1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.228 2005/09/13 11:27:45 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added the "success_on_redirect" address verification option. See
10 NewStuff for rationale and an example.
12 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
14 PH/02 Patch to exigrep to allow it to work on syslog lines.
16 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
17 fread() to skip over the body file's header line, because in Cygwin the
18 header line is locked and is inaccessible.
20 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
21 co-exist for some time) to make it clear that it is the Exim ID that is
22 referenced, not the Message-ID: header line.
24 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
25 string_format() function, because snprintf() does not exist on all
28 PH/06 The use of forbid_filter_existstest now also locks out the use of the
29 ${stat: expansion item.
31 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
32 protocol synchronization error", to keep the pedants happy.
34 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
35 well as for IRIX systems, when gcc is being used. See the host.c source
38 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
40 PH/10 Named domain lists were not working if used in a queue_smtp_domains
43 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
44 transport and to the smtp transport in LMTP mode.
46 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
48 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
49 run a filter in a subprocess. This could lead to confusion in subsequent
50 lookups in the parent process. There should also be a search_tidyup() at
51 the end of the subprocess.
53 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
54 only if the host matched helo_try_verify_hosts, which caused the
55 verification to occur when the EHLO/HELO command was issued. The ACL just
56 tested the remembered result. Now, if a previous verification attempt has
57 not happened, "verify = helo" does it there and then.
59 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
61 TK/03 Fix log output including CR from clamd.
63 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
64 exist provoked a memory error which could cause a segfault.
66 PH/15 Installed PCRE 6.2
68 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
70 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
71 of the problem. Specifically, suggested +O2 rather than +O1 for the
74 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
76 PH/20 If a delivery was routed to a non-standard port by means of an SRV
77 record, the port was not correctly logged when the outgoing_port log
78 selector was set (it logged the transort's default port).
80 PH/21 Added support for host-specific ports to manualroute, queryprogram,
81 fallback_hosts, and "hosts" in the smtp transport.
83 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
84 host errors such as "Connection refused".
86 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
87 authentication with radiusclient 0.4.9:
89 - Error returned from rc_read_config was caught wrongly
90 - Username/password not passed on to radius server due to wrong length.
92 The presumption is that some radiusclient API changes for 4.51/PH/17
93 were not taken care of correctly. The code is still untested by me (my
94 Linux distribution still has 0.3.2 of radiusclient), but it was
95 contributed by a Radius user.
97 PH/24 When doing a callout, the value of $domain wasn't set correctly when
98 expanding the "port" option of the smtp transport.
100 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
101 while reading a MIME header. Thanks to Tom Hughes for a patch.
103 PH/24 Include config.h inside local_scan.h so that configuration settings are
106 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
107 that in an ACL for RCPT (for example), you can examine exactly what was
110 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
111 commands, but it was not correctly comparing the address with the actual
112 client host address. Thus, it would show the EHLO address in Received:
113 header lines when this was not necessary.
115 PH/27 Added the % operator to ${eval:}.
117 PH/28 Exim tries to create and chdir to its spool directory when it starts;
118 it should be ignoring failures (because with -C, for example, it has lost
119 privilege). It wasn't ignoring creation failures other than "already
122 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
125 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
126 because an input line was too long, either on its own, or by virtue of
127 too many continuations, the temporary file was not being removed, and the
128 return code was incorrect.
130 PH/31 Missing "BOOL" in function definition in filtertest.c.
132 PH/32 Applied Sieve patches from the maintainer.
134 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
136 PH/33 Added "verify = not_blind".
138 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
139 Local/Makefile (with some defaults set). These are used in built scripts
140 such as exicyclog, but they have never been used in the exim_install
141 script (though there are many overriding facilities there). I have
142 arranged that the exim_install script now takes note of these two
145 PH/35 Installed configuration files for Dragonfly.
147 PH/36 When a locally submitted message by a trusted user did not contain a
148 From: header, and the sender address was obtained from -f or from an SMTP
149 MAIL command, and the trusted user did not use -F to supply a sender
150 name, $originator_name was incorrectly used when constructing a From:
151 header. Furthermore, $originator_name was used for submission mode
152 messages from external hosts without From: headers in a similar way,
153 which is clearly wrong.
155 PH/37 Added control=suppress_local_fixups.
157 PH/38 When log_selector = +received_sender was set, and the addition of the
158 sender made the log line's construction buffer exactly full, or one byte
159 less than full, an overflow happened when the terminating "\n" was
162 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
163 when the result of a list match is failure because a DNS lookup failed.
165 PH/40 RM_COMMAND is now used in the building process.
167 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
168 the "build-* directories that it finds.
170 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
171 address in a domain literal was a prefix of an interface address.
173 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
174 when verifying a sender address, unless rewrite_headers is false.
176 PH/44 Wrote a long comment about why errors_to addresses are verified as
177 recipients, not senders.
179 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
180 the ratelimit ACL was added.
182 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
184 PH/46 Added extra information about PostgreSQL errors to the error string.
190 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
192 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
193 command itself ended in error, the underlying message about the transport
194 filter timeout was being overwritten with the pipe command error. Now the
195 underlying error message should be appended to the second error message.
197 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
199 PH/02 Reduce the amount of output that "make" produces by default. Full output
200 can still be requested.
202 PH/03 The warning log line about a condition test deferring for a "warn" verb
203 was being output only once per connection, rather than after each
204 occurrence (because it was using the same function as for successful
205 "warn" verbs). This seems wrong, so I have changed it.
207 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
208 it should not have, which might have caused a crash in the right
209 circumstances, but probably never did.
211 PH/04 Installed a modified version of Tony Finch's patch to make submission
212 mode fix the return path as well as the Sender: header line, and to
213 add a /name= option so that you can make the user's friendly name appear
216 TF/03 Added the control = fakedefer ACL modifier.
218 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
219 Mark Lowes for thorough testing.
221 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
223 TK/03 Merged latest SRS patch from Miles Wilton.
225 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
226 with the definition in sysexits.h (which is #included earlier).
227 Fortunately, Exim does not actually use EX_OK. The code used to try to
228 preserve the sysexits.h value, by assumimg that macro definitions were
229 scanned for macro replacements. I have been disabused of this notion,
230 so now the code just undefines EX_OK before #including unistd.h.
232 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
233 in the smtp transport. When a block could not be written in a single
234 write() function, the timeout was being re-applied to each part-write.
235 This seems wrong - if the receiver was accepting one byte at a time it
236 would take for ever. The timeout is now adjusted when this happens. It
237 doesn't have to be particularly precise.
239 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
240 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
242 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
243 without a domain if the check to <postmaster@domain> fails.
245 SC/01 Eximstats: added -xls and the ability to specify output files
246 (patch written by Frank Heydlauf).
248 SC/02 Eximstats: use FileHandles for outputing results.
250 SC/03 Eximstats: allow any combination of xls, txt, and html output.
252 SC/04 Eximstats: fixed display of large numbers with -nvr option
254 SC/05 Eximstats: fixed merging of reports with empty tables.
256 SC/06 Eximstats: added the -include_original_destination flag
258 SC/07 Eximstats: removed tabs and trailing whitespace.
260 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
262 TK/06 MBOX spool code: Add real "From " MBOX separator line
263 so the .eml file is really in mbox format (even though
264 most programs do not really care). Patch from Alex Miller.
266 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
267 The latter is generated from $received_to and is only set if the
268 message has one envelope recipient. SA can use these headers,
269 obviously out-of-the-box. Patch from Alex Miller.
271 PH/08 The ${def test on a variable was returning false if the variable's
272 value was "0", contrary to what the specification has always said!
273 The result should be true unless the variable is empty.
275 PH/09 The syntax error of a character other than { following "${if
276 def:variable_name" (after optional whitespace) was not being diagnosed.
277 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
278 accidental colon was present, for example, could give incorrect results.
280 PH/10 Tidied the code in a number of places where the st_size field of a stat()
281 result is used (not including appendfile, where other changes are about
284 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
285 This involved changing a lot of size variables from int to off_t. It
286 should work with maildirs and everything.
288 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
289 spamd dying while we are connected to it.
291 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
294 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
295 function for BDB 4.3.
297 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
299 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
300 the sources for PCRE have been re-arranged and more files are now
303 PH/15 The code I had for printing potentially long long variables in PH/11
304 above was not the best (it lost precision). The length of off_t variables
305 is now inspected at build time, and an appropriate printing format (%ld
306 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
307 to be "long long int" or "long int". This is needed for the internal
308 formatting function string_vformat().
310 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
311 the configuration file to be ":syslog", then the script "guesses" where
312 the logs files are, rather than using the compiled in default. In our
313 case the guess is not the same as the compiled default, so the script
314 suddenly stopped working when I started to use syslog. The patch checks
315 to see if log_file_path is "". If so, it attempts to read it from exim
316 with no configuration file to get the compiled in version, before it
317 falls back to the previous guessing code."
319 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
320 implementing BATV in an Exim configuration. See NewStuff for the gory
323 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
324 Makefile that are specific to HP-UX.
326 PH/18 If the "use_postmaster" option was set for a recipient callout together
327 with the "random" option, the postmaster address was used as the MAIL
328 FROM address for the random test, but not for the subsequent recipient
329 test. It is now used for both.
331 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
332 patch removes a few documentation additions to RFC 3028, because the
333 latest draft now contains them. It adds the new en;ascii-case comparator
334 and a new error check for 8bit text in MIME parts. Comparator and
335 require names are now matched exactly. I enabled the subaddress
336 extension, but it is not well tested yet (read: it works for me)."
338 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
339 rework some of the code of TK/09 above to avoid the hardwired use of
340 "%lld" and "long long". Replaced the call to snprintf() with a call to
343 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
344 records point to non-existent hosts", "retry timeout exceeded", and
345 "retry time not reached for any host after a long failure period".
347 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
348 experimental DomainKeys support:
350 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
351 (2) On an error such as an illegally used "control", the wrong name for
352 the control was given.
354 These problems did NOT occur unless DomainKeys support was compiled.
356 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
358 PH/24 Added ${if match_ip condition.
360 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
361 databases so that it will be absolutely obvious if a crash occurs in the
362 DB library. This is a regular occurrence (often caused by mis-matched
365 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
366 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
367 picked up on a user's system that detects such things. There doesn't seem
368 to be a gcc warning option for this - only an attribute that has to be
369 put on the function's prototype. It seems that in Fedora Core 4 they have
370 set this on a number of new functions. No doubt there will be more in due
373 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
374 specify a transport. However, if an address that was verified by such a
375 router was the subject of a callout, Exim crashed because it tried to
376 read the rcpt_include_affixes from the non-existent transport. Now it
377 just assumes that the setting of that option is false. This bug was
378 introduced by 4.51/PH/31.
380 PH/28 Changed -d+all to exclude +memory, because that information is very
381 rarely of interest, but it makes the output a lot bigger. People tend to
382 do -d+all out of habit.
384 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
385 code in os-type was giving problems when libc.so lives in lib64, like on
388 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
389 aren't the modern standard, and it seems that some systems' include files
390 don't always have them. Exim was already checking for some of the newer
391 ones like T_AAAA, and defining it itself. I've added checks for all the
392 record types that Exim uses.
394 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
395 not automatically generating a new one, as it is supposed to. This
396 prevented TLS from working. If the file did exist, but contained invalid
397 data, a new version was generated, as expected. It was only the case of a
398 non-existent file that was broken.
400 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
401 with a change in libdomainkeys > 0.64.
403 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
404 from DNS. If the selector record carries the flag, it now has
405 precedence over the domain-wide flag.
407 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
409 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
410 the use of a port name that isn't defined in /etc/services) occurred, the
411 message was deferred as in a normal delivery, and thus remained on the
412 spool, instead of being failed because of the mua_wrapper setting. This
413 is now fixed, and I tidied up some of the mua_wrapper messages at the
416 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
417 the messages in a hash of arrays rather than using individual hashes.
418 This is a bit cleaner and results in dramatic memory savings, albeit
419 at a slight CPU cost.
421 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
422 as requested by Marc Sherman.
424 SC/10 Eximstats: added histograms for user specified patterns as requested
427 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
429 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
430 fopen() in the content-scanning modules that did not already have it.
436 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
437 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
439 TK/02 Fix ACL "control" statement not being available in MIME ACL.
441 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
443 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
444 to test Sieve filters that use "vacation".
446 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
447 that changes the way the GnuTLS parameters are stored in the cache file.
448 The new format can be generated externally. For backward compatibility,
449 if the data in the cache doesn't make sense, Exim assumes it has read an
450 old-format file, and it generates new data and writes a new file. This
451 means that you can't go back to an older release without removing the
454 PH/03 A redirect router that has both "unseen" and "one_time" set does not
455 work if there are any delivery delays because "one_time" forces the
456 parent to be marked "delivered", so its unseen clone is never tried
457 again. For this reason, Exim now forbids the simultaneous setting of
460 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
461 redirected to themselves ("homonym" addresses). Read the long ChangeLog
462 entry if you want to know the details. The fix, however, neglected to
463 consider the case when local delivery batching is involved. The test for
464 "previously delivered" was not happening when checking to see if an
465 address could be batched with a previous (undelivered) one; under
466 certain circumstances this could lead to multiple deliveries to the same
469 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
470 in its include files, and this causes problems building Exim.
472 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
473 header_syntax) but Exim was just ignoring anything given after a slash.
474 In particular, this caused confusion with an attempt to use "verify =
475 reverse_host_lookup/defer_ok". An error is now given when options are
476 supplied for verify items that do not have them. (Maybe reverse_host_
477 lookup should have a defer_ok option, but that's a different point.)
479 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
480 defined by RFC 821) to 2048, because there were problems with some AUTH
481 commands, and RFC 1869 says the size should be increased for extended
482 SMTP commands that take arguments.
484 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
487 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
488 "unknown" error; now it says that the functionality isn't in the binary.
490 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
491 an address' error message when a string expansion fails (syntax or
492 whatever). Otherwise the password may appear in the log. Following change
493 PH/42 below, there is no longer a chance of it appearing in a bounce
496 PH/11 Installed exipick version 20050225.0 from John Jetmore.
498 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
499 of its addresses was ever tried. (Bugzilla bug #2.)
501 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
502 the result incorrectly in the debug output. (It correctly added a newline
503 to what was transported.)
505 TF/01 Added $received_time.
507 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
508 commented out examples of how to interface to a virus scanner and to
509 SpamAssassin. Also added commented examples of av_scanner and
510 spamd_address settings.
512 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
513 and controls are allowed in which ACLs. There were a couple of minor
514 errors. Some of the entries in the conditions table (which is a table of
515 where they are NOT allowed) were getting very unwieldy; rewrote them as a
516 negation of where the condition IS allowed.
518 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
520 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
521 header file does not have a version number, so I've had to invent a new
522 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
523 API. The code is untested by me (my Linux distribution still has 0.3.2 of
524 radiusclient), but it was contributed by a Radius user.
526 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
527 files or directories, for OpenSSL.
529 PH/19 When an Exim process that is running as root has to create an Exim log
530 file, it does so in a subprocess that runs as exim:exim so as to get the
531 ownership right at creation (otherwise, other Exim processes might see
532 the file with the wrong ownership). There was no test for failure of this
533 fork() call, which would lead to the process getting stuck as it waited
534 for a non-existent subprocess. Forks do occasionally fail when resources
535 run out. I reviewed all the other calls to fork(); they all seem to check
538 PH/20 When checking for unexpected SMTP input at connect time (before writing
539 the banner), Exim was not dealing correctly with a non-positive return
540 from the read() function. If the client had disconnected by this time,
541 the result was a log entry for a synchronization error with an empty
542 string after "input=" when read() returned zero. If read() returned -1
543 (an event I could not check), uninitialized data bytes were printed.
544 There were reports of junk text (parts of files, etc) appearing after
547 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
549 PH/22 Added support for macro redefinition, and (re)definition in between
550 driver and ACL definitions.
552 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
553 forgetting to use the resulting value; it was using the unexpanded value.
555 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
556 hadn't been configured. The fix is from Juergen Kreileder, who
557 understands it better than I do:
559 "Here's what I see happening with three configured cyrus_sasl
560 authenticators configured (plain, login, cram-md5):
562 On startup auth_cyrus_sasl_init() gets called for each of these.
563 This means three calls to sasl_listmech() without a specified mech_list.
564 => SASL tests which mechs of all available mechs actually work
565 => three warnings about OTP not working
566 => the returned list contains: plain, login, cram-md5, digest-md5, ...
568 With the patch, sasl_listmech() also gets called three times. But now
569 SASL's mech_list option is set to the server_mech specified in the the
570 authenticator. Or in other words, the answer from sasl_listmech()
571 gets limited to just the mech you're testing for (which is different
573 => the return list contains just 'plain' or 'login', 'cram-md5' or
574 nothing depending on the value of ob->server_mech.
576 I've just tested the patch: Authentication still works fine,
577 unavailable mechs specified in the exim configuration are still
578 caught, and the auth.log warnings about OTP are gone."
580 PH/25 When debugging is enabled, the contents of the command line are added
581 to the debugging output, even when log_selector=+arguments is not
584 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
585 answer is "GNU", and only if the return is "GNU/something" is the answer
588 PH/27 $acl_verify_message is now set immediately after the failure of a
589 verification in an ACL, and so is available in subsequent modifiers. In
590 particular, the message can be preserved by coding like this:
592 warn !verify = sender
593 set acl_m0 = $acl_verify_message
595 Previously, $acl_verify_message was set only while expanding "message"
596 and "log_message" when a very denied access.
598 PH/28 Modified OS/os.c-Linux with
600 -#ifndef OS_LOAD_AVERAGE
601 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
603 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
604 nomenclature these days.)
606 PH/29 Installed patch from the Sieve maintainer that adds the options
607 sieve_useraddress and sieve_subaddress to the redirect router.
609 PH/30 In these circumstances:
610 . Two addresses routed to the same list of hosts;
611 . First host does not offer TLS;
612 . First host accepts first address;
613 . First host gives temporary error to second address;
614 . Second host offers TLS and a TLS session is established;
615 . Second host accepts second address.
616 Exim incorrectly logged both deliveries with the TLS parameters (cipher
617 and peerdn, if requested) that were in fact used only for the second
620 PH/31 When doing a callout as part of verifying an address, Exim was not paying
621 attention to any local part prefix or suffix that was matched by the
622 router that accepted the address. It now behaves in the same way as it
623 does for delivery: the affixes are removed from the local part unless
624 rcpt_include_affixes is set on the transport.
626 PH/32 Add the sender address, as F=<...>, to the log line when logging a
627 timeout during the DATA phase of an incoming message.
629 PH/33 Sieve envelope tests were broken for match types other than :is. I have
630 applied a patch sanctioned by the Sieve maintainer.
632 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
633 the uid or gid is negative. A case of a negative gid caused this to be
634 noticed. The fix allows for either to be negative.
636 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
637 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
640 PH/36 The change PH/12 above was broken. Fixed it.
642 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
643 the grounds that routing the same address twice would always produce the
644 same answer. This might have been true once, but it is certainly no
645 longer true now. Routing a child address may depend on the previous
646 routing that produced that child. Some complicated redirection strategies
647 went wrong when messages had multiple recipients, and made Exim's
648 behaviour dependent on the order in which the addresses were given.
650 I have moved the duplicate checking until after the routing is complete.
651 Exim scans the addresses that are assigned to local and remote
652 transports, and removes any duplicates. This means that more work will be
653 done, as duplicates will always all be routed, but duplicates are
654 presumably rare, so I don't expect this is of any significance.
656 For deliveries to pipes, files, and autoreplies, the duplicate checking
657 still happens during the routing process, since they are not going to be
660 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
661 It corrects a timeout issue with spamd. This is Ian's comment: "The
662 background is that sometimes spamd either never reads data from a
663 connection it has accepted, or it never writes response data. The exiscan
664 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
665 blindly assumes that writes won't block so it may never time out."
667 PH/39 Allow G after quota size as well as K and M.
669 PH/40 The value set for $authenticated_id in an authenticator may not contain
670 binary zeroes or newlines because the value is written to log lines and
671 to spool files. There was no check on this. Now the value is run through
672 the string_printing() function so that such characters are converted to
673 printable escape sequences.
675 PH/41 $message_linecount is a new variable that contains the total number of
676 lines in the message. Compare $body_linecount, which is the count for the
679 PH/42 Exim no longer gives details of delivery errors for specific addresses in
680 bounce and delay warning messages, except in certain special cases, which
683 (a) An SMTP error message from a remote host;
684 (b) A message specified in a :fail: redirection;
685 (c) A message specified in a "fail" command in a system filter;
686 (d) A message specified in a FAIL return from the queryprogram router;
687 (e) A message specified by the cannot_route_message router option.
689 In these cases only, Exim does include the error details in bounce and
690 warning messages. There are also a few cases where bland messages such
691 as "unrouteable address" or "local delivery error" are given.
693 PH/43 $value is now also set for the "else" part of a ${run expansion.
695 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
696 being worked on, but at least Exim now implements the latest version to
699 PH/45 In a pipe transport, although a timeout while waiting for the pipe
700 process to complete was treated as a delivery failure, a timeout while
701 writing the message to the pipe was logged, but erroneously treated as a
702 successful delivery. Such timeouts include transport filter timeouts. For
703 consistency with the overall process timeout, these timeouts are now
704 treated as errors, giving rise to delivery failures by default. However,
705 there is now a new Boolean option for the pipe transport called
706 timeout_defer, which, if set TRUE, converts the failures into defers for
707 both kinds of timeout. A transport filter timeout is now identified in
710 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
711 systems where "make" and "gmake" are different, calling "gmake" at top
712 level broke things. I've arranged for the value of $(MAKE) to be passed
713 from the Makefile to this script so that it can call the same version of
717 A note about Exim versions 4.44 and 4.50
718 ----------------------------------------
720 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
721 changes of various kinds. As a consequence, a big documentation update was
722 needed. This delayed the release for rather longer than seemed good, especially
723 in the light of a couple of (minor) security issues. Therefore, the changes
724 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
725 release. So 4.44 and 4.50 are in effect two different branches that both start
728 I have left the 4.50 change log unchanged; it contains all the changes since
729 4.43. The change log for 4.44 is below; many of its items are identical to
730 those for 4.50. This seems to be the most sensible way to preserve the
731 historical information.
737 1. Minor wording change to the doc/README.SIEVE file.
739 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
740 computation of the current number of files was incorrect.
742 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
743 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
746 4. Give more explanation in the error message when the command for a transport
747 filter fails to execute.
749 5. There are several places where Exim runs a non-Exim command in a
750 subprocess. The SIGUSR1 signal should be disabled for these processes. This
751 was being done only for the command run by the queryprogram router. It is
752 now done for all such subprocesses. The other cases are: ${run, transport
753 filters, and the commands run by the lmtp and pipe transports.
755 6. Added CONFIGURE_GROUP build-time option.
757 7. Some older OS have a limit of 256 on the maximum number of file
758 descriptors. Exim was using setrlimit() to set 1000 as a large value
759 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
760 systems. I've change it so that if it can't get 1000, it tries for 256.
762 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
763 was an oversight, and furthermore, ever since the addition of extra
764 controls (e.g. 4.43/32), the checks on when to allow different forms of
765 "control" were broken. There should now be diagnostics for all cases when a
766 control that does not make sense is encountered.
768 9. Added the /retain_sender option to "control=submission".
770 10. $recipients is now available in the predata ACL (oversight).
772 11. Tidy the search cache before the fork to do a delivery from a message
773 received from the command line. Otherwise the child will trigger a lookup
774 failure and thereby defer the delivery if it tries to use (for example) a
775 cached ldap connection that the parent has called unbind on.
777 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
778 of $address_data from the recipient verification was clobbered by the
781 13. The value of address_data from a sender verification is now available in
782 $sender_address_data in subsequent conditions in the ACL statement.
784 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
786 15. Added a new option "connect=<time>" to callout options, to set a different
789 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
790 was its contents. (It was OK if the option was not defined at all.)
792 17. A "Completed" log line is now written for messages that are removed from
793 the spool by the -Mrm option.
795 18. New variables $sender_verify_failure and $recipient_verify_failure contain
796 information about exactly what failed.
798 19. Added -dd to debug only the daemon process.
800 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
801 handles timeouts, both on the server side and network timeouts. Renamed the
802 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
804 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
805 It is now set to "smtps".
807 22. $host_address is now set to the target address during the checking of
810 23. When checking ignore_target_hosts for an ipliteral router, no host name was
811 being passed; this would have caused $sender_host_name to have been used if
812 matching the list had actually called for a host name (not very likely,
813 since this list is usually IP addresses). A host name is now passed as
816 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
817 code that specifies a non-restarting handler (typically sigaction() in
818 modern systems) in an attempt to fix a rare and obscure crash bug.
820 25. Narrowed the window for a race in the daemon that could cause it to ignore
821 SIGCHLD signals. This is not a major problem, because they are used only to
822 wake it up if nothing else does.
824 26. A malformed maildirsize file could cause Exim to calculate negative values
825 for the mailbox size or file count. Odd effects could occur as a result.
826 The maildirsize information is now recalculated if the size or filecount
829 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
830 support for a long time. Removed HAVE_SYS_VFS_H.
832 28. Installed the latest version of exipick from John Jetmore.
834 29. In an address list, if the pattern was not a regular expression, an empty
835 subject address (from a bounce message) matched only if the pattern was an
836 empty string. Non-empty patterns were not even tested. This was the wrong
837 because it is perfectly reasonable to use an empty address as part of a
838 database query. An empty address is now tested by patterns that are
839 lookups. However, all the other forms of pattern expect the subject to
840 contain a local part and a domain, and therefore, for them, an empty
841 address still always fails if the pattern is not itself empty.
843 30. Exim went into a mad DNS loop when attempting to do a callout where the
844 host was specified on an smtp transport, and looking it up yielded more
847 31. Re-factored the code for checking spool and log partition space into a
848 function that finds that data and another that does the check. The former
849 is then used to implement four new variables: $spool_space, $log_space,
850 $spool_inodes, and $log_inodes.
852 32. The RFC2047 encoding function was originally intended for short strings
853 such as real names; it was not keeping to the 75-character limit for
854 encoded words that the RFC imposes. It now respects the limit, and
855 generates multiple encoded words if necessary. To be on the safe side, I
856 have increased the buffer size for the ${rfc2047: expansion operator from
859 33. It is now permitted to omit both strings after an "if" condition; if the
860 condition is true, the result is "true". As before, when the second string
861 is omitted, a false condition yields an empty string. This makes it less
862 cumbersome to write custom ACL and router conditions.
864 34. Failure to deliver a bounce message always caused it to be frozen, even if
865 there was an errors_to setting on the router. The errors_to setting is now
868 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
869 canonical form (fully expanded) before being placed in
870 $sender_host_address.
872 36. The table in the code that translates DNS record types into text (T_A to
873 "A" for instance) was missing entries for NS and CNAME. It is just possible
874 that this could have caused confusion if both these types were looked up
875 for the same domain, because the text type is used as part of Exim's
876 per-process caching. But the chance of anyone hitting this buglet seems
879 37. The dnsdb lookup has been extended in a number of ways.
881 (1) There is a new type, "zns", which walks up the domain tree until it
882 finds some nameserver records. It should be used with care.
884 (2) There is a new type, "mxh", which is like "mx" except that it returns
885 just the host names, not the priorities.
887 (3) It is now possible to give a list of domains (or IP addresses) to be
888 looked up. The behaviour when one of the lookups defers can be
889 controlled by a keyword.
891 (4) It is now possible to specify the separator character for use when
892 multiple records are returned.
894 38. The dnslists ACL condition has been extended: it is now possible to supply
895 a list of IP addresses and/or domains to be looked up in a particular DNS
898 39. Added log_selector=+queue_time_overall.
900 40. When running the queue in the test harness, wait just a tad after forking a
901 delivery process, to get repeatability of debugging output.
903 41. Include certificate and key file names in error message when GnuTLS fails
904 to set them up, because the GnuTLS error message doesn't include the name
905 of the failing file when there is a problem reading it.
907 42. Allow both -bf and -bF in the same test run.
909 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
911 44. The "Exiscan patch" is now merged into the mainline Exim source.
913 45. Sometimes the final signoff response after QUIT could fail to get
914 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
915 < 0 before doing a fflush(). This bug looks as though it goes back to the
916 introduction of TLS in release 3.20, but "sometimes" must have been rare
917 because the tests only now provoked it.
919 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
920 (this can affect the format of dates).
922 47. exim_tidydb, when checking for the continued existence of a message for
923 which it has found a message-specific retry record, was not finding
924 messages that were in split spool directories. Consequently, it was
925 deleting retry records that should have stayed in existence.
927 48. Steve fixed some bugs in eximstats.
929 49. The SPA authentication driver was not abandoning authentication and moving
930 on to the next authenticator when an expansion was forced to fail,
931 contradicting the general specification for all authenticators. Instead it
932 was generating a temporary error. It now behaves as specified.
934 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
935 (the order specifies the preference for clients). The order is now AES256,
936 AES128, 3DES, ARCFOUR128.
938 51. Small patch to Sieve code - explicitly set From: when generating an
941 52. Exim crashed if a remote delivery caused a very long error message to be
942 recorded - for instance if somebody sent an entire SpamAssassin report back
943 as a large number of 550 error lines. This bug was coincidentally fixed by
944 increasing the size of one of Exim's internal buffers (big_buffer) that
945 happened as part of the Exiscan merge. However, to be on the safe side, I
946 have made the code more robust (and fixed the comments that describe what
949 53. Now that there can be additional text after "Completed" in log lines (if
950 the queue_time_overall log selector is set), a one-byte patch to exigrep
951 was needed to allow it to recognize "Completed" as not the last thing in
954 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
955 patch that reportedly fixes this has been added. I am not expert enough to
956 create a test for it. This is what the patch creator wrote:
958 "I found a little strange behaviour of ldap code when working with
959 Windows 2003 AD Domain, where users was placed in more than one
960 Organization Units. When I tried to give exim partial DN, the exit code
961 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
962 But simultaneously result of request was absolutely normal ldap result,
963 so I produce this patch..."
965 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
966 REFERENCE, so I have modified the code to exclude the patch when that macro
969 55. Some experimental protocols are using DNS PTR records for new purposes. The
970 keys for these records are domain names, not reversed IP addresses. The
971 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
972 leaves it alone. Component reversal etc. now happens only for IP addresses.
975 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
977 57. Double the size of the debug message buffer (to 2048) so that more of very
978 long debug lines gets shown.
980 58. The exicyclog utility now does better if the number of log files to keep
981 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
983 59. Two changes related to the smtp_active_hostname option:
985 (1) $smtp_active_hostname is now available as a variable.
986 (2) The default for smtp_banner uses $smtp_active_hostname instead
987 of $primary_hostname.
989 60. The host_aton() function is supposed to be passed a string that is known
990 to be a valid IP address. However, in the case of IPv6 addresses, it was
991 not checking this. This is a hostage to fortune. Exim now panics and dies
992 if the condition is not met. A case was found where this could be provoked
993 from a dnsdb PTR lookup with an IPv6 address that had more than 8
994 components; fortuitously, this particular loophole had already been fixed
995 by change 4.50/55 above.
997 If there are any other similar loopholes, the new check in host_aton()
998 itself should stop them being exploited. The report I received stated that
999 data on the command line could provoke the exploit when Exim was running as
1000 exim, but did not say which command line option was involved. All I could
1001 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1002 running as the user.
1005 61. There was a buffer overflow vulnerability in the SPA authentication code
1006 (which came originally from the Samba project). I have added a test to the
1007 spa_base64_to_bits() function which I hope fixes it.
1010 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1011 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1013 63. The daemon start-up calls getloadavg() while still root for those OS that
1014 need the first call to be done as root, but it missed one case: when
1015 deliver_queue_load_max is set with deliver_drop_privilege. This is
1016 necessary for the benefit of the queue runner, because there is no re-exec
1017 when deliver_drop_privilege is set.
1019 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1020 This has been fixed.
1022 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1023 was in use, was not putting the data itself into the right store pool;
1024 consequently, it could be overwritten for a subsequent message in the same
1025 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1028 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1030 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1031 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1032 the function were treating the return as a boolean value, which happened to
1033 work because 0=false and not-0=true, but is not correct code.
1035 68. The host_aton() function was not handling scoped IPv6 addresses (those
1036 with, for example, "%eth0" on the end) correctly.
1038 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1039 negated items (that is, ~something) in unsigned ints. Some compilers
1040 apparently mutter when there is no cast.
1042 70. If an address verification called from an ACL failed, and did not produce a
1043 user-specific message (i.e. there was only a "system" message), nothing was
1044 put in $acl_verify_message. In this situation, it now puts the system
1047 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1048 session; change 4.31/43 added the unwanted input to the log line - except
1049 that it did not do this in the start of session case. It now does.
1051 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1052 This is wrong and can cause the other end to generate a synchronization
1053 error if it is another Exim or anything else that does the synchronization
1054 check. A QUIT command is no longer sent after a timeout.
1056 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1057 during host lookups.
1059 74. The defer_ok option of callout verification was not working if it was used
1060 when verifying addresses in header lines, that is, for this case:
1062 verify = header_sender/callout=defer_ok
1064 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1065 those file descriptors could be used for SMTP connections. If anything
1066 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1067 could be sent to the SMTP client, causing chaos. The daemon now opens
1068 stdin, stdout, and stderr to /dev/null when it puts itself into the
1071 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1072 log by default. The user can override this with suitable Perl magic.
1074 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1075 the log message when discard triggers, was not working for the DATA ACL or
1076 for the non-SMTP ACL.
1078 78. Error message wording change in sieve.c.
1080 79. If smtp_accept_max_per_host was set, the number of connections could be
1081 restricted to fewer than expected, because the daemon was trying to set up
1082 a new connection before checking whether the processes handling previous
1083 connections had finished. The check for completed processes is now done
1084 earlier. On busy systems, this bug wouldn't be noticed because something
1085 else would have woken the daemon, and it would have reaped the completed
1088 80. If a message was submitted locally by a user whose login name contained one
1089 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1090 It caused a spool format error. I have fixed the spool reading code. A
1091 related problem was that the "from" clause in the Received: line became
1092 illegal because of the space(s). It is now covered by ${quote_local_part.
1094 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1097 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1099 83. Patch from Sieve maintainer to fix unterminated string problem in
1100 "vacation" handling.
1102 84. Some minor changes to the Linux configuration files to help with other
1103 OS variants using glibc.
1105 85. One more patch for Sieve to update vacation handling to latest spec.
1108 ----------------------------------------------------
1109 See the note above about the 4.44 and 4.50 releases.
1110 ----------------------------------------------------
1116 1. Change 4.43/35 introduced a bug that caused file counts to be
1117 incorrectly computed when quota_filecount was set in an appendfile
1120 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1121 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1124 3. Give more explanation in the error message when the command for a transport
1125 filter fails to execute.
1127 4. There are several places where Exim runs a non-Exim command in a
1128 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1129 was being done only for the command run by the queryprogram router. It is
1130 now done for all such subprocesses. The other cases are: ${run, transport
1131 filters, and the commands run by the lmtp and pipe transports.
1133 5. Some older OS have a limit of 256 on the maximum number of file
1134 descriptors. Exim was using setrlimit() to set 1000 as a large value
1135 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1136 systems. I've change it so that if it can't get 1000, it tries for 256.
1138 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1139 was an oversight, and furthermore, ever since the addition of extra
1140 controls (e.g. 4.43/32), the checks on when to allow different forms of
1141 "control" were broken. There should now be diagnostics for all cases when a
1142 control that does not make sense is encountered.
1144 7. $recipients is now available in the predata ACL (oversight).
1146 8. Tidy the search cache before the fork to do a delivery from a message
1147 received from the command line. Otherwise the child will trigger a lookup
1148 failure and thereby defer the delivery if it tries to use (for example) a
1149 cached ldap connection that the parent has called unbind on.
1151 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1152 of $address_data from the recipient verification was clobbered by the
1153 sender verification.
1155 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1156 was its contents. (It was OK if the option was not defined at all.)
1158 11. A "Completed" log line is now written for messages that are removed from
1159 the spool by the -Mrm option.
1161 12. $host_address is now set to the target address during the checking of
1162 ignore_target_hosts.
1164 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1165 being passed; this would have caused $sender_host_name to have been used if
1166 matching the list had actually called for a host name (not very likely,
1167 since this list is usually IP addresses). A host name is now passed as
1170 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1171 code that specifies a non-restarting handler (typically sigaction() in
1172 modern systems) in an attempt to fix a rare and obscure crash bug.
1174 15. Narrowed the window for a race in the daemon that could cause it to ignore
1175 SIGCHLD signals. This is not a major problem, because they are used only to
1176 wake it up if nothing else does.
1178 16. A malformed maildirsize file could cause Exim to calculate negative values
1179 for the mailbox size or file count. Odd effects could occur as a result.
1180 The maildirsize information is now recalculated if the size or filecount
1183 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1184 support for a long time. Removed HAVE_SYS_VFS_H.
1186 18. Updated exipick to current release from John Jetmore.
1188 19. Allow an empty sender to be matched against a lookup in an address list.
1189 Previously the only cases considered were a regular expression, or an
1192 20. Exim went into a mad DNS lookup loop when doing a callout where the
1193 host was specified on the transport, if the DNS lookup yielded more than
1196 21. The RFC2047 encoding function was originally intended for short strings
1197 such as real names; it was not keeping to the 75-character limit for
1198 encoded words that the RFC imposes. It now respects the limit, and
1199 generates multiple encoded words if necessary. To be on the safe side, I
1200 have increased the buffer size for the ${rfc2047: expansion operator from
1203 22. Failure to deliver a bounce message always caused it to be frozen, even if
1204 there was an errors_to setting on the router. The errors_to setting is now
1207 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1208 canonical form (fully expanded) before being placed in
1209 $sender_host_address.
1211 24. Updated eximstats to version 1.33
1213 25. Include certificate and key file names in error message when GnuTLS fails
1214 to set them up, because the GnuTLS error message doesn't include the name
1215 of the failing file when there is a problem reading it.
1217 26. Expand error message when OpenSSL has problems setting up cert/key files.
1220 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1221 (this can affect the format of dates).
1223 28. exim_tidydb, when checking for the continued existence of a message for
1224 which it has found a message-specific retry record, was not finding
1225 messages that were in split spool directories. Consequently, it was
1226 deleting retry records that should have stayed in existence.
1228 29. eximstats updated to version 1.35
1229 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1230 1.35 - bugfix such that pie charts by volume are generated correctly
1232 30. The SPA authentication driver was not abandoning authentication and moving
1233 on to the next authenticator when an expansion was forced to fail,
1234 contradicting the general specification for all authenticators. Instead it
1235 was generating a temporary error. It now behaves as specified.
1237 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1238 (the order specifies the preference for clients). The order is now AES256,
1239 AES128, 3DES, ARCFOUR128.
1241 31. Small patch to Sieve code - explicitly set From: when generating an
1244 32. Exim crashed if a remote delivery caused a very long error message to be
1245 recorded - for instance if somebody sent an entire SpamAssassin report back
1246 as a large number of 550 error lines. This bug was coincidentally fixed by
1247 increasing the size of one of Exim's internal buffers (big_buffer) that
1248 happened as part of the Exiscan merge. However, to be on the safe side, I
1249 have made the code more robust (and fixed the comments that describe what
1252 33. Some experimental protocols are using DNS PTR records for new purposes. The
1253 keys for these records are domain names, not reversed IP addresses. The
1254 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1255 leaves it alone. Component reversal etc. now happens only for IP addresses.
1258 34. The host_aton() function is supposed to be passed a string that is known
1259 to be a valid IP address. However, in the case of IPv6 addresses, it was
1260 not checking this. This is a hostage to fortune. Exim now panics and dies
1261 if the condition is not met. A case was found where this could be provoked
1262 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1263 components; fortuitously, this particular loophole had already been fixed
1264 by change 4.50/55 or 4.44/33 above.
1266 If there are any other similar loopholes, the new check in host_aton()
1267 itself should stop them being exploited. The report I received stated that
1268 data on the command line could provoke the exploit when Exim was running as
1269 exim, but did not say which command line option was involved. All I could
1270 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1271 running as the user.
1274 35. There was a buffer overflow vulnerability in the SPA authentication code
1275 (which came originally from the Samba project). I have added a test to the
1276 spa_base64_to_bits() function which I hope fixes it.
1279 36. The daemon start-up calls getloadavg() while still root for those OS that
1280 need the first call to be done as root, but it missed one case: when
1281 deliver_queue_load_max is set with deliver_drop_privilege. This is
1282 necessary for the benefit of the queue runner, because there is no re-exec
1283 when deliver_drop_privilege is set.
1285 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1286 was in use, was not putting the data itself into the right store pool;
1287 consequently, it could be overwritten for a subsequent message in the same
1288 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1291 38. Sometimes the final signoff response after QUIT could fail to get
1292 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1293 < 0 before doing a fflush(). This bug looks as though it goes back to the
1294 introduction of TLS in release 3.20, but "sometimes" must have been rare
1295 because the tests only now provoked it.
1301 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1302 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1303 Now it returns an integer. A number of calls were still expecting a T/F
1304 return. Fortuitously, in all cases, the tests worked in OK situations,
1305 which is the norm. However, things would have gone wrong on any write
1306 failures on the smtp file descriptor. This function is used when sending
1307 messages over SMTP and also when doing verify callouts.
1309 2. When Exim is called to do synchronous delivery of a locally submitted
1310 message (the -odf or -odi options), it no longer closes stderr before doing
1313 3. Implemented the mua_wrapper option.
1315 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1317 5. Implemented the functions header_remove(), header_testname(),
1318 header_add_at_position(), and receive_remove_recipient(), and exported them
1321 6. If an ACL "warn" statement specified the addition of headers, Exim already
1322 inserted X-ACL-Warn: at the start if there was no header name. However, it
1323 was not making this test for the second and subsequent header lines if
1324 there were newlines in the string. This meant that an invalid header could
1325 be inserted if Exim was badly configured.
1327 7. Allow an ACL "warn" statement to add header lines at the start or after all
1328 the Received: headers, as well as at the end.
1330 8. Added the rcpt_4xx retry error code.
1332 9. Added postmaster_mailfrom=xxx to callout verification option.
1334 10. Added mailfrom=xxxx to the callout verification option, for verify=
1337 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1338 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1340 12. Inserted some casts to stop certain compilers warning when using pointer
1341 differences as field lengths or precisions in printf-type calls (mostly
1342 affecting debugging statements).
1344 13. Added optional readline() support for -be (dynamically loaded).
1346 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1347 same clock tick as a message's arrival, so that its received time was the
1348 same as the "first fail" time on the retry record, and that message
1349 remained on the queue past the ultimate address timeout, every queue runner
1350 would try a delivery (because it was past the ultimate address timeout) but
1351 after another failure, the ultimate address timeout, which should have then
1352 bounced the address, did not kick in. This was a "< instead of <=" error;
1353 in most cases the first failure would have been in the next clock tick
1354 after the received time, and all would be well.
1356 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1357 being recognized when the domain list was tested by the match_domain
1358 condition in an expansion string.
1360 16. Added the ${str2b64: operator.
1362 17. Exim was always calling setrlimit() to set a large limit for the number of
1363 processes, without checking whether the existing limit was already
1364 adequate. (It did check for the limit on file descriptors.) Furthermore,
1365 errors from getrlimit() and setrlimit() were being ignored. Now they are
1366 logged to the main and panic logs, but Exim does carry on, to try to do its
1367 job under whatever limits there are.
1369 18. Imported PCRE 5.0.
1371 19. Trivial typo in log message " temporarily refused connection" (the leading
1374 20. If the log selector return_path_on_delivery was set and an address was
1375 redirected to /dev/null, the delivery process crashed because it assumed
1376 that a return path would always be set for a "successful" delivery. In this
1377 case, the whole delivery is bypassed as an optimization, and therefore no
1380 21. Internal re-arrangement: the function for sending a challenge and reading
1381 a response while authentication was assuming a zero-terminated challenge
1382 string. It's now changed to take a pointer and a length, to allow for
1383 binary data in such strings.
1385 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1387 23. Exim was not respecting finduser_retries when seeking the login of the
1388 uid under which it was called; it was always trying 10 times. (The default
1389 setting of finduser_retries is zero.) Also, it was sleeping after the final
1390 failure, which is pointless.
1392 24. Implemented tls_on_connect_ports.
1394 25. Implemented acl_smtp_predata.
1396 26. If the domain in control=submission is set empty, Exim assumes that the
1397 authenticated id is a complete email address when it generates From: or
1398 Sender: header lines.
1400 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1401 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1402 chown and chgrp in /bin and hostname in /usr/bin.
1404 28. Exim was keeping the "process log" file open after each use, just as it
1405 does for the main log. This opens the possibility of it remaining open for
1406 long periods when the USR1 signal hits a daemon. Occasional processlog
1407 errors were reported, that could have been caused by this. Anyway, it seems
1408 much more sensible not to leave this file open at all, so that is what now
1411 29. The long-running daemon process does not normally write to the log once it
1412 has entered its main loop, and it closes the log before doing so. This is
1413 so that log files can straightforwardly be renamed and moved. However,
1414 there are a couple of unusual error situations where the daemon does write
1415 log entries, and I had neglected to close the log afterwards.
1417 30. The text of an SMTP error response that was received during a remote
1418 delivery was being truncated at 512 bytes. This is too short for some of
1419 the long messages that one sometimes sees. I've increased the limit to
1422 31. It is now possible to make retry rules that apply only when a message has a
1423 specific sender, in particular, an empty sender.
1425 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1426 it possible to be selective about when SMTP synchronization is enforced.
1428 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1430 32. Implemented hosts_connection_nolog.
1432 33. Added an ACL for QUIT.
1434 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1437 35. Added mailbox_size and mailbox_filecount to appendfile.
1439 36. Added control = no_multiline_responses to ACLs.
1441 37. There was a bug in the logic of the code that waits for the clock to tick
1442 in the case where the clock went backwards by a substantial amount such
1443 that the microsecond fraction of "now" was more than the microsecond
1444 fraction of "then" (but the whole seconds number was less).
1446 38. Added support for the libradius Radius client library this is found on
1447 FreeBSD (previously only the radiusclient library was supported).
1453 1. When certain lookups returned multiple values in the form name=value, the
1454 quoting of the values was not always being done properly. Specifically:
1455 (a) If the value started with a double quote, but contained no whitespace,
1457 (b) If the value contained whitespace other than a space character (i.e.
1458 tabs or newlines or carriage returns) it was not quoted.
1459 This fix has been applied to the mysql and pgsql lookups by writing a
1460 separate quoting function and calling it from the lookup code. The fix
1461 should probably also be applied to nisplus, ibase and oracle lookups, but
1462 since I cannot test any of those, I have not disturbed their existing code.
1464 2. A hit in the callout cache for a specific address caused a log line with no
1465 reason for rejecting RCPT. Now it says "Previous (cached) callout
1466 verification failure".
1468 3. There was an off-by-one bug in the queryprogram router. An over-long
1469 return line was truncated at 256 instead of 255 characters, thereby
1470 overflowing its buffer with the terminating zero. As well as fixing this, I
1471 have increased the buffer size to 1024 (and made a note to document this).
1473 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1474 when Exim is waiting for an SMTP response from a remote server, Exim
1475 restarts its select() call on the socket, thereby resetting its timeout.
1476 This is not a problem when such interrupts are rare. Somebody set up a cron
1477 job to run exiwhat every 2 minutes, which is less than the normal select()
1478 timeout (5 or 10 minutes). This meant that the select() timeout never
1479 kicked in because it was always reset. I have fixed this by comparing the
1480 time when an interrupt arrives with the time at the start of the first call
1481 to select(). If more time than the timeout has elapsed, the interrupt is
1482 treated as a timeout.
1484 5. Some internal re-factoring in preparation for the addition of Sieve
1485 extensions (by MH). In particular, the "personal" test is moved to a
1486 separate function, and given an option for scanning Cc: and Bcc: (which is
1487 not set for Exim filters).
1489 6. When Exim created an email address using the login of the caller as the
1490 local part (e.g. when creating a From: or Sender: header line), it was not
1491 quoting the local part when it contained special characters such as @.
1493 7. Installed new OpenBSD configuration files.
1495 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1496 try to make them clearer.
1498 9. Callout options, other than the timeout value, were being ignored when
1499 verifying sender addresses in header lines. For example, when using
1501 verify = header_sender/callout=no_cache
1503 the cache was (incorrectly) being used.
1505 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1506 only the Cygwin environment.
1508 11. When return_path_on_delivery was set as a log selector, if different remote
1509 addresses in the same message used different return paths and parallel
1510 remote delivery occurred, the wrong values would sometimes be logged.
1511 (Whenever a remote delivery process finished, the return path value from
1512 the most recently started remote delivery process was logged.)
1514 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1515 lines when AUTH and/or TLS are in use. This is the "received protocol"
1516 field. Exim used to use "asmtp" for authenticated SMTP, without any
1517 indication (in the protocol name) for TLS use. Now it follows the RFC and
1518 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1519 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1520 names appear in log lines as well as in Received: header lines.
1522 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1523 extensions, and comparison tests, and to fix some bugs.
1525 14. Changes to the "personal" filter test:
1527 (1) The test was buggy in that it was just doing the equivalent of
1528 "contains" tests on header lines. For example, if a user's address was
1529 anne@some.where, the "personal" test would incorrectly be true for
1531 To: susanne@some.where
1533 This test is now done by extracting each address from the header in turn,
1534 and checking the entire address. Other tests that are part of "personal"
1535 are now done using regular expressions (for example, to check local parts
1536 of addresses in From: header lines).
1538 (2) The list of non-personal local parts in From: addresses has been
1539 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1540 taken from the Sieve specification recommendations.
1542 (3) If the message contains any header line starting with "List-" it is
1543 treated as non-personal.
1545 (4) The test for "circular" in the Subject: header line has been removed
1546 because it now seems ill-conceived.
1548 15. Minor typos in src/EDITME comments corrected.
1550 16. Installed latest exipick from John Jetmore.
1552 17. If headers_add on a router specified a text string that was too long for
1553 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1554 of string_sprintf() is now avoided.
1556 18. $message_body_size was not set (it was always zero) when running the DATA
1557 ACL and the local_scan() function.
1559 19. For the "mail" command in an Exim filter, no default was being set for
1560 the once_repeat time, causing a random time value to be used if "once" was
1561 specified. (If the value happened to be <= 0, no repeat happened.) The
1562 default is now 0s, meaning "never repeat". The "vacation" command was OK
1563 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1564 (I found it when inspecting the code).
1566 20. There is now an overall timeout for performing a callout verification. It
1567 defaults to 4 times the callout timeout, which applies to individual SMTP
1568 commands during the callout. The overall timeout applies when there is more
1569 than one host that can be tried. The timeout is checked before trying the
1570 next host. This prevents very long delays if there are a large number of
1571 hosts and all are timing out (e.g. when the network connections are timing
1572 out). The value of the overall timeout can be changed by specifying an
1573 additional sub-option for "callout", called "maxwait". For example:
1575 verify = sender/callout=5s,maxwait=20s
1577 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1578 to the end before writing, but this should make it even safer).
1580 22. Exim was forgetting that it had advertised PIPELINING for the second and
1581 subsequent messages on an SMTP connection. It was also not resetting its
1582 memory on STARTTLS and an internal HELO.
1584 23. When Exim logs an SMTP synchronization error within a session, it now
1585 records whether PIPELINING has been advertised or not.
1587 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1588 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1589 rather than long int.
1591 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1593 26. Added the never_mail option to autoreply.
1599 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1600 crash if the getsockname() call failed; this can happen if a connection is
1601 closed very soon after it is established. The problem was simply in the
1602 order in which certain operations were done, causing Exim to try to write
1603 to the SMTP stream before it had set up the file descriptor. The bug has
1604 been fixed by making things happen in the correct order.
1610 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1611 before the connection was closed, thus losing the rejection response.
1613 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1614 some early Solaris releases, but causes trouble in current releases where
1615 socklen_t is defined.
1617 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1620 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1623 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1624 that releasing the top of it at the end releases what was used for sub-
1625 expansions (unless the block got too big). However, discard this block if
1626 the first thing is a variable or header, so that we can use its block when
1627 it is dynamic (useful for very large $message_headers, for example).
1629 6. Lookups now cache *every* query, not just the most recent. A new, separate
1630 store pool is used for this. It can be recovered when all lookup caches are
1631 flushed. Lookups now release memory at the end of their result strings.
1632 This has involved some general refactoring of the lookup sources.
1634 7. Some code has been added to the store_xxx() functions to reduce the amount
1635 of flapping under certain conditions.
1637 8. log_incoming_interface used to affect only the <= reception log lines. Now
1638 it causes the local interface and port to be added to several more SMTP log
1639 lines, for example "SMTP connection from", and rejection lines.
1641 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1643 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1645 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1646 could be overwritten at the end of the current message (or the start of a
1647 new message if it was set in a HELO ACL). The value is now preserved for
1648 the duration of the SMTP connection.
1650 12. If a transport had a headers_rewrite setting, and a matching header line
1651 contained an unqualified address, that address was qualified, even if it
1652 did not match any rewriting rules. The underlying bug was that the values
1653 of the flags that permit the existence of unqualified sender and recipient
1654 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1655 non-local messages, and by -bnq for local messages) were not being
1656 preserved with the message after it was received.
1658 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1659 "next input=" as part of the text comprising the host identity instead of
1660 the correct text. The code was using the same buffer for two different
1661 strings. However, depending on which order the printing function evaluated
1662 its arguments, the bug did not always show up. Under Linux, for example, my
1663 test suite worked just fine.
1665 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1666 doesn't work with some older versions of Perl. It has been changed to "my",
1667 which in any case is probably the better facility to use.
1669 15. A really picky compiler found some instances of statements for creating
1670 error messages that either had too many or two few arguments for the format
1673 16. The size of the buffer for calls to the DNS resolver has been increased
1674 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1675 for addresses that have a lot of PTR records. This alleviates a problem; it
1676 does not fully solve it.
1678 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1679 buffer now truncates the list and logs the incident, which is the same
1680 action as happens when Exim is looking up a host name and its aliases.
1681 Previously in this situation something unpredictable would happen;
1682 sometimes it was "internal error: store_reset failed".
1684 18. If a server dropped the connection unexpectedly when an Exim client was
1685 using GnuTLS and trying to read a response, the client delivery process
1686 crashed while trying to generate an error log message.
1688 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1689 string, for example:
1691 warn message = H1: something\nH2: something
1693 the text was added as a single header line from Exim's point of view
1694 though it ended up OK in the delivered message. However, searching for the
1695 second and subsequent header lines using $h_h2: did not work. This has been
1696 fixed. Similarly, if a system filter added multiple headers in this way,
1697 the routers could not see them.
1699 20. Expanded the error message when iplsearch is called with an invalid key to
1700 suggest using net-iplsearch in a host list.
1702 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1703 ACLs are no longer actually imposed (and a message to that effect is
1706 22. If a "gecos" field in a passwd entry contained escaped characters, in
1707 particular, if it contained a \" sequence, Exim got it wrong when building
1708 a From: or a Sender: header from that name. A second bug also caused
1709 incorrect handling when an unquoted " was present following a character
1710 that needed quoting.
1712 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1713 was not being matched caselessly.
1715 24. Arranged for all hyphens in the exim.8 source to be escaped with
1718 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1719 properly. Recipient callout cache records were still being keyed to include
1720 the sender, even when use_sender was set false. This led to far more
1721 callouts that were necessary. The sender is no longer included in the key
1722 when use_sender is false.
1724 26. Added "control = submission" modifier to ACLs.
1726 27. Added the ${base62d: operator to decode base 62 numbers.
1728 28. dnsdb lookups can now access SRV records.
1730 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1731 the configuration file.
1733 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1734 (-v) mode. This makes the output for a verbose queue run more intelligible.
1736 31. Added a use_postmaster feature to recipient callouts.
1738 32. Added the $body_zerocount variable, containing the number of binary zero
1739 bytes in the message body.
1741 33. The time of last modification of the "new" subdirectory is now used as the
1742 "mailbox time last read" when there is a quota error for a maildir
1745 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1747 35. Added +ignore_unknown as a special item in host lists.
1749 36. Code for decoding IPv6 addresses in host lists is now included, even if
1750 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1751 address was recognized as an IP address, but was then not correctly decoded
1752 into binary, causing unexpected and incorrect effects when compared with
1759 1. Very minor rewording of debugging text in manualroute to say "list of
1760 hosts" instead of "hostlist".
1762 2. If verify=header_syntax was set, and a header line with an unqualified
1763 address (no domain) and a large number of spaces between the end of the
1764 name and the colon was received, the reception process suffered a buffer
1765 overflow, and (when I tested it) crashed. This was caused by some obsolete
1766 code that should have been removed. The fix is to remove it!
1768 3. When running in the test harness, delay a bit after writing a bounce
1769 message to get a bit more predictability in the log output.
1771 4. Added a call to search_tidyup() just before forking a reception process. In
1772 theory, someone could use a lookup in the expansion of smtp_accept_max_
1773 per_host which, without the tidyup, could leave open a database connection.
1775 5. Added the variables $recipient_data and $sender_data which get set from a
1776 lookup success in an ACL "recipients" or "senders" condition, or a router
1777 "senders" option, similar to $domain_data and $local_part_data.
1779 6. Moved the writing of debug_print from before to after the "senders" test
1782 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1783 problems for message scanning, either using a data ACL, or using
1784 local_scan() because the Received: header was not generated till after they
1785 were called (in order to set the time as the time of reception completion).
1786 I have revised the way this works. The header is now generated after the
1787 body is received, but before the ACL or local_scan() are called. After they
1788 are run, the timestamp in the header is updated.
1794 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1795 before starting a queue runner without re-exec. This happened only when
1796 deliver_drop_privilege was set or when the Exim user was set to root. The
1797 effect of the bug was that timeouts during subsequent deliveries caused
1798 crashes instead of being properly handled. The handler is now left at its
1799 default (and expected) setting.
1801 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1802 message, again when deliver_drop_privilege is set or Exim is run as root.
1803 The bug described in (1) was not present in this case, but the tidying up
1804 of the other signals was missing. I have made the two cases consistent.
1806 3. The ignore_target_hosts setting on a manualroute router was being ignored
1807 for hosts that were looked up using the /MX notation.
1809 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1812 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1813 operated on the sender address. After changing the $sender_address to <>
1814 for the sender address verify, Exim was re-instated it as the original
1815 (before rewriting) address, but remembering that it had rewritten it, so it
1816 wasn't rewriting it again. This bug also had the effect of breaking the
1817 sender address verification caching when the sender address was rewritten.
1819 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1820 This has been changed so that if the ip literal address matches
1821 ignore_target_hosts, the router declines.
1823 7. Added expansion conditions match_domain, match_address, and match_local_
1824 part (NOT match_host).
1826 8. The placeholder for the Received: header didn't have a length field set.
1828 9. Added code to Exim itself and to exim_lock to test for a specific race
1829 condition that could lead to file corruption when using MBX delivery. The
1830 issue is with the lockfile that is created in /tmp. If this file is removed
1831 after a process has opened it but before that process has acquired a lock,
1832 there is the potential for a second process to recreate the file and also
1833 acquire a lock. This could lead to two Exim processes writing to the file
1834 at the same time. The added code performs the same test as UW imapd; it
1835 checks after acquiring the lock that its file descriptor still refers to
1836 the same named file.
1838 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1839 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1842 11. Added the smtp_active_hostname option. If used, this will typically be made
1843 to depend on the incoming interface address. Because $interface_address is
1844 not set up until the daemon has forked a reception process, error responses
1845 that can happen earlier (such as "too many connections") no longer contain
1848 12. If an expansion in a condition on a "warn" statement fails because a lookup
1849 defers, the "warn" statement is abandoned, and the next ACL statement is
1850 processed. Previously this caused the whole ACL to be aborted.
1852 13. Added the iplsearch lookup type.
1854 14. Added ident_timeout as a log selector.
1856 15. Added tls_certificate_verified as a log selector.
1858 16. Added a global option tls_require_ciphers (compare the smtp transport
1859 option of the same name). This controls incoming TLS connections.
1861 17. I finally figured out how to make tls_require_ciphers do a similar thing
1862 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1863 before starting the TLS session.
1865 18. Tabs are now shown as \t in -bP output.
1867 19. If the log selector return_path_on_delivery was set, Exim crashed when
1868 bouncing a message because it had too many Received: header lines.
1870 20. If two routers both had headers_remove settings, and the first one included
1871 a superfluous trailing colon, the final name in the first list and the
1872 first name in the second list were incorrectly joined into one item (with a
1873 colon in the middle).
1879 1. Added -C and -D options to the exinext utility, mainly to make it easier
1880 to include in the automated testing, but these could be helpful when
1881 multiple configurations are in use.
1883 2. The exinext utility was not formatting the output nicely when there was
1884 an alternate port involved in the retry record key, nor when there was a
1885 message id as well (for retries that were specific to a specific message
1886 and a specific host). It was also confused by IPv6 addresses, because of
1887 the additional colons they contain. I have fixed the IPv4 problem, and
1888 patched it up to do a reasonable job for IPv6.
1890 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1891 delivery, the log line now contains "pipelined" if PIPELINING was used.
1893 4. An SMTP transport process used to panic and die if the bind() call to set
1894 an explicit outgoing interface failed. This has been changed; it is now
1895 treated in the same way as a connect() failure.
1897 5. A reference to $sender_host_name in the part of a conditional expansion
1898 that was being skipped was still causing a DNS lookup. This no longer
1901 6. The def: expansion condition was not recognizing references to header lines
1902 that used bh_ and bheader_.
1904 7. Added the _cache feature to named lists.
1906 8. The code for checking quota_filecount in the appendfile transport was
1907 allowing one more file than it should have been.
1909 9. For compatibility with Sendmail, the command line option
1917 and sets the incoming protocol and host name (for trusted callers). The
1918 host name and its colon can be omitted when only the protocol is to be set.
1919 Note the Exim already has two private options, -pd and -ps, that refer to
1920 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1921 "s", but I don't think that's a major issue.
1923 10. A number of refactoring changes to the code, none of which should affect
1926 (a) The number of logging options was getting close to filling up the
1927 32-bit word that was used as a bit map. I have split them into two classes:
1928 those that are passed in the argument to log_write(), and those that are
1929 only ever tested independently outside of that function. These are now in
1930 separate 32-bit words, so there is plenty of room for expansion again.
1931 There is no change in the user interface or the logging behaviour.
1933 (b) When building, for example, log lines, the code previously used a
1934 macro that called string_cat() twice, in order to add two strings. This is
1935 not really sufficiently general. Furthermore, there was one instance where
1936 it was actually wrong because one of the argument was used twice, and in
1937 one call a function was used. (As it happened, calling the function twice
1938 did not affect the overall behaviour.) The macro has been replaced by a
1939 function that can join an arbitrary number of extra strings onto a growing
1942 (c) The code for expansion conditions now uses a table and a binary chop
1943 instead of a serial search (which was left over from when there were very
1944 few conditions). Also, it now recognizes conditions like "pam" even when
1945 the relevant support is not compiled in: a suitably worded error message is
1946 given if an attempt is made to use such a condition.
1948 11. Added ${time_interval:xxxxx}.
1950 12. A bug was causing one of the ddress fields not to be passed back correctly
1951 from remote delivery subprocesses. The field in question was not being
1952 subsequently used, so this caused to problems in practice.
1954 13. Added new log selectors queue_time and deliver_time.
1956 14. Might have fixed a bug in maildirsizefile handling that threw up
1957 "unexpected character" debug warnings, and recalculated the data
1958 unnecessarily. In any case, I expanded the warning message to give more
1961 15. Added the message "Restricted characters in address" to the statements in
1962 the default ACL that block characters like @ and % in local parts.
1964 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1965 Three changes have been made:
1967 (a) There was a serious bug; a negative response to MAIL caused the whole
1968 recipient domain to be cached as invalid, thereby blocking all messages
1969 to all local parts at the same domain, from all senders. This bug has
1970 been fixed. The domain is no longer cached after a negative response to
1971 MAIL if the sender used is not empty.
1973 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1976 (c) A new callout option, "use_sender" has been added for people who want
1977 the modified behaviour.
1983 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1986 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1987 indeed breaks things for older releases.
1989 3. Added additional logging to the case where there is a problem reading data
1990 from a filter that is running in a subprocess using a pipe, in order to
1991 try to track down a specific problem.
1993 4. Testing facility fudge: when running in the test harness and attempting
1994 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1995 getting "No route to host". Convert this to a timeout.
1997 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2000 6. Some OS don't have socklen_t but use size_t instead. This affects the
2001 fifth argument of getsockopt() amongst other things. This is now
2002 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2003 can be set for individual OS. I have set it for SunOS5, OSF1, and
2004 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2005 some earlier ones do not.
2007 7. Change 4.30/15 was not doing the test caselessly.
2009 8. The standard form for an IPv6 address literal was being rejected by address
2010 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2011 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2012 this, as well as the form without the "IPv6" on the front (but only when
2013 address literals are enabled, of course).
2015 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2017 10. Exim crashed if a message with an empty sender address specified by -f
2018 encountered a router with an errors_to setting. This could be provoked only
2019 by a command such as
2023 where an empty string was supplied; "<>" did not hit this bug.
2025 11. Installed PCRE release 4.5.
2027 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2028 remained set. It is now erased.
2030 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2031 times from message ids (which are base 36 rather than the normal 62).
2033 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2034 were being counted as actual protocol errors, and logged if the log
2035 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2036 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2037 and DATA following a set of rejected RCPTs do not count as protocol errors.
2038 In other words, Exim assumes they were pipelined, though this may not
2039 actually be the case. Of course, in all cases the client gets an
2040 appropriate error code.
2042 15. If a lookup fails in an ACL condition, a message about the failure may
2043 be available; it is used if testing the ACL cannot continue, because most
2044 such messages specify what the cause of the deferral is. However, some
2045 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2046 that caused an old message to be retained and used if a later statement
2047 caused a defer, replacing the real cause of the deferral.
2049 16. If an IP address had so many PTR records that the DNS lookup buffer
2050 was not large enough to hold them, Exim could crash while trying to process
2051 the truncated data. It now detects and logs this case.
2053 17. Further to 4.21/58, another change has been made: if (and only if) the
2054 first line of a message (the first header line) ends with CRLF, a bare LF
2055 in a subsequent header line has a space inserted after it, so as not to
2056 terminate the header.
2058 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2059 unnecessarily, used atoi() instead of strtol(), and didn't check the
2060 termination when getting file sizes from file names by regex.
2062 19. Completely re-implemented the support for maildirsize files, in the light
2063 of a number of problems with the previous contributed implementation
2064 (4.30/29). In particular:
2066 . If the quota is zero, the maildirsize file is maintained, but no quota is
2069 . If the maildir directory does not exist, it is created before any attempt
2070 to write a maildirsize file.
2072 . The quota value in the file is just a cache; if the quota is changed in
2073 the transport, the new value overrides.
2075 . A regular expression is available for excluding directories from the
2078 20. The autoreply transport checks the characters in options that define the
2079 message's headers; it allows continued headers, but it was checking with
2080 isspace() after an embedded newline instead of explicitly looking for a
2083 21. If all the "regular" hosts to which an address was routed had passed their
2084 expiry times, and had not reached their retry times, the address was
2085 bounced, even if fallback hosts were defined. Now Exim should go on to try
2088 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2089 equivalent code in the SMTP transport. Some hosts send humungous responses
2090 to HELO/EHLO, more than 1024 it seems.
2092 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2093 gives compiler warnings in some environments. I've now done it "properly",
2096 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2097 (because of problems with the built-in one) was declared to return uschar *
2098 instead of char *, causing compiler failure.
2100 25. Fixed a file descriptor leak when processing alias/forward files.
2102 26. Fixed a minor format string issue in dbfn.c.
2104 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2106 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2107 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2110 29. When a sender address is verified, it is cached, to save repeating the test
2111 when there is more than one recipient in a message. However, when the
2112 verification involves a callout, it is possible for different callout
2113 options to be set for different recipients. It is too complicated to keep
2114 track of this in the cache, so now Exim always runs a verification when a
2115 callout is required, relying on the callout cache for the optimization.
2116 The overhead is duplication of the address routing, but this should not be
2119 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2120 to be verified with callout=postmaster, and the main callout worked but the
2121 postmaster check failed, the verification correctly failed. However, if a
2122 subsequent RCPT command asked for sender verification *without* the
2123 postmaster check, incorrect caching caused this verification also to fail,
2126 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2127 it was not caching the DNS options (qualify_single, search_parents) that
2128 were used when the lookup failed. A subsequent lookup with different
2129 options therefore always gave the same answer, though there were cases
2130 where it should not have. (Example: a "domains = !$mx_any" option on a
2131 dnslookup router: the "domains" option is always processed without any
2132 widening, but the router might have qualify_single set.) Now Exim uses the
2133 cached value only when the same options are set.
2135 32. Added John Jetmore's "exipick" utility to the distribution.
2137 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2138 than a timeout (e.g. a certificate is required, and is not provided), an
2139 Exim server now closes the connection immediately. Previously it waited for
2140 the client to close - but if the client is SSL, it seems that they each
2141 wait for each other, leading to a delay before one of them times out.
2143 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2144 maintained 0.8.x compatibility because I don't think many are using it, and
2145 it is clearly obsolete.
2147 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2150 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2151 client certificate was expired. A simple patch fixes this, though I don't
2152 understand the full logic of why the verify callback is called multiple
2155 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2156 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2157 which causes problems with some clients (such as the Certicom SSL Plus
2158 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2159 disables the coutermeasure allowing Eudora to connect."
2161 38. Exim was not checking that a write() to a log file succeeded. This could
2162 lead to Bad Things if a log got too big, in particular if it hit a file
2163 size limit. Exim now panics and dies if it cannot write to a log file, just
2164 as it does if it cannot open a log file.
2166 39. Modified OS/Makefile-Linux so that it now contains
2168 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2170 The two -D definitions ensure that Exim is compiled with large file
2171 support, which makes it possible to handle log files that are bigger than
2174 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2175 instance) a domain was checked against a named list that involved a lookup,
2176 causing $domain_data to be set, then another domain was checked against the
2177 same list, then the first domain was re-checked, the value of $domain_data
2178 after the final check could be wrong. In particular, if the second check
2179 failed, it could be set empty. This bug probably also applied to
2182 41. The strip_trailing_dot option was not being applied to the address given
2183 with the -f command-line option.
2185 42. The code for reading a message's header from the spool was incrementing
2186 $received_count, but never initializing it. This meant that the value was
2187 incorrect (doubled) while delivering a message in the same process in which
2188 it was received. In the most common configuration of Exim, this never
2189 happens - a fresh exec is done - but it can happen when
2190 deliver_drop_privilege is set.
2192 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2193 it now includes up to 150 characters of the unexpected data in the log
2196 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2197 and building data strings. The size of both of these buffers was 10 000
2198 bytes - far larger than anybody would *ever* want, thought I. Needless to
2199 say, somebody hit the limit. I have increased the maximum line length to
2200 20 000 and the maximum data length of concatenated lines to 100 000. I have
2201 also fixed two bugs, because there was no checking on these buffers. Tsk,
2202 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2203 buffer is too small.
2205 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2206 lsearch lookups. Now it does.
2208 46. When parsing a route_list item in a manualroute router, a fixed-length
2209 buffer was used for the list of hosts. I made this 1024 bytes long,
2210 thinking that nobody would ever have a list of hosts that long. Wrong.
2211 Somebody had a whole pile of complicated expansion conditions, and the
2212 string was silently truncated, leading to an expansion error. It turns out
2213 that it is easier to change to an unlimited length (owing to other changes
2214 that have happened since this code was originally written) than to build
2215 structure for giving a limitation error. The length of the item that
2216 expands into the list of hosts is now unlimited.
2218 47. The lsearch lookup could not handle data where the length of text line was
2219 more than 4095 characters. Such lines were truncated, leading to shortened
2220 data being returned. It should now handle lines of any length.
2222 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2223 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2226 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2227 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2228 no longer happens in comment lines. A list of replacements is now placed
2229 at the head of all of the source files, except those whose only change is
2230 to replace PERL_COMMAND in the very first #! line.
2232 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2233 messages on the queue, with a bottom-up merge sort, using code contributed
2234 by Michael Haardt. This should make operations like -bp somewhat faster on
2235 large queues. It won't affect queue runners, except when queue_run_in_order
2238 51. Installed eximstats 1.31 in the distribution.
2240 52. Added support for SRV lookups to the dnslookup router.
2242 53. If an ACL referred to $message_body or $message_body_end, the value was not
2243 reset for any messages that followed in the same SMTP session.
2245 54. The store-handling optimization for building very long strings was not
2246 differentiating between the different store pools. I don't think this
2247 actually made any difference in practice, but I've tidied it.
2249 55. While running the routers to verify a sender address, $sender_address
2250 was still set to the sender address. This is wrong, because when routing to
2251 send a bounce to the sender, it would be empty. Therefore, I have changed
2252 it so that, while verifying a sender address, $sender_address is set to <>.
2253 (There is no change to what happens when verifying a recipient address.)
2255 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2256 target A or AAAA records (if not already returned) without resetting the
2257 qualify_single or search_parents options of the DNS resolver. These are
2258 inappropriate in this case because the targets of MX and SRV records must
2259 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2260 target that, when qualified, matched something in the local domain. These
2261 two options are now turned off when doing these lookups.
2263 57. It seems that at least some releases of Reiserfs (which does not have the
2264 concept of a fixed number of inodes) returns zero and not -1 for the
2265 number of available inodes. This interacted badly with check_spool_inodes,
2266 which assumed that -1 was the "no such thing" setting. What I have done is
2267 to check that the total number of inodes is greater than zero before doing
2268 the test of how many are available.
2270 58. When a "warn" ACL statement has a log_message modifier, the message is
2271 remembered, and not repeated. This is to avoid a lot of repetition when a
2272 message has many recipients that cause the same warning to be written.
2273 Howewer, Exim was preserving the list of already written lines for an
2274 entire SMTP session, which doesn't seem right. The memory is now reset if a
2275 new message is started.
2277 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2278 debugging output unless log_rewrite was also set.
2280 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2281 of (int)(handle) when we know that handle contains (void *)(-1).
2283 61. The Exim daemon panic-logs an error return when it closes the incoming
2284 connection. However "connection reset by peer" seems to be common, and
2285 isn't really an error worthy of noting specially, so that particular error
2288 62. When Exim is trying to find all the local interfaces, it used to panic and
2289 die if the ioctl to get the interface flags failed. However, it seems that
2290 on at least one OS (Solaris 9) it is possible to have an interface that is
2291 included in the list of interfaces, but for which you get a failure error
2292 for this call. This happens when the interface is not "plumbed" into a
2293 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2294 failure of the "get flags" call assumes that the interface is down.
2296 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2297 makes life easier for people who are doing arithmetic on fields extracted
2298 from dates, where you often get leading zeros that should not be
2299 interpreted as octal.
2301 64. Added qualify_domain to the redirect router, to override the global
2304 65. If a pathologically long header line contained very many addresses (the
2305 report of this problem mentioned 10 000) and each of them was rewritten,
2306 Exim could use up a very large amount of memory. (It kept on making new
2307 copies of the header line as it rewrote, and never released the old ones.)
2308 At the expense of a bit more processing, the header rewriting function has
2309 been changed so that it no longer eats memory in this way.
2311 66. The generation of the Received: header has been moved from the time that a
2312 message starts to be received, to the time that it finishes. The timestamp
2313 in the Received: header should now be very close to that of the <= log
2314 line. There are two side-effects of this change:
2316 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2317 logged header lines no longer include the local Received: line, because
2318 it has not yet been created. The same applies to a copy of the message
2319 that is returned to a non-SMTP sender when a message is rejected.
2321 (b) When a filter file is tested using -bf, no additional Received: header
2322 is added to the test message. After some thought, I decided that this
2325 This change does not affect the value of $received_for. It is still set
2326 after address rewriting, but before local_scan() is called.
2328 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2330 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2331 gave an unhelpful panic error message, and a defer error. I have managed to
2332 change this behaviour so that it now rejects any supplied certificate,
2333 which seems right, as the list of acceptable certificates is empty.
2335 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2336 gave an unhelpful defer error. I have not managed to make this reject any
2337 supplied certificates, but the error message it gives is "no certificate
2338 supplied", which is not helpful.
2340 70. exigrep's output now also includes lines that are not associated with any
2341 message, but which match the given pattern. Implemented by a patch from
2342 Martin Sluka, which also tidied up the Perl a bit.
2344 71. Recipient callout verification, like sender verification, was using <> in
2345 the MAIL FROM command. This isn't really the right thing, since the actual
2346 sender may affect whether the remote host accepts the recipient or not. I
2347 have changed it to use the actual sender in the callout; this means that
2348 the cache record is now keyed on a recipient/sender pair, not just the
2349 recipient address. There doesn't seem to be a real danger of callout loops,
2350 since a callout by the remote host to check the sender would use <>.
2351 [SEE ABOVE: changed after hitting problems.]
2353 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2354 temporary errors. However, in the case of such a code being given after
2355 the end of a data transmission (i.e. after ".") Exim was failing to write
2356 a retry record for the message. (Yes, there was some broken host that was
2357 actually sending 8xx at this point.)
2359 73. An unknown lookup type in a host list could cause Exim to panic-die when
2360 the list was checked. (An example that provoked this was putting <; in the
2361 middle of a list instead of at the start.) If this happened during a DATA
2362 ACL check, a -D file could be left lying around. This kind of configuration
2363 error no longer causes Exim to die; instead it causes a defer errror. The
2364 incident is still logged to the main and panic logs.
2366 74. Buglet left over from Exim 3 conversion. The message "too many messages
2367 in one connection" was written to the rejectlog but not the mainlog, except
2368 when address rewriting (yes!) was being logged.
2370 75. Added write_rejectlog option.
2372 76. When a system filter was run not as root (that is, when system_filter_user
2373 was set), the values of the $n variables were not being returned to the
2374 main process; thus, they were not subsequently available in the $sn
2377 77. Added +return_path_on_delivery log selector.
2379 78. A connection timeout was being treated differently from recipients deferred
2380 when testing hosts_max_try with a message that was older than the host's
2381 retry timeout. (The host should not be counted, thus allowing all hosts to
2382 be tried at least once before bouncing.) This may have been the cause of an
2383 occasionally reported bug whereby a message would remain on the queue
2384 longer than the retry timeout, but would be bounced if a delivery was
2385 forced. I say "may" because I never totally pinned down the problem;
2386 setting up timeout/retry tests is difficult. See also the next item.
2388 79. The ultimate address timeout was not being applied to errors that involved
2389 a combination of host plus message (for example, a timeout on a MAIL
2390 command). When an address resolved to a number of possible hosts, and they
2391 were not all tried for each delivery (e.g. because of hosts_max_try), a
2392 message could remain on the queue longer than the retry timeout.
2394 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2397 81. Fixed an obscure SMTP outgoing bug which required at least the following
2398 conditions: (a) there was another message waiting for the same server;
2399 (b) the server returned 5xx to all RCPT commands in the first message so
2400 that the message was not completed; (c) the server dropped the connection
2401 or gave a negative response to the RSET that Exim sends to abort the
2402 transaction. The observed case was a dropped connection after DATA that had
2403 been sent in pipelining mode. That is, the server had advertised PIPELINING
2404 but was not implementing it correctly. The effect of the bug was incorrect
2405 behaviour, such as trying another host, and this could lead to a crash.
2411 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2412 and daemon.c were passed as pointers to ints; they should have been
2413 pointers to socklen_t variables (which are typically unsigned ints).
2415 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2418 3. Fixed a really odd bug that affected only the testing scheme; patching a
2419 certain fixed string in the binary changed the value of another string that
2420 happened to be identical to the end of the original first string.
2422 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2423 name", it returns that address as the IP address. On some operating
2424 systems (e.g. Solaris), it also passes back the IP address string as the
2425 "host name". However, on others (e.g. Linux), it passes back an empty
2426 string. Exim wasn't checking for this, and was changing the host name to an
2427 empty string, assuming it had been canonicized.
2429 5. Although rare, it is permitted to have more than one PTR record for a given
2430 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2431 all the names associated with an address, because they do in Solaris.
2432 However, it seems that they do not in Linux for data that comes from the
2433 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2434 I found this out when I moved to a new Linux workstation and tried to run
2435 the Exim test suite.
2437 To get round this problem I have changed the code so that it now does its
2438 own call to the DNS to look up PTR records when searching for a host name.
2439 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2440 addresses that are only in /etc/hosts are still found.
2442 This behaviour is, however, controlled by an option called host_lookup_
2443 order, which defaults to "bydns:byaddr". If people want to use the other
2444 order, or indeed, just use one or the other means of lookup, they can
2445 specify it in this variable.
2447 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2448 some operating systems, this comes back from gethostbyaddr() as an empty
2449 string, and this is what Exim used to test for. However, it seems that in
2450 other systems, "." is yielded. Exim now tests for this case too.
2452 7. The values of check_spool_space and check_log_space are now held internally
2453 as a number of kilobytes instead of an absolute number of bytes. If a
2454 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2455 kilobyte. This means that much larger values can be stored.
2457 8. Exim monitor: an attempt to get the action menu when not actually pointing
2458 at a message produces an empty menu entitled "No message selected". This
2459 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2460 no entries in it ("Shell widget menu has zero width and/or height"). So I
2461 have added a single, blank menu entry in this case.
2463 9. Added ${quote_local_part.
2465 10. MIME decoding is now applied to the contents of Subject: header lines when
2468 11. Now that a reference to $sender_host_address automatically causes a reverse
2469 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2470 host lookup before query-style lookups in lists that might use this
2471 variable. This has therefore been abolished, and the "net-" prefix is no
2472 longer necessary for query-style lookups.
2474 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2475 have been a typo for LFLAGS, so it has been changed.
2477 13. The install script calls Exim with "-C /dev/null" in order to find the
2478 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2479 to be output. Howeve, since Exim outputs its version number before the
2480 error, it didn't break the script. It just looked ugly. I fixed this by
2481 always allowing "-C /dev/null" if the caller is root.
2483 14. Ignore overlarge ACL variable number when reading spool file - insurance
2484 against a later release with more variables having written the file.
2486 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2487 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2488 this, as well as the form without the "IPv6" on the front.
2490 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2491 OS/Makefile-Darwin file.
2493 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2494 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2496 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2497 variable. However, some OS use ints for inodes. I've added cast to long int
2498 to get rid of the compiler warning.
2500 19. I had forgotten to lock out "/../" in configuration file names when
2501 ALT_CONFIG_PREFIX was set.
2503 20. Routers used for verification do not need to specify transports. However,
2504 if such a router generated a host list, and callout was configured, Exim
2505 crashed, because it could not find a port number from the (non-existent)
2506 transport. It now assumes port 25 in this circumstance.
2508 21. Added the -t option to exigrep.
2510 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2511 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2512 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2513 them is set, LOOKUP_LSEARCH is forced.
2515 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2517 24. Added sender and host information to the "rejected by local_scan()" log
2518 line; previously there was no indication of these.
2520 25. Added .include_if_exists.
2522 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2523 Linux. It turns out that not all file systems support this. Apparently some
2524 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2525 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2526 not supported on the file descriptor, is now ignored when Exim is trying to
2527 sync a directory. This applies only to Linux.
2529 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2531 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2532 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2535 29. Added support for maildirsize files from supplied patch (modified a bit).
2537 30. The use of :fail: followed by an empty string could lead Exim to respond to
2538 sender verification failures with (e.g.):
2540 550 Verification failed for <xxx>
2541 550 Sender verify failed
2543 where the first response line was missing the '-' that indicates it is not
2544 the final line of the response.
2546 31. The loop for finding the name of the user that called Exim had a hardwired
2547 limit of 10; it now uses the value of finduser_retries, which is used for
2548 all other user lookups.
2550 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2553 33. Exim was neglecting to zero errno before one call of strtol() when
2554 expanding a string and expecting an integer value. On some systems this
2555 resulted in spurious "integer overflow" errors. Also, it was casting the
2556 result into an int without checking.
2558 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2559 did not work. The code looks as if it has *never* worked, though it appears
2560 to have been documented since at least releast 1.62. I have made it work.
2562 35. The "timeout_DNS" error in retry rules, also documented since at least
2563 1.62, also never worked. As it isn't clear exactly what this means, and
2564 clearly it isn't a major issue, I have abolished the feature by treating it
2565 as "timeout", and writing a warning to the main and panic logs.
2567 36. The display of retry rules for -brt wasn't always showing the error code
2570 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2571 timeout_connect_A, timeout_connect_MX.
2573 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2574 to the empty sender.
2576 39. The daemon was not analysing the content of -oX till after it had closed
2577 stderr and disconnected from the controlling terminal. This meant that any
2578 syntax errors were only noted on the panic log, and the return code from
2579 the command was 0. By re-arranging the code a little, I've made the
2580 decoding happen first, so such errors now appear on stderr, and the return
2581 code is 1. However, the actual setting up of the sockets still happens in
2582 the disconnected process, so errors there are still only recorded on the
2585 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2586 connections (as happens on some IP stacks) was logged at start up time as
2587 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2588 it from "IPv6 and IPv4", which means that two separate sockets are being
2591 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2592 says whether AF_INET or AF_INET6 was passed as an argument.
2594 42. Exiwhat output was messed up when time zones were included in log
2597 43. Exiwhat now gives more information about the daemon's listening ports,
2598 and whether -tls-on-connect was used.
2600 44. The "port" option of the smtp transport is now expanded.
2602 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2603 silently ignored. Now an error message is written to the main and panic
2606 46. There's a new ACL modifier called "logwrite" which writes to a log file
2607 as soon as it is encountered.
2609 47. Added $local_user_uid and $local_user_gid at routing time.
2611 48. Exim crashed when trying to verify a sender address that was being
2614 49. Exim was recognizing only a space character after ".include". It now also
2615 recognizes a tab character.
2617 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2618 extracting the relevant information from the specification. The man page no
2619 longer contains scrambled data for the -d option, and I've added a section
2620 at the front about calling Exim under different names.
2622 51. Added "extra_headers" argument to the "mail" command in filter files.
2624 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2627 53. Installed eximstats 1.29.
2629 54. Added transport_filter_timeout as a generic transport option.
2631 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2632 Cc: header lines. This was required by RFC 822, but it not required by RFC
2635 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2636 incoming messages that did not have them. Now it does so only if the
2637 message originates locally, that is, if there is no associated remote host
2638 address. When Resent- header lines are present, this applies to the Resent-
2639 lines rather than the non-Resent- lines.
2641 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2642 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2644 58. Messages for configuration errors now include the name of the main
2645 configuration file - useful now that there may be more than one file in a
2646 list (.included file names were always shown).
2648 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2649 for those rare installations that do not start the daemon as root or run it
2650 setuid root. I've cut out the call to initgroups() if the daemon is not
2653 60. The Exim user and group can now be bound into the binary as text strings
2654 that are looked up at the start of Exim's processing.
2656 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2658 62. Added $mailstore_basename variable.
2660 63. Installed patch to sieve.c from Michael Haardt.
2662 64. When Exim failed to open the panic log after failing to open the main log,
2663 the original message it was trying to log was written to stderr and debug
2664 output, but if they were not available (the usual case in production), it
2665 was lost. Now it is written to syslog before the two lines that record the
2666 failures to open the logs.
2668 65. Users' Exim filters run in subprocesses under the user's uid. It is
2669 possible for a "deliver" command or an alias in a "personal" command to
2670 provoke an address rewrite. If logging of address rewriting is configured,
2671 this fails because the process is not running as root or exim. There may be
2672 a better way of dealing with this, but for the moment (because 4.30 needs
2673 to be released), I have disabled address rewrite logging when running a
2674 filter in a non-root, non-exim process.
2680 1. The buildconfig auxiliary program wasn't quoting the value set for
2681 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2682 not defined. This bug was masked in 4.22 by the effect that was fixed in
2685 2. Some messages that were rejected after a message id was allocated were
2686 shown as "incomplete" by exigrep. It no longer does this for messages that
2687 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2689 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2690 have allow_domain_literals set, the ID did not get logged in the <= line.
2691 Domain literals are now always recognized in Message-ID: header lines.
2693 4. The first argument for a ${extract expansion item is the key name or field
2694 number. Leading and trailing spaces in this item were not being ignored,
2695 causing some misleading effects.
2697 5. When deliver_drop_privilege was set, single queue runner processes started
2698 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2699 same command in the process it spins off) were not dropping privilege.
2701 6. When the daemon running as "exim" started a queue runner, it always
2702 re-executed Exim in the spun-off process. This is a waste of effort when
2703 deliver_drop_privilege is set. The new process now just calls the
2704 queue-runner function directly.
2710 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2713 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2714 ignored. Though the use of -oP was forcing the writing of a pid file, it
2715 was always written to the default place.
2717 3. If the message "no IP address found for host xxxx" is generated during
2718 incoming verification, it is now followed by identification of the incoming
2719 connection (so you can more easily find what provoked it).
2721 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2723 5. Added some features to "harden" Exim a bit more against certain attacks:
2725 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2726 be put in Local/Makefile. This is like the never_users runtime option,
2727 but it cannot be overridden. The default setting is "root".
2729 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2730 prefix string with which any file named in a -C command line option
2733 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2734 is retained for -C and -D only if the caller of Exim is root. Without
2735 it, the exim user may also use -C and -D and retain privilege.
2737 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2738 command line option is disabled.
2740 6. Macro names set by the -D option must start with an upper case letter, just
2741 like macro names defined in the configuration file.
2743 7. Added "dereference=" facility to LDAP.
2745 8. Two instances of the typo "uknown" in the source files are fixed.
2747 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2748 the Configure-Makefile script screwed up while processing it.
2750 10. Incorporated PCRE 4.4.
2752 11. The SMTP synchronization check was not operating right at the start of an
2753 SMTP session. For example, it could not catch a HELO sent before the client
2754 waited for the greeting. There is now a check for outstanding input at the
2755 point when the greeting is written. Because of the duplex, asynchronous
2756 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2757 way, but not yet received, when the check is performed.
2759 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2760 on TCP/IP sockets, because this apparently causes some broken clients to
2763 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2764 unchanged) from the Cygwin maintainer.
2766 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2767 was supported instead of testing for mbx. Effectively a typo.
2769 15. The spa authenticator server code was not checking that the input it
2770 received was valid base64.
2772 16. The debug output line for the "set" modifier in ACLs was not showing the
2773 name of the variable that was being set.
2775 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2777 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2778 lookup on its own. Something else was needed to trigger the lookup. For
2779 example, a match in host_lookup or the need for a host name in a host list.
2780 Now, if $sender_host_name is referenced and the host name has not yet been
2781 looked up, a lookup is performed. If the lookup fails, the variable remains
2782 empty, and $host_lookup_failed is set to "1".
2784 19. Added "eqi" as a case-independent comparison operator.
2786 20. The saslauthd authentication condition could segfault if neither service
2787 nor realm was specified.
2789 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2790 error message that was logged was misleading, and incoming SMTP
2791 connections were dropped. The message is now more accurate, and temporary
2792 errors are given to SMTP connections.
2794 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2795 (except RSET) with a 421 error, until QUIT is received. However, it was
2796 failing to send a response to QUIT.
2798 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2799 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2800 ACL is now run after the helo_try_verify_hosts code.
2802 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2803 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2804 case-independent, so other case variants are also recognized). Apparently
2805 some systems use these upper case variants.
2807 25. If more than two messages were waiting for the same host, and a transport
2808 filter was specified for the transport, Exim sent two messages over the
2809 same TCP/IP connection, and then failed with "socket operation on non-
2810 socket" when it tried to send the third.
2812 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2814 27. The extern definition of crypt16() in expand.c was not being excluded when
2815 the OS had its own crypt16() function.
2817 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2818 as a preferred synonym for return_size_limit, both as an option and as an
2821 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2823 30. Changed the default configuration ACL to relax the local part checking rule
2824 for addresses that are not in any local domains. For these addresses,
2825 slashes and pipe symbols are allowed within local parts, but the sequence
2826 /../ is explicitly forbidden.
2828 31. SPA server authentication was not clearing the challenge buffer before
2831 32. log_message in a "warn" ACL statement was writing to the reject log as
2832 well as to the main log, which contradicts the documentation and doesn't
2833 seem right (because no rejection is happening). So I have stopped it.
2835 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2836 However, I am unable to do any testing of this.
2838 34. Fixed an infelicity in the appendfile transport. When checking directories
2839 for a mailbox, to see if any needed to be created, it was accidentally
2840 using path names with one or more superfluous leading slashes; tracing
2841 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2843 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2844 contents are added to the log line that is written for every discarded
2845 recipient. (Previously a log_message setting was ignored.)
2847 36. The ${quote: operator now quotes the string if it is empty.
2849 37. The install script runs exim in order to find its version number. If for
2850 some reason other than non-existence or emptiness, which it checks, it
2851 could not run './exim', it was installing it with an empty version number,
2852 i.e. as "exim-". This error state is now caught, and the installation is
2855 38. An argument was missing from the function that creates an error message
2856 when Exim fails to connect to the socket for saslauthd authentication.
2857 This could cause Exim to crash, or give a corrupted message.
2859 39. Added isip, isip4, and isip6 to ${if conditions.
2861 40. The ACL variables $acl_xx are now saved with the message, and can be
2862 accessed later in routers, transports, and filters.
2864 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2865 strings in the file are not string-expanded.
2867 42. If a MAIL command specified a SIZE value that was too large to fit into an
2868 int variable, the check against message_size_limit failed. Such values are
2869 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2870 day this will have to be increased, but I don't think I want to be around
2871 when emails are that large.
2878 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2879 iconv() is not standard in FreeBSD.
2881 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2882 IPv6 enabled. The observed symptom was a segmentation fault on return from
2883 the function os_common_find_running_interfaces() in src/os.c.
2885 3. In the check_special_case() function in daemon.c I had used "errno" as an
2886 argument name, which causes warnings on some systems. This was basically a
2887 typo, since it was named "eno" in the comments!
2889 4. The code that waits for the clock to tick (at a resolution of some fraction
2890 of a second) so as to ensure message-id uniqueness was always waiting for
2891 at least one whole tick, when it could have waited for less. [This is
2892 almost certainly not relevant at current processor speeds, where it is
2893 unlikely to ever wait at all. But we try to future-proof.]
2895 5. The function that sleeps for a time interval that includes fractions of a
2896 second contained a race. It did not block SIGALRM between setting the
2897 timer, and suspending (a couple of lines later). If the interval was short
2898 and the sigsuspend() was delayed until after it had expired, the suspension
2899 never ended. On busy systems this could lead to processes getting stuck for
2902 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2903 process, before it forks any delivery processes. The open lookup caching
2904 mechanism meant that the open file or database connection was passed into
2905 the delivery process. The problem was that delivery processes always tidy
2906 up cached lookup data. This could cause a problem for the next delivery
2907 process started by the queue runner, because the external queue runner
2908 process does not know about the closure. So the next delivery process
2909 still has data in the lookup cache. In the case of a file lookup, there was
2910 no problem because closing a file descriptor in a subprocess doesn't affect
2911 the parent. However, if the lookup was caching a connection to a database,
2912 the connection was closed, and the second delivery process was likely to
2913 see errors such as "PGSQL: query failed: server closed the connection
2914 unexpectedly". The problem has been fixed by closing all cached lookups
2915 in a queue runner before running a delivery process.
2917 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2918 seem to have the "const" qualifier which it has on other OS. I've
2921 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2922 *of the same type* (client or server) with the same public name that an
2923 error should be diagnosed.
2925 9. When Exim looked up a host name for an IP address, but failed to find the
2926 original IP address when looking up the host name (a safety check), it
2927 output the message "<ip address> does not match any IP for NULL", which was
2928 confusing, to say the least. The bug was that the host name should have
2929 appeared instead of "NULL".
2931 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2932 user that is built into the binary, and the -C or -D options is used, root
2933 privilege is dropped before the configuration file is read. In addition,
2934 logging is switched to stderr instead of the normal log files. If the
2935 configuration then re-defines the Exim user, the unprivileged environment
2936 is probably not what is expected, so Exim logs a panic warning message (but
2939 However, if deliver_drop_privilege is set, the unprivileged state may well
2940 be exactly what is intended, so the warning has been cut out in that case,
2941 and Exim is allowed to try to write to its normal log files.
2947 1. smtp_return_error_details was not giving details for temporary sender
2948 or receiver verification errors.
2950 2. Diagnose a configuration error if two authenticators have the same public
2953 3. Exim used not to create the message log file for a message until the first
2954 delivery attempt. This could be confusing when incoming messages were held
2955 for policy or load reasons. The message log file is now created at the time
2956 the message is received, and an initial "Received" line is written to it.
2958 4. The automatically generated man page for command line options had a minor
2959 bug that caused no ill effects; however, a more serious problem was that
2960 the procedure for building the man page automatically didn't always
2961 operate. Consequently, release 4.20 contains an out-of-date version. This
2962 shouldn't happen again.
2964 5. When building Exim with embedded Perl support, the script that builds the
2965 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2966 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2968 6. The freeze_tell option was not being used for messages that were frozen on
2969 arrival, either by an ACL or by local_scan().
2971 7. Added the smtp_incomplete_transaction log selector.
2973 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2974 was accepting AUTH without a new EHLO.
2976 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2977 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2980 10. Logging of TCP/IP connections (when configured) now happens in the main
2981 daemon process instead of the child process, so that the TCP/IP connection
2982 count is more accurate (but it can never be perfect).
2984 11. The use of "drop" in a nested ACL was not being handled correctly in the
2985 outer ACL. Now, if condition failure induced by the nested "drop" causes
2986 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2987 or "require"), the connection is dropped.
2989 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2990 that yield "discard" can now be used with an "accept" or a "discard" verb,
2991 but an error is generated for any others (because I can't see a useful way
2992 to define what should happen).
2994 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2995 are now processed in the same way as lines in the Exim configuration file.
2996 In particular, continuation lines are supported.
2998 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3000 15. Added -ti meaning -t -i.
3002 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3003 domains, and warn by logging if others are found.
3005 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3006 returned from GIFCONF when Exim is trying to find the list of interfaces on
3007 a host. The code in os.c has been modified to copy each ifreq to an aligned
3008 structure in all cases.
3010 Also, in some cases, the returned ifreq's were being copied to a 'struct
3011 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3012 means the last couple of bytes of an IPv6 address could be chopped if the
3013 ifreq contained only a normal sockaddr (14 bytes storage).
3015 18. Named domain lists were not supported in the hosts_treat_as_local option.
3016 An entry such as +xxxx was not recognized, and was treated as a literal
3019 19. Ensure that header lines added by a DATA ACL are included in the reject log
3020 if the ACL subsequently rejects the message.
3022 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3023 MD5 (which is deprecated).
3025 21. When testing a filter file using -bf, Exim was writing a message when it
3026 took the sender from a "From " line in the message, but it was not doing so
3027 when it took $return_path from a Return-Path: header line. It now does.
3029 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3030 with a valid header line field name (a series of printing characters
3031 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3033 23. Changed "disc" in the source to "disk" to conform to the documentation and
3034 the book and for uniformity.
3036 24. Ignore Sendmail's -Ooption=value command line item.
3038 25. When execve() failed while trying to run a command in a pipe transport,
3039 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3040 could be confused with a return value of 69 from the command itself. This
3041 has been changed to 127, the value the shell returns if it is asked to run
3042 a non-existent command. The wording for the related log line suggests a
3043 non-existent command as the problem.
3045 26. If received_header_text expands to an empty string, do not add a Received:
3046 header line to the message. (Well, it adds a token one on the spool, but
3047 marks it "old" so that it doesn't get used or transmitted.)
3049 27. Installed eximstats 1.28 (addition of -nt option).
3051 28. There was no check for failure on the call to getsockname() in the daemon
3052 code. This can fail if there is a shortage of resources on the system, with
3053 ENOMEM, for example. A temporary error is now given on failure.
3055 29. Contrary to the C standard, it seems that in some environments, the
3056 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3057 program. Exim now does this explicitly; it affects the formatting of
3058 timestamps using strftime().
3060 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3061 complaints. I've now initialized all the variables, to avoid this.
3063 32. Header lines added by a system filter were not being "seen" during
3064 transport-time rewrites.
3066 33. The info_callback() function passed to OpenSSL is set up with type void
3067 (*)(SSL *, int, int), as described somewhere. However, when calling the
3068 function (actually a macro) that sets it up, the type void(*)() is
3069 expected. I've put in a cast to prevent warnings from picky compilers.
3071 34. If a DNS black list lookup found a CNAME record, but there were no A
3072 records associated with the domain it pointed at, Exim crashed.
3074 35. If a DNS black list lookup returned more than one A record, Exim ignored
3075 all but the first. It now scans all returned addresses if a particular IP
3076 value is being sought. In this situation, the contents of the
3077 $dnslist_value variable are a list of all the addresses, separated by a
3080 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3081 to accept a host name and all its aliases if the forward lookup for any of
3082 them yielded the IP address of the incoming connection. Now it accepts only
3083 those names whose forward lookup yields the correct IP address. Any other
3084 names are discarded. This closes a loophole whereby a rogue DNS
3085 administrator could create reverse DNS records to break through a
3086 wildcarded host restriction in an ACL.
3088 37. If a user filter or a system filter that ran in a subprocess used any of
3089 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3090 the wrong values were passed to the pipe command ($thisaddress had the
3091 value of $0, $0 had the value of $1, etc). This bug was introduced by
3092 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3094 38. Improved the line breaking for long SMTP error messages from ACLs.
3095 Previously, if there was no break point between 40 and 75 characters, Exim
3096 left the rest of the message alone. Two changes have been made: (a) I've
3097 reduced the minimum length to 35 characters; (b) if it can't find a break
3098 point between 35 and 75 characters, it looks ahead and uses the first one
3099 that it finds. This may give the occasional overlong line, but at least the
3100 remaining text gets split now.
3102 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3103 file descriptors might be low, and that setting 1000 would always raise it.
3104 It turns out that in some environments, the limit is already over 1000 and
3105 that lowering it causes trouble. So now Exim takes care not to decrease it.
3107 40. When delivering a message, the value of $return_path is set to $sender_
3108 address at the start of routing (routers may change the value). By an
3109 oversight, this default was not being set up when an address was tested by
3110 -bt or -bv, which affected the outcome if any router or filter referred to
3113 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3114 log only when "message" or "log_message" are set. However, if one of the
3115 conditions was an address verification, or a call to a nested ACL, the
3116 messages generated by the underlying test were being passed through. This
3117 no longer happens. The underlying message is available in $acl_verify_
3118 message for both "message" and "log_message" expansions, so it can be
3119 passed through if needed.
3121 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3122 new expansion $bh_ to give the encoded byte string without charset
3123 translation. Translation happens only if iconv() is available; HAVE_ICONV
3124 indicates this at build time. HEADERS_CHARSET gives the charset to
3125 translate to; headers_charset can change it in the configuration, and
3126 "headers charset" can change it in an individual filter file.
3128 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3129 that creates RFC 2047 encoded "words" labels them as that charset instead
3130 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3131 expansion operator; (ii) when Exim creates a From: line for a local
3132 message; (iii) when a header line is rewritten to include a "phrase" part.
3134 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3135 buggy, causing it to skip the first lines of messages whose message ID
3136 ended in 'D'. This would not have bitten before Exim release 4.14, because
3137 message IDs were unlikely to end in 'D' before then. The effect was to have
3138 incorrect size information for certain domains.
3140 45. #include "config.h" was missing at the start of the crypt16.c module. This
3141 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3144 46. If there was a timeout during a "random" callout check, Exim treated it as
3145 a failure of the random address, and carried on sending RSET and the real
3146 address. If the delay was just some slowness somewhere, the response to the
3147 original RCPT would be taken as a response to RSET and so on, causing
3148 mayhem of various kinds.
3150 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3151 when I implemented it. It didn't allow for the fact that some option values
3152 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3153 the right test for positive values.
3155 48. Domain names in DNS records are case-independent. Exim always looks them up
3156 in lower case. Some resolvers return domain names in exactly the case they
3157 appear in the zone file, that is, they may contain uppercase letters. Not
3158 all resolvers do this - some return always lower case. Exim was treating a
3159 change of case by a resolver as a change of domain, similar to a widening
3160 of a domain abbreviation. This triggered its re-routing code and so it was
3161 trying to route what was effectively the same domain again. This normally
3162 caused routing to fail (because the router wouldn't handle the domain
3163 twice). Now Exim checks for this case specially, and just changes the
3164 casing of the domain that it ultimately uses when it transmits the message
3167 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3170 50. If a filter generated a file delivery with a non-absolute name (possible if
3171 no home directory exists for the router), the forbid_file option was not
3174 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3175 the existing equality matching.
3177 52. Exim was using ints instead of ino_t variables in some places where it was
3178 dealing with inode numbers.
3180 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3181 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3182 called TMPDIR, and if it finds it is different, it changes its value.
3184 54. The smtp_printf() function is now made available to local_scan() so
3185 additional output lines can be written before returning. There is also an
3186 smtp_fflush() function to enable the detection of a dropped connection.
3187 The variables smtp_input and smtp_batched_input are exported to
3190 55. Changed the default runtime configuration: the message "Unknown user"
3191 has been removed from the ACL, and instead placed on the localuser router,
3192 using the cannot_route_message feature. This means that any verification
3193 failures that generate their own messages won't get overridden. Similarly,
3194 the "Unrouteable address" message that was in the ACL for unverifiable
3195 relay addresses has also been removed.
3197 56. Added hosts_avoid_esmtp to the smtp transport.
3199 57. The exicyclog script was not checking for the esoteric option
3200 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3201 will work only if exicyclog is run under the appropriate euid.
3203 58. Following a discussion on the list, the rules by which Exim recognises line
3204 endings on incoming messages have been changed. The -dropcr and drop_cr
3205 options are now no-ops, retained only for backwards compatibility. The
3206 following line terminators are recognized: LF CRLF CR. However, special
3207 processing applies to CR:
3209 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3210 nor a local message in the state where . is a terminator.
3212 (ii) If a bare CR is encountered in a header line, an extra space is added
3213 after the line terminator so as not to end the header. The reasoning
3214 behind this is that bare CRs in header lines are most likely either
3215 to be mistakes, or people trying to play silly games.
3217 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3218 was being incorrectly given as 18 bytes larger than it should have been.
3219 This is a VOB (very old bug).
3221 60. This may never have affected anything current, but just in case it has:
3222 When the local host is found other than at the start of a list of hosts,
3223 the local host, those with the same MX, and any that follow, are discarded.
3224 When the list in question was part of a longer list of hosts, the following
3225 hosts (not currently being processed) were also being discarded. This no
3226 longer happens. I'm not sure if this situation could ever has previously
3229 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3232 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3233 header. This is something that is recommended in a new Internet Draft, and
3234 is something that is documented as being done by Sendmail. There are two
3235 possible values. For messages generated by the autoreply transport, Exim
3238 Auto-Submitted: auto-replied
3240 whereas for all other generated messages (e.g. bounces) it adds
3242 Auto-Submitted: auto-generated
3244 63. The "personal" condition in filters now includes a test for the
3245 Auto-Submitted: header. If it contains the string "auto-" the message it
3246 not considered personal.
3248 64. Added rcpt_include_affixes as a generic transport option.
3250 65. Added queue_only_override (default true).
3252 66. Added the syslog_duplication option.
3254 67. If what should have been the first header line of a message consisted of
3255 a space followed by a colon, Exim was mis-interpreting it as a header line.
3256 It isn't of course - it is syntactically invalid and should therefore be
3257 treated as the start of the message body. The misbehaviour could have
3258 caused a number of strange effects, including loss of data in subsequent
3259 header lines, and spool format errors.
3261 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3262 client host had authenticated. This control can now be exercised by an ACL
3263 for more flexibility.
3265 69. By default, callouts do not happen when testing with -bh. There is now a
3266 variant, -bhc, which does actually run the callout code, including
3267 consulting and updating the callout cache.
3269 70. Added support for saslauthd authentication, courtesy of Alexander
3272 71. If statvfs() failed on the spool or log directories while checking their
3273 size for availability, Exim confusingly gave the error "space shortage".
3274 Furthermore, in debugging mode it crashed with a floating point exception.
3275 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3276 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3277 serious problem, Exim now writes to the main and panic logs when this
3278 happens, with details of the failure. It then refuses to accept the
3279 incoming message, giving the message "spool directory problem" or "log
3280 directory problem" with a 421 code for SMTP messages.
3282 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3283 0, 1, and 2 exist, because some OS complain for execs without them (see
3284 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3285 descriptors. However, the code omitted to check that the open succeeded,
3286 causing mysterious errors if for some reason the permissions on /dev/null
3287 got screwed. Now Exim writes a message to the main and panic logs, and
3288 bombs out if it can't open /dev/null.
3290 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3291 interact so that it is all more flexible. It is supposed to remain
3292 backwards compatible. Also added extra_local_interfaces.
3294 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3295 to bomb out with an assertion failure - to the client this appears as a
3296 connection drop. This problem occurs in the part of the code that was taken
3297 from the Samba project. Fortunately, the assertion is in a very simple
3298 function, so I have fixed this by reproducing the function inline in the
3299 one place where it is called, and arranging for authentication to fail
3300 instead of killing the process with assert().
3302 75. The SPA client code was not working when the server requested OEM rather
3303 than Unicode encoding.
3305 76. Added code to make require_files with a specific uid setting more usable in
3306 the case where statting the file as root fails - usually a non-root-mounted
3307 NFS file system. When this happens and the failure is EACCES, Exim now
3308 forks a subprocess and does the per-uid checking as the relevant uid.
3310 77. Added process_log_path.
3312 78. If log_file_path was not explicitly set, a setting of check_log_space or
3313 check_log_inodes was ignored.
3315 79. If a space check for the spool or log partitions fails, the incident is now
3316 logged. Of course, in the latter case the data may get lost...
3318 80. Added the %p formatting code to string_format() so that it can be used to
3319 print addresses in debug_print(). Adjusted all the address printing in the
3320 debugging in store.c to use %p rather than %d.
3322 81. There was a concern that a line of code in smtp_in.c could overflow a
3323 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3324 initially expressed, the concern was not well-founded, because trailing
3325 spaces are removed early. However, if the trailing spaces were followed by
3326 a NULL, they did not get removed, so the overflow was possible. Two fixes
3329 (a) I re-wrote the offending code in a cleaner fashion.
3330 (b) If an incoming SMTP command contains a NULL character, it is rejected
3333 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3334 runs initgroups(), so that if the Exim user is in any additional groups,
3335 they will be used during message reception.
3341 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob