possible hosts, and they
2384 were not all tried for each delivery (e.g. because of hosts_max_try), a
2385 message could remain on the queue longer than the retry timeout.
2387 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2390 81. Fixed an obscure SMTP outgoing bug which required at least the following
2391 conditions: (a) there was another message waiting for the same server;
2392 (b) the server returned 5xx to all RCPT commands in the first message so
2393 that the message was not completed; (c) the server dropped the connection
2394 or gave a negative response to the RSET that Exim sends to abort the
2395 transaction. The observed case was a dropped connection after DATA that had
2396 been sent in pipelining mode. That is, the server had advertised PIPELINING
2397 but was not implementing it correctly. The effect of the bug was incorrect
2398 behaviour, such as trying another host, and this could lead to a crash.
2404 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2405 and daemon.c were passed as pointers to ints; they should have been
2406 pointers to socklen_t variables (which are typically unsigned ints).
2408 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2411 3. Fixed a really odd bug that affected only the testing scheme; patching a
2412 certain fixed string in the binary changed the value of another string that
2413 happened to be identical to the end of the original first string.
2415 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2416 name", it returns that address as the IP address. On some operating
2417 systems (e.g. Solaris), it also passes back the IP address string as the
2418 "host name". However, on others (e.g. Linux), it passes back an empty
2419 string. Exim wasn't checking for this, and was changing the host name to an
2420 empty string, assuming it had been canonicized.
2422 5. Although rare, it is permitted to have more than one PTR record for a given
2423 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2424 all the names associated with an address, because they do in Solaris.
2425 However, it seems that they do not in Linux for data that comes from the
2426 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2427 I found this out when I moved to a new Linux workstation and tried to run
2428 the Exim test suite.
2430 To get round this problem I have changed the code so that it now does its
2431 own call to the DNS to look up PTR records when searching for a host name.
2432 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2433 addresses that are only in /etc/hosts are still found.
2435 This behaviour is, however, controlled by an option called host_lookup_
2436 order, which defaults to "bydns:byaddr". If people want to use the other
2437 order, or indeed, just use one or the other means of lookup, they can
2438 specify it in this variable.
2440 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2441 some operating systems, this comes back from gethostbyaddr() as an empty
2442 string, and this is what Exim used to test for. However, it seems that in
2443 other systems, "." is yielded. Exim now tests for this case too.
2445 7. The values of check_spool_space and check_log_space are now held internally
2446 as a number of kilobytes instead of an absolute number of bytes. If a
2447 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2448 kilobyte. This means that much larger values can be stored.
2450 8. Exim monitor: an attempt to get the action menu when not actually pointing
2451 at a message produces an empty menu entitled "No message selected". This
2452 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2453 no entries in it ("Shell widget menu has zero width and/or height"). So I
2454 have added a single, blank menu entry in this case.
2456 9. Added ${quote_local_part.
2458 10. MIME decoding is now applied to the contents of Subject: header lines when
2461 11. Now that a reference to $sender_host_address automatically causes a reverse
2462 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2463 host lookup before query-style lookups in lists that might use this
2464 variable. This has therefore been abolished, and the "net-" prefix is no
2465 longer necessary for query-style lookups.
2467 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2468 have been a typo for LFLAGS, so it has been changed.
2470 13. The install script calls Exim with "-C /dev/null" in order to find the
2471 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2472 to be output. Howeve, since Exim outputs its version number before the
2473 error, it didn't break the script. It just looked ugly. I fixed this by
2474 always allowing "-C /dev/null" if the caller is root.
2476 14. Ignore overlarge ACL variable number when reading spool file - insurance
2477 against a later release with more variables having written the file.
2479 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2480 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2481 this, as well as the form without the "IPv6" on the front.
2483 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2484 OS/Makefile-Darwin file.
2486 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2487 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2489 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2490 variable. However, some OS use ints for inodes. I've added cast to long int
2491 to get rid of the compiler warning.
2493 19. I had forgotten to lock out "/../" in configuration file names when
2494 ALT_CONFIG_PREFIX was set.
2496 20. Routers used for verification do not need to specify transports. However,
2497 if such a router generated a host list, and callout was configured, Exim
2498 crashed, because it could not find a port number from the (non-existent)
2499 transport. It now assumes port 25 in this circumstance.
2501 21. Added the -t option to exigrep.
2503 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2504 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2505 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2506 them is set, LOOKUP_LSEARCH is forced.
2508 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2510 24. Added sender and host information to the "rejected by local_scan()" log
2511 line; previously there was no indication of these.
2513 25. Added .include_if_exists.
2515 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2516 Linux. It turns out that not all file systems support this. Apparently some
2517 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2518 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2519 not supported on the file descriptor, is now ignored when Exim is trying to
2520 sync a directory. This applies only to Linux.
2522 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2524 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2525 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2528 29. Added support for maildirsize files from supplied patch (modified a bit).
2530 30. The use of :fail: followed by an empty string could lead Exim to respond to
2531 sender verification failures with (e.g.):
2533 550 Verification failed for <xxx>
2534 550 Sender verify failed
2536 where the first response line was missing the '-' that indicates it is not
2537 the final line of the response.
2539 31. The loop for finding the name of the user that called Exim had a hardwired
2540 limit of 10; it now uses the value of finduser_retries, which is used for
2541 all other user lookups.
2543 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2546 33. Exim was neglecting to zero errno before one call of strtol() when
2547 expanding a string and expecting an integer value. On some systems this
2548 resulted in spurious "integer overflow" errors. Also, it was casting the
2549 result into an int without checking.
2551 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2552 did not work. The code looks as if it has *never* worked, though it appears
2553 to have been documented since at least releast 1.62. I have made it work.
2555 35. The "timeout_DNS" error in retry rules, also documented since at least
2556 1.62, also never worked. As it isn't clear exactly what this means, and
2557 clearly it isn't a major issue, I have abolished the feature by treating it
2558 as "timeout", and writing a warning to the main and panic logs.
2560 36. The display of retry rules for -brt wasn't always showing the error code
2563 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2564 timeout_connect_A, timeout_connect_MX.
2566 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2567 to the empty sender.
2569 39. The daemon was not analysing the content of -oX till after it had closed
2570 stderr and disconnected from the controlling terminal. This meant that any
2571 syntax errors were only noted on the panic log, and the return code from
2572 the command was 0. By re-arranging the code a little, I've made the
2573 decoding happen first, so such errors now appear on stderr, and the return
2574 code is 1. However, the actual setting up of the sockets still happens in
2575 the disconnected process, so errors there are still only recorded on the
2578 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2579 connections (as happens on some IP stacks) was logged at start up time as
2580 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2581 it from "IPv6 and IPv4", which means that two separate sockets are being
2584 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2585 says whether AF_INET or AF_INET6 was passed as an argument.
2587 42. Exiwhat output was messed up when time zones were included in log
2590 43. Exiwhat now gives more information about the daemon's listening ports,
2591 and whether -tls-on-connect was used.
2593 44. The "port" option of the smtp transport is now expanded.
2595 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2596 silently ignored. Now an error message is written to the main and panic
2599 46. There's a new ACL modifier called "logwrite" which writes to a log file
2600 as soon as it is encountered.
2602 47. Added $local_user_uid and $local_user_gid at routing time.
2604 48. Exim crashed when trying to verify a sender address that was being
2607 49. Exim was recognizing only a space character after ".include". It now also
2608 recognizes a tab character.
2610 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2611 extracting the relevant information from the specification. The man page no
2612 longer contains scrambled data for the -d option, and I've added a section
2613 at the front about calling Exim under different names.
2615 51. Added "extra_headers" argument to the "mail" command in filter files.
2617 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2620 53. Installed eximstats 1.29.
2622 54. Added transport_filter_timeout as a generic transport option.
2624 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2625 Cc: header lines. This was required by RFC 822, but it not required by RFC
2628 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2629 incoming messages that did not have them. Now it does so only if the
2630 message originates locally, that is, if there is no associated remote host
2631 address. When Resent- header lines are present, this applies to the Resent-
2632 lines rather than the non-Resent- lines.
2634 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2635 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2637 58. Messages for configuration errors now include the name of the main
2638 configuration file - useful now that there may be more than one file in a
2639 list (.included file names were always shown).
2641 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2642 for those rare installations that do not start the daemon as root or run it
2643 setuid root. I've cut out the call to initgroups() if the daemon is not
2646 60. The Exim user and group can now be bound into the binary as text strings
2647 that are looked up at the start of Exim's processing.
2649 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2651 62. Added $mailstore_basename variable.
2653 63. Installed patch to sieve.c from Michael Haardt.
2655 64. When Exim failed to open the panic log after failing to open the main log,
2656 the original message it was trying to log was written to stderr and debug
2657 output, but if they were not available (the usual case in production), it
2658 was lost. Now it is written to syslog before the two lines that record the
2659 failures to open the logs.
2661 65. Users' Exim filters run in subprocesses under the user's uid. It is
2662 possible for a "deliver" command or an alias in a "personal" command to
2663 provoke an address rewrite. If logging of address rewriting is configured,
2664 this fails because the process is not running as root or exim. There may be
2665 a better way of dealing with this, but for the moment (because 4.30 needs
2666 to be released), I have disabled address rewrite logging when running a
2667 filter in a non-root, non-exim process.
2673 1. The buildconfig auxiliary program wasn't quoting the value set for
2674 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2675 not defined. This bug was masked in 4.22 by the effect that was fixed in
2678 2. Some messages that were rejected after a message id was allocated were
2679 shown as "incomplete" by exigrep. It no longer does this for messages that
2680 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2682 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2683 have allow_domain_literals set, the ID did not get logged in the <= line.
2684 Domain literals are now always recognized in Message-ID: header lines.
2686 4. The first argument for a ${extract expansion item is the key name or field
2687 number. Leading and trailing spaces in this item were not being ignored,
2688 causing some misleading effects.
2690 5. When deliver_drop_privilege was set, single queue runner processes started
2691 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2692 same command in the process it spins off) were not dropping privilege.
2694 6. When the daemon running as "exim" started a queue runner, it always
2695 re-executed Exim in the spun-off process. This is a waste of effort when
2696 deliver_drop_privilege is set. The new process now just calls the
2697 queue-runner function directly.
2703 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2706 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2707 ignored. Though the use of -oP was forcing the writing of a pid file, it
2708 was always written to the default place.
2710 3. If the message "no IP address found for host xxxx" is generated during
2711 incoming verification, it is now followed by identification of the incoming
2712 connection (so you can more easily find what provoked it).
2714 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2716 5. Added some features to "harden" Exim a bit more against certain attacks:
2718 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2719 be put in Local/Makefile. This is like the never_users runtime option,
2720 but it cannot be overridden. The default setting is "root".
2722 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2723 prefix string with which any file named in a -C command line option
2726 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2727 is retained for -C and -D only if the caller of Exim is root. Without
2728 it, the exim user may also use -C and -D and retain privilege.
2730 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2731 command line option is disabled.
2733 6. Macro names set by the -D option must start with an upper case letter, just
2734 like macro names defined in the configuration file.
2736 7. Added "dereference=" facility to LDAP.
2738 8. Two instances of the typo "uknown" in the source files are fixed.
2740 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2741 the Configure-Makefile script screwed up while processing it.
2743 10. Incorporated PCRE 4.4.
2745 11. The SMTP synchronization check was not operating right at the start of an
2746 SMTP session. For example, it could not catch a HELO sent before the client
2747 waited for the greeting. There is now a check for outstanding input at the
2748 point when the greeting is written. Because of the duplex, asynchronous
2749 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2750 way, but not yet received, when the check is performed.
2752 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2753 on TCP/IP sockets, because this apparently causes some broken clients to
2756 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2757 unchanged) from the Cygwin maintainer.
2759 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2760 was supported instead of testing for mbx. Effectively a typo.
2762 15. The spa authenticator server code was not checking that the input it
2763 received was valid base64.
2765 16. The debug output line for the "set" modifier in ACLs was not showing the
2766 name of the variable that was being set.
2768 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2770 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2771 lookup on its own. Something else was needed to trigger the lookup. For
2772 example, a match in host_lookup or the need for a host name in a host list.
2773 Now, if $sender_host_name is referenced and the host name has not yet been
2774 looked up, a lookup is performed. If the lookup fails, the variable remains
2775 empty, and $host_lookup_failed is set to "1".
2777 19. Added "eqi" as a case-independent comparison operator.
2779 20. The saslauthd authentication condition could segfault if neither service
2780 nor realm was specified.
2782 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2783 error message that was logged was misleading, and incoming SMTP
2784 connections were dropped. The message is now more accurate, and temporary
2785 errors are given to SMTP connections.
2787 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2788 (except RSET) with a 421 error, until QUIT is received. However, it was
2789 failing to send a response to QUIT.
2791 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2792 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2793 ACL is now run after the helo_try_verify_hosts code.
2795 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2796 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2797 case-independent, so other case variants are also recognized). Apparently
2798 some systems use these upper case variants.
2800 25. If more than two messages were waiting for the same host, and a transport
2801 filter was specified for the transport, Exim sent two messages over the
2802 same TCP/IP connection, and then failed with "socket operation on non-
2803 socket" when it tried to send the third.
2805 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2807 27. The extern definition of crypt16() in expand.c was not being excluded when
2808 the OS had its own crypt16() function.
2810 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2811 as a preferred synonym for return_size_limit, both as an option and as an
2814 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2816 30. Changed the default configuration ACL to relax the local part checking rule
2817 for addresses that are not in any local domains. For these addresses,
2818 slashes and pipe symbols are allowed within local parts, but the sequence
2819 /../ is explicitly forbidden.
2821 31. SPA server authentication was not clearing the challenge buffer before
2824 32. log_message in a "warn" ACL statement was writing to the reject log as
2825 well as to the main log, which contradicts the documentation and doesn't
2826 seem right (because no rejection is happening). So I have stopped it.
2828 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2829 However, I am unable to do any testing of this.
2831 34. Fixed an infelicity in the appendfile transport. When checking directories
2832 for a mailbox, to see if any needed to be created, it was accidentally
2833 using path names with one or more superfluous leading slashes; tracing
2834 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2836 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2837 contents are added to the log line that is written for every discarded
2838 recipient. (Previously a log_message setting was ignored.)
2840 36. The ${quote: operator now quotes the string if it is empty.
2842 37. The install script runs exim in order to find its version number. If for
2843 some reason other than non-existence or emptiness, which it checks, it
2844 could not run './exim', it was installing it with an empty version number,
2845 i.e. as "exim-". This error state is now caught, and the installation is
2848 38. An argument was missing from the function that creates an error message
2849 when Exim fails to connect to the socket for saslauthd authentication.
2850 This could cause Exim to crash, or give a corrupted message.
2852 39. Added isip, isip4, and isip6 to ${if conditions.
2854 40. The ACL variables $acl_xx are now saved with the message, and can be
2855 accessed later in routers, transports, and filters.
2857 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2858 strings in the file are not string-expanded.
2860 42. If a MAIL command specified a SIZE value that was too large to fit into an
2861 int variable, the check against message_size_limit failed. Such values are
2862 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2863 day this will have to be increased, but I don't think I want to be around
2864 when emails are that large.
2871 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2872 iconv() is not standard in FreeBSD.
2874 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2875 IPv6 enabled. The observed symptom was a segmentation fault on return from
2876 the function os_common_find_running_interfaces() in src/os.c.
2878 3. In the check_special_case() function in daemon.c I had used "errno" as an
2879 argument name, which causes warnings on some systems. This was basically a
2880 typo, since it was named "eno" in the comments!
2882 4. The code that waits for the clock to tick (at a resolution of some fraction
2883 of a second) so as to ensure message-id uniqueness was always waiting for
2884 at least one whole tick, when it could have waited for less. [This is
2885 almost certainly not relevant at current processor speeds, where it is
2886 unlikely to ever wait at all. But we try to future-proof.]
2888 5. The function that sleeps for a time interval that includes fractions of a
2889 second contained a race. It did not block SIGALRM between setting the
2890 timer, and suspending (a couple of lines later). If the interval was short
2891 and the sigsuspend() was delayed until after it had expired, the suspension
2892 never ended. On busy systems this could lead to processes getting stuck for
2895 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2896 process, before it forks any delivery processes. The open lookup caching
2897 mechanism meant that the open file or database connection was passed into
2898 the delivery process. The problem was that delivery processes always tidy
2899 up cached lookup data. This could cause a problem for the next delivery
2900 process started by the queue runner, because the external queue runner
2901 process does not know about the closure. So the next delivery process
2902 still has data in the lookup cache. In the case of a file lookup, there was
2903 no problem because closing a file descriptor in a subprocess doesn't affect
2904 the parent. However, if the lookup was caching a connection to a database,
2905 the connection was closed, and the second delivery process was likely to
2906 see errors such as "PGSQL: query failed: server closed the connection
2907 unexpectedly". The problem has been fixed by closing all cached lookups
2908 in a queue runner before running a delivery process.
2910 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2911 seem to have the "const" qualifier which it has on other OS. I've
2914 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2915 *of the same type* (client or server) with the same public name that an
2916 error should be diagnosed.
2918 9. When Exim looked up a host name for an IP address, but failed to find the
2919 original IP address when looking up the host name (a safety check), it
2920 output the message "<ip address> does not match any IP for NULL", which was
2921 confusing, to say the least. The bug was that the host name should have
2922 appeared instead of "NULL".
2924 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2925 user that is built into the binary, and the -C or -D options is used, root
2926 privilege is dropped before the configuration file is read. In addition,
2927 logging is switched to stderr instead of the normal log files. If the
2928 configuration then re-defines the Exim user, the unprivileged environment
2929 is probably not what is expected, so Exim logs a panic warning message (but
2932 However, if deliver_drop_privilege is set, the unprivileged state may well
2933 be exactly what is intended, so the warning has been cut out in that case,
2934 and Exim is allowed to try to write to its normal log files.
2940 1. smtp_return_error_details was not giving details for temporary sender
2941 or receiver verification errors.
2943 2. Diagnose a configuration error if two authenticators have the same public
2946 3. Exim used not to create the message log file for a message until the first
2947 delivery attempt. This could be confusing when incoming messages were held
2948 for policy or load reasons. The message log file is now created at the time
2949 the message is received, and an initial "Received" line is written to it.
2951 4. The automatically generated man page for command line options had a minor
2952 bug that caused no ill effects; however, a more serious problem was that
2953 the procedure for building the man page automatically didn't always
2954 operate. Consequently, release 4.20 contains an out-of-date version. This
2955 shouldn't happen again.
2957 5. When building Exim with embedded Perl support, the script that builds the
2958 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2959 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2961 6. The freeze_tell option was not being used for messages that were frozen on
2962 arrival, either by an ACL or by local_scan().
2964 7. Added the smtp_incomplete_transaction log selector.
2966 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2967 was accepting AUTH without a new EHLO.
2969 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2970 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2973 10. Logging of TCP/IP connections (when configured) now happens in the main
2974 daemon process instead of the child process, so that the TCP/IP connection
2975 count is more accurate (but it can never be perfect).
2977 11. The use of "drop" in a nested ACL was not being handled correctly in the
2978 outer ACL. Now, if condition failure induced by the nested "drop" causes
2979 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2980 or "require"), the connection is dropped.
2982 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2983 that yield "discard" can now be used with an "accept" or a "discard" verb,
2984 but an error is generated for any others (because I can't see a useful way
2985 to define what should happen).
2987 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2988 are now processed in the same way as lines in the Exim configuration file.
2989 In particular, continuation lines are supported.
2991 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
2993 15. Added -ti meaning -t -i.
2995 16. Check for letters, digits, hyphens, and dots in the names of dnslist
2996 domains, and warn by logging if others are found.
2998 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
2999 returned from GIFCONF when Exim is trying to find the list of interfaces on
3000 a host. The code in os.c has been modified to copy each ifreq to an aligned
3001 structure in all cases.
3003 Also, in some cases, the returned ifreq's were being copied to a 'struct
3004 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3005 means the last couple of bytes of an IPv6 address could be chopped if the
3006 ifreq contained only a normal sockaddr (14 bytes storage).
3008 18. Named domain lists were not supported in the hosts_treat_as_local option.
3009 An entry such as +xxxx was not recognized, and was treated as a literal
3012 19. Ensure that header lines added by a DATA ACL are included in the reject log
3013 if the ACL subsequently rejects the message.
3015 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3016 MD5 (which is deprecated).
3018 21. When testing a filter file using -bf, Exim was writing a message when it
3019 took the sender from a "From " line in the message, but it was not doing so
3020 when it took $return_path from a Return-Path: header line. It now does.
3022 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3023 with a valid header line field name (a series of printing characters
3024 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3026 23. Changed "disc" in the source to "disk" to conform to the documentation and
3027 the book and for uniformity.
3029 24. Ignore Sendmail's -Ooption=value command line item.
3031 25. When execve() failed while trying to run a command in a pipe transport,
3032 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3033 could be confused with a return value of 69 from the command itself. This
3034 has been changed to 127, the value the shell returns if it is asked to run
3035 a non-existent command. The wording for the related log line suggests a
3036 non-existent command as the problem.
3038 26. If received_header_text expands to an empty string, do not add a Received:
3039 header line to the message. (Well, it adds a token one on the spool, but
3040 marks it "old" so that it doesn't get used or transmitted.)
3042 27. Installed eximstats 1.28 (addition of -nt option).
3044 28. There was no check for failure on the call to getsockname() in the daemon
3045 code. This can fail if there is a shortage of resources on the system, with
3046 ENOMEM, for example. A temporary error is now given on failure.
3048 29. Contrary to the C standard, it seems that in some environments, the
3049 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3050 program. Exim now does this explicitly; it affects the formatting of
3051 timestamps using strftime().
3053 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3054 complaints. I've now initialized all the variables, to avoid this.
3056 32. Header lines added by a system filter were not being "seen" during
3057 transport-time rewrites.
3059 33. The info_callback() function passed to OpenSSL is set up with type void
3060 (*)(SSL *, int, int), as described somewhere. However, when calling the
3061 function (actually a macro) that sets it up, the type void(*)() is
3062 expected. I've put in a cast to prevent warnings from picky compilers.
3064 34. If a DNS black list lookup found a CNAME record, but there were no A
3065 records associated with the domain it pointed at, Exim crashed.
3067 35. If a DNS black list lookup returned more than one A record, Exim ignored
3068 all but the first. It now scans all returned addresses if a particular IP
3069 value is being sought. In this situation, the contents of the
3070 $dnslist_value variable are a list of all the addresses, separated by a
3073 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3074 to accept a host name and all its aliases if the forward lookup for any of
3075 them yielded the IP address of the incoming connection. Now it accepts only
3076 those names whose forward lookup yields the correct IP address. Any other
3077 names are discarded. This closes a loophole whereby a rogue DNS
3078 administrator could create reverse DNS records to break through a
3079 wildcarded host restriction in an ACL.
3081 37. If a user filter or a system filter that ran in a subprocess used any of
3082 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3083 the wrong values were passed to the pipe command ($thisaddress had the
3084 value of $0, $0 had the value of $1, etc). This bug was introduced by
3085 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3087 38. Improved the line breaking for long SMTP error messages from ACLs.
3088 Previously, if there was no break point between 40 and 75 characters, Exim
3089 left the rest of the message alone. Two changes have been made: (a) I've
3090 reduced the minimum length to 35 characters; (b) if it can't find a break
3091 point between 35 and 75 characters, it looks ahead and uses the first one
3092 that it finds. This may give the occasional overlong line, but at least the
3093 remaining text gets split now.
3095 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3096 file descriptors might be low, and that setting 1000 would always raise it.
3097 It turns out that in some environments, the limit is already over 1000 and
3098 that lowering it causes trouble. So now Exim takes care not to decrease it.
3100 40. When delivering a message, the value of $return_path is set to $sender_
3101 address at the start of routing (routers may change the value). By an
3102 oversight, this default was not being set up when an address was tested by
3103 -bt or -bv, which affected the outcome if any router or filter referred to
3106 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3107 log only when "message" or "log_message" are set. However, if one of the
3108 conditions was an address verification, or a call to a nested ACL, the
3109 messages generated by the underlying test were being passed through. This
3110 no longer happens. The underlying message is available in $acl_verify_
3111 message for both "message" and "log_message" expansions, so it can be
3112 passed through if needed.
3114 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3115 new expansion $bh_ to give the encoded byte string without charset
3116 translation. Translation happens only if iconv() is available; HAVE_ICONV
3117 indicates this at build time. HEADERS_CHARSET gives the charset to
3118 translate to; headers_charset can change it in the configuration, and
3119 "headers charset" can change it in an individual filter file.
3121 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3122 that creates RFC 2047 encoded "words" labels them as that charset instead
3123 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3124 expansion operator; (ii) when Exim creates a From: line for a local
3125 message; (iii) when a header line is rewritten to include a "phrase" part.
3127 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3128 buggy, causing it to skip the first lines of messages whose message ID
3129 ended in 'D'. This would not have bitten before Exim release 4.14, because
3130 message IDs were unlikely to end in 'D' before then. The effect was to have
3131 incorrect size information for certain domains.
3133 45. #include "config.h" was missing at the start of the crypt16.c module. This
3134 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3137 46. If there was a timeout during a "random" callout check, Exim treated it as
3138 a failure of the random address, and carried on sending RSET and the real
3139 address. If the delay was just some slowness somewhere, the response to the
3140 original RCPT would be taken as a response to RSET and so on, causing
3141 mayhem of various kinds.
3143 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3144 when I implemented it. It didn't allow for the fact that some option values
3145 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3146 the right test for positive values.
3148 48. Domain names in DNS records are case-independent. Exim always looks them up
3149 in lower case. Some resolvers return domain names in exactly the case they
3150 appear in the zone file, that is, they may contain uppercase letters. Not
3151 all resolvers do this - some return always lower case. Exim was treating a
3152 change of case by a resolver as a change of domain, similar to a widening
3153 of a domain abbreviation. This triggered its re-routing code and so it was
3154 trying to route what was effectively the same domain again. This normally
3155 caused routing to fail (because the router wouldn't handle the domain
3156 twice). Now Exim checks for this case specially, and just changes the
3157 casing of the domain that it ultimately uses when it transmits the message
3160 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3163 50. If a filter generated a file delivery with a non-absolute name (possible if
3164 no home directory exists for the router), the forbid_file option was not
3167 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3168 the existing equality matching.
3170 52. Exim was using ints instead of ino_t variables in some places where it was
3171 dealing with inode numbers.
3173 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3174 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3175 called TMPDIR, and if it finds it is different, it changes its value.
3177 54. The smtp_printf() function is now made available to local_scan() so
3178 additional output lines can be written before returning. There is also an
3179 smtp_fflush() function to enable the detection of a dropped connection.
3180 The variables smtp_input and smtp_batched_input are exported to
3183 55. Changed the default runtime configuration: the message "Unknown user"
3184 has been removed from the ACL, and instead placed on the localuser router,
3185 using the cannot_route_message feature. This means that any verification
3186 failures that generate their own messages won't get overridden. Similarly,
3187 the "Unrouteable address" message that was in the ACL for unverifiable
3188 relay addresses has also been removed.
3190 56. Added hosts_avoid_esmtp to the smtp transport.
3192 57. The exicyclog script was not checking for the esoteric option
3193 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3194 will work only if exicyclog is run under the appropriate euid.
3196 58. Following a discussion on the list, the rules by which Exim recognises line
3197 endings on incoming messages have been changed. The -dropcr and drop_cr
3198 options are now no-ops, retained only for backwards compatibility. The
3199 following line terminators are recognized: LF CRLF CR. However, special
3200 processing applies to CR:
3202 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3203 nor a local message in the state where . is a terminator.
3205 (ii) If a bare CR is encountered in a header line, an extra space is added
3206 after the line terminator so as not to end the header. The reasoning
3207 behind this is that bare CRs in header lines are most likely either
3208 to be mistakes, or people trying to play silly games.
3210 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3211 was being incorrectly given as 18 bytes larger than it should have been.
3212 This is a VOB (very old bug).
3214 60. This may never have affected anything current, but just in case it has:
3215 When the local host is found other than at the start of a list of hosts,
3216 the local host, those with the same MX, and any that follow, are discarded.
3217 When the list in question was part of a longer list of hosts, the following
3218 hosts (not currently being processed) were also being discarded. This no
3219 longer happens. I'm not sure if this situation could ever has previously
3222 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3225 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3226 header. This is something that is recommended in a new Internet Draft, and
3227 is something that is documented as being done by Sendmail. There are two
3228 possible values. For messages generated by the autoreply transport, Exim
3231 Auto-Submitted: auto-replied
3233 whereas for all other generated messages (e.g. bounces) it adds
3235 Auto-Submitted: auto-generated
3237 63. The "personal" condition in filters now includes a test for the
3238 Auto-Submitted: header. If it contains the string "auto-" the message it
3239 not considered personal.
3241 64. Added rcpt_include_affixes as a generic transport option.
3243 65. Added queue_only_override (default true).
3245 66. Added the syslog_duplication option.
3247 67. If what should have been the first header line of a message consisted of
3248 a space followed by a colon, Exim was mis-interpreting it as a header line.
3249 It isn't of course - it is syntactically invalid and should therefore be
3250 treated as the start of the message body. The misbehaviour could have
3251 caused a number of strange effects, including loss of data in subsequent
3252 header lines, and spool format errors.
3254 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3255 client host had authenticated. This control can now be exercised by an ACL
3256 for more flexibility.
3258 69. By default, callouts do not happen when testing with -bh. There is now a
3259 variant, -bhc, which does actually run the callout code, including
3260 consulting and updating the callout cache.
3262 70. Added support for saslauthd authentication, courtesy of Alexander
3265 71. If statvfs() failed on the spool or log directories while checking their
3266 size for availability, Exim confusingly gave the error "space shortage".
3267 Furthermore, in debugging mode it crashed with a floating point exception.
3268 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3269 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3270 serious problem, Exim now writes to the main and panic logs when this
3271 happens, with details of the failure. It then refuses to accept the
3272 incoming message, giving the message "spool directory problem" or "log
3273 directory problem" with a 421 code for SMTP messages.
3275 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3276 0, 1, and 2 exist, because some OS complain for execs without them (see
3277 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3278 descriptors. However, the code omitted to check that the open succeeded,
3279 causing mysterious errors if for some reason the permissions on /dev/null
3280 got screwed. Now Exim writes a message to the main and panic logs, and
3281 bombs out if it can't open /dev/null.
3283 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3284 interact so that it is all more flexible. It is supposed to remain
3285 backwards compatible. Also added extra_local_interfaces.
3287 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3288 to bomb out with an assertion failure - to the client this appears as a
3289 connection drop. This problem occurs in the part of the code that was taken
3290 from the Samba project. Fortunately, the assertion is in a very simple
3291 function, so I have fixed this by reproducing the function inline in the
3292 one place where it is called, and arranging for authentication to fail
3293 instead of killing the process with assert().
3295 75. The SPA client code was not working when the server requested OEM rather
3296 than Unicode encoding.
3298 76. Added code to make require_files with a specific uid setting more usable in
3299 the case where statting the file as root fails - usually a non-root-mounted
3300 NFS file system. When this happens and the failure is EACCES, Exim now
3301 forks a subprocess and does the per-uid checking as the relevant uid.
3303 77. Added process_log_path.
3305 78. If log_file_path was not explicitly set, a setting of check_log_space or
3306 check_log_inodes was ignored.
3308 79. If a space check for the spool or log partitions fails, the incident is now
3309 logged. Of course, in the latter case the data may get lost...
3311 80. Added the %p formatting code to string_format() so that it can be used to
3312 print addresses in debug_print(). Adjusted all the address printing in the
3313 debugging in store.c to use %p rather than %d.
3315 81. There was a concern that a line of code in smtp_in.c could overflow a
3316 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3317 initially expressed, the concern was not well-founded, because trailing
3318 spaces are removed early. However, if the trailing spaces were followed by
3319 a NULL, they did not get removed, so the overflow was possible. Two fixes
3322 (a) I re-wrote the offending code in a cleaner fashion.
3323 (b) If an incoming SMTP command contains a NULL character, it is rejected
3326 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3327 runs initgroups(), so that if the Exim user is in any additional groups,
3328 they will be used during message reception.
3334 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob