1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.337 2006/04/04 09:09:44 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 The code for finding all the local interface addresses on a FreeBSD
10 system running IPv6 was broken. This may well have applied to all BSD
11 systems, as well as to others that have similar system calls. The broken
12 code found IPv4 interfaces correctly, but gave incorrect values for the
13 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
14 that it would not match correctly against @[] and not recognize the IPv6
17 PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
20 PH/03 Added disable_ipv6.
22 PH/04 Changed $reply_address to use the raw form of the headers instead of the
23 decoded form, because it is most often used to construct To: headers
24 lines in autoreplies, and the decoded form may well be syntactically
25 invalid. However, $reply_address has leading white space removed, and all
26 newlines turned into spaces so that the autoreply transport does not
29 PH/05 If group was specified without a user on a router, and no group or user
30 was specified on a transport, the group from the router was ignored.
32 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
33 for visible compile-time settings that can be used to change these
34 numbers, for those that want even more. Backwards compatibility with old
35 spool files has been maintained. However, going back to a previous Exim
36 release will lost any variables that are in spool files.
38 PH/07 Two small changes when running in the test harness: increase delay when
39 passing a TCP/IP connection to a new process, in case the original
40 process has to generate a bounce, and remove special handling of
41 127.0.0.2 (sic), which is no longer necessary.
43 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
44 be the same on different OS.
46 PH/09 Moved a debug statement in filter processing to avoid a race problem when
49 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
50 whether --show-vars was specified or not
52 JJ/02 exipick: Added support for new ACL variable spool format introduced
55 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
56 syntactically invalid From: or Reply-to: line, and a filter used this to
57 generate an autoreply, and therefore failed to obtain an address for the
58 autoreply, Exim could try to deliver to a non-existent relative file
59 name, causing unrelated and misleading errors. What now happens is that
60 it logs this as a hard delivery error, but does not attempt to create a
63 PH/11 The exinext utility has a -C option for testing purposes, but although
64 the given file was scanned by exinext itself; it wasn't being passed on
67 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
68 an end-of-file indication when reading a command response.
70 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
71 compiled. In many other places in Exim, IPv6 addresses are always
72 recognized, so I have changed this. It also means that IPv4 domain
73 literals of the form [IPV4:n.n.n.n] are now always recognized.
75 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
76 used if the router is not running as root, for example, when verifying at
77 ACL time, or when using -bh. The debugging output from this situation was
78 non-existent - all you got was a failure to exec. I have made two
81 (a) Failures to set uid/gid, the current directory, or a process leader
82 in a subprocess such as that created by queryprogram now generate
83 suitable debugging ouput when -d is set.
85 (b) The queryprogram router detects when it is not running as root,
86 outputs suitable debugging information if -d is set, and then runs
87 the subprocess without attempting to change uid/gid.
89 PH/15 Minor change to Makefile for building test_host (undocumented testing
92 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
93 additional section of a DNS packet that returns MX or SRV records.
94 Instead, it always explicitly searches for A/AAAA records. This avoids
95 major problems that occur when a DNS server includes only records of one
96 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
97 fixed another bug: if SRV records were looked up and the corresponding
98 address records were *not* found in the additional section, the port
99 values from the SRV records were lost.
101 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
102 using the correct key (the original address) when searching the retry
103 rules in order to find which one to use for generating the retry hint.
105 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
106 adding the default one. Similarly, if it contains a Reply-To: header, the
107 errors_reply_to option, if set, is not used.
109 PH/19 When calculating a retry time, Exim used to measure the "time since
110 failure" by looking at the "first failed" field in the retry record. Now
111 it does not use this if it is later than than the arrival time of the
112 message. Instead it uses the arrival time. This makes for better
113 behaviour in cases where some deliveries succeed, thus re-setting the
114 "first failed" field. An example is a quota failure for a huge message
115 when small messages continue to be delivered. Without this change, the
116 "time since failure" will always be short, possible causing more frequent
117 delivery attempts for the huge message than are intended.
119 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
120 $1, $2, $3) because the numerical variables can be reset during some
121 expansion items (e.g. "match"), thereby losing the authentication data.
123 PH/21 Make -bV show the size of off_t variables so that the test suite can
124 decide whether to run tests for quotas > 2G.
126 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
127 mailbox_size, and mailbox_filecount in the appendfile transport. If a
128 filecount value is greater than 2G or if a quota value is greater than 2G
129 on a system where the size of off_t is not greater than 4, a panic error
132 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
133 never match. The debug and -bh output now contains an explicit error
134 message indicating a malformed IPv4 address or mask.
136 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
137 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
140 PH/25 Do not write to syslog when running in the test harness. The only
141 occasion when this arises is a failure to open the main or panic logs
142 (for which there is an explicit test).
144 PH/26 Added the /no_tell option to "control=freeze".
146 PH/27 If a host name lookup failed very early in a connection, for example, if
147 the IP address matched host_lookup and the reverse lookup yielded a name
148 that did not have a forward lookup, an error message of the form "no IP
149 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
150 could be logged. Now it outputs the IP address instead of "NULL".
152 PH/28 An enabling patch from MH: add new function child_open_exim2() which
153 allows the sender and the authenticated sender to be set when
154 submitting a message from within Exim. Since child_open_exim() is
155 documented for local_scan(), the new function should be too.
157 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
158 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
159 results in an empty string is now treated as unset.
161 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
163 PH/31 Added sender_verify_fail logging option.
165 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
166 needed by RFC 822 but not by RFC 2822 was commented out. I have now
167 tidied the source and removed it altogether.
169 PH/33 When a queue run was abandoned because the load average was too high, a
170 log line was always written; now it is written only if the queue_run log
171 selector is set. In addition, the log line for abandonment now contains
172 information about the queue run such as the pid. This is always present
173 in "start" and "stop" lines but was omitted from the "abandon" line.
175 PH/34 Omit spaces between a header name and the colon in the error message that
176 is given when verify = headers_syntax fails (if there are lots of them,
177 the message gets confusing).
179 PH/35 Change the default for dns_check_names_pattern to allow slashes within
180 names, as there are now some PTR records that contain slashes. This check
181 is only to protect against broken name servers that fall over on strange
182 characters, so the fact that it applies to all lookups doesn't matter.
184 PH/36 Now that the new test suite is complete, we can remove some of the
185 special code in Exim that was needed for the old test suite. For example,
186 sorting DNS records because real resolvers return them in an arbitrary
187 order. The new test suite's fake resolver always returns records in the
190 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
191 bounces) except when queue_only is set, to avoid logging races between
192 the different processes.
194 PH/38 Panic-die if .include specifies a non-absolute path.
196 PH/39 A tweak to the "H" retry rule from its user.
198 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
199 a label. They prevented compilation on older perls.
201 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
202 a warning to be raised on newish perls.
204 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
205 on queue. Changes to match documented behaviour of showing count of
206 messages matching specified criteria.
208 PH/40 Changed the default ident timeout from 30s to 5s.
210 PH/41 Added support for the use of login_cap features, on those BSD systems
211 that have them, for controlling the resources used by pipe deliveries.
213 PH/42 The content-scanning code uses fopen() to create files in which to put
214 message data. Previously it was not paying any attention to the mode of
215 the files. Exim runs with umask(0) because the rest of the code creates
216 files with open(), and sets the required mode explicitly. Thus, these
217 files were ending up world-writeable. This was not a big issue, because,
218 being within the spool directory, they were not world-accessible. I have
219 created a function called modefopen, which takes an additional mode
220 argument. It sets umask(777), creates the file, chmods it to the required
221 mode, then resets the umask. All the relevant calls to fopen() in the
222 content scanning code have been changed to use this function.
224 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
225 to 24 hours. This avoids potential overflow problems when processing G
226 and H retry rules. I suspect nobody ever tinkers with this value.
228 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
230 PH/45 When the plaintext authenticator is running as a client, the server's
231 challenges are checked to ensure they are valid base64 strings. By
232 default, the authentication attempt is cancelled if an invalid string is
233 received. Setting client_ignore_invalid_base64 true ignores these errors.
234 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
235 they are received. Thus, the responses can be made to depend on the
236 challenges. If an invalid string is ignored, an empty string is placed in
239 PH/46 Messages that are created by the autoreply transport now contains a
240 References: header, in accordance with RFCs 2822 and 3834.
242 PH/47 Added authenticated_sender_force to the smtp transport.
244 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
246 PH/49 Installed latest patch from the Sieve maintainer.
248 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
249 was also set, the appendfile transport was unnecessarily scanning a
250 directory of message files (e.g. for maildir delivery) to find the count
251 of files (along with the size), even though it did not need this
252 information. It now does the scan only if it needs to find either the
253 size of the count of files.
255 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
257 PH/52 Two bugs concerned with error handling when the smtp transport is
260 (i) Exim was not creating retry information for temporary errors given
261 for individual recipients after the DATA command when the smtp transport
262 was used in LMTP mode. This meant that they could be retried too
263 frequently, and not timed out correctly.
265 (ii) Exim was setting the flag that allows error details to be returned
266 for LMTP errors on RCPT commands, but not for LMTP errors for individual
267 recipients that were returned after the DATA command.
269 PH/53 This is related to PH/52, but is more general: for any failing address,
270 when detailed error information was permitted to be returned to the
271 sender, but the error was temporary, then after the final timeout, only
272 "retry timeout exceeded" was returned. Now it returns the full error as
273 well as "retry timeout exceeded".
275 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
276 do this, and (what is worse) MTAs that accept it.
278 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
279 will now be deprecated.
281 PH/56 New os.c-cygwin from the Cygwin maintainer.
283 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
284 formats (previously only available in exim formats via -bpr, -bpru,
285 and -bpra. Now also available in native and exiqgrep formats)
287 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
288 with very large, slow to parse queues
290 JJ/08 exipick: added ! as generic prefix to negate any criteria format
292 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
294 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
295 responses to authentication challenges, though it was showing the
296 challenges; (ii) I've removed the CR characters from the debug output for
299 PH/58 Allow for the insertion of a newline as well as a space when a string
300 is turned into more than one encoded-word during RFC 2047 encoding. The
301 Sieve code now uses this.
303 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
304 data_4xx, lost_connection, tls_required.
306 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
307 message was being sent as an SMTP response.
309 PH/61 Add -l and -k options to exicyclog.
311 PH/62 When verifying, if an address was redirected to one new address, so that
312 verification continued, and the new address failed or deferred after
313 having set something in $address_data, the value of $address_data was not
314 passed back to the ACL. This was different to the case when no
315 redirection occurred. The value is now passed back in both cases.
317 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
318 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
319 use login_cap.h, so on its own it isn't the distinguishing feature. The
320 new name refers directly to the setclassresources() function.
322 PH/65 Added configuration files for NetBSD3.
324 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
326 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
327 is preferred over IPv4.
329 PH/68 The bounce_return_message and bounce_return_body options were not being
330 honoured for bounces generated during the reception of non-SMTP messages.
331 In particular, this applied to messages rejected by the ACL. This bug has
332 been fixed. However, if bounce_return_message is true and bounce_return_
333 body is false, the headers that are returned for a non-SMTP message
334 include only those that have been read before the error was detected.
335 (In the case of an ACL rejection, they have all been read.)
337 PH/69 The HTML version of the specification is now built in a directory called
338 spec_html instead of spec.html, because the latter looks like a path with
339 a MIME-type, and this confuses some software.
341 PH/70 Catch two compiler warnings in sieve.c.
343 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
344 function verify_get_ident() calls ip_connect() to connect a socket, but
345 if the "connect()" function timed out, ip_connect() used to close the
346 socket. However, verify_get_ident() also closes the socket later, and in
347 between Exim writes to the log, which may get opened at this point. When
348 the socket was closed in ip_connect(), the log could get the same file
349 descriptor number as the socket. This naturally causes chaos. The fix is
350 not to close the socket in ip_connect(); the socket should be closed by
351 the function that creates it. There was only one place in the code where
352 this was missing, in the iplookup router, which I don't think anybody now
353 uses, but I've fixed it anyway.
359 PH/01 Two changes to the default runtime configuration:
361 (1) Move the checks for relay_from_hosts and authenticated clients from
362 after to before the (commented out) DNS black list checks.
364 (2) Add control=submission to the relay_from_hosts and authenticated
365 clients checks, on the grounds that messages accepted by these
366 statements are most likely to be submissions.
368 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
370 (1) Generate an error if the third argument for the ${prvs expansion is
373 (2) Treat a missing third argument of ${prvscheck as if it were an empty
376 (3) Reset the variables that are obtained from the first argument of
377 ${prvscheck and used in the second argument before leaving the code,
378 because their memory is reclaimed, so using them afterwards may do
381 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
382 one (it's much easier than Tom thought :-).
384 (5) Because of (4), we can now allow for the use of $prvscheck_result
385 inside the third argument.
387 PH/03 For some reason, the default setting of PATH when running a command from
388 a pipe transport was just "/usr/bin". I have changed it to
391 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
392 anything to be listed in the output from -bV.
394 PH/05 When a filter generated an autoreply, the entire To: header line was
395 quoted in the delivery log line, like this:
397 => >A.N.Other <ano@some.domain> <original@ddress> ...
399 This has been changed so that it extracts the operative address. There
400 may be more than one such address. If so, they are comma-separated, like
403 => >ano@some.domain,ona@other.domain <original@ddress> ...
405 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
406 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
407 not being looked up in the rDNS to get a host name, Exim was showing the
408 IP address twice in Received: lines, even though the IP addresses were
409 identical. For example:
411 Received: from [1.2.3.4] (helo=[1.2.3.4])
413 However, if the real host name was known, it was omitting the HELO data
414 if it matched the actual IP address. This has been tidied up so that it
415 doesn't show the same IP address twice.
417 PH/07 When both +timestamp and +memory debugging was on, the value given by
418 $tod_xxx expansions could be wrong, because the tod_stamp() function was
419 called by the debug printing, thereby overwriting the timestamp buffer.
420 Debugging no longer uses the tod_stamp() function when +timestamp is set.
422 PH/08 When the original message was included in an autoreply transport, it
423 always said "this is a copy of the message, including all the headers",
424 even if body_only or headers_only was set. It now gives an appropriate
427 PH/09 Applied a patch from the Sieve maintainer which:
429 o fixes some comments
430 o adds the (disabled) notify extension core
431 o adds some debug output for the result of if/elsif tests
432 o points to the current vacation draft in the documentation
433 and documents the missing references header update
437 o fixes a bug in processing the envelope test (when testing
438 multiple envelope elements, the last element determinted the
441 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
442 Electronic Mail") by including:
444 Auto-submitted: auto-generated
446 in the messages that it generates (bounce messages and others, such as
447 warnings). In the case of bounce messages for non-SMTP mesages, there was
448 also a typo: it was using "Auto_submitted" (underscore instead of
449 hyphen). Since every message generated by Exim is necessarily in response
450 to another message, thes have all been changed to:
452 Auto-Submitted: auto-replied
454 in accordance with these statements in the RFC:
456 The auto-replied keyword:
458 - SHOULD be used on messages sent in direct response to another
459 message by an automatic process,
461 - MUST NOT be used on manually-generated messages,
463 - MAY be used on Delivery Status Notifications (DSNs) and Message
464 Disposition Notifications (MDNs),
466 - MUST NOT be used on messages generated by automatic or periodic
467 processes, except for messages which are automatic responses to
470 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
471 to the default Received: header definition.
473 PH/12 Added log selector acl_warn_skipped (default on).
475 PH/13 After a successful wildlsearch lookup, discard the values of numeric
476 variables because (a) they are in the wrong storage pool and (b) even if
477 they were copied, it wouldn't work properly because of the caching.
479 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
480 checking when decoding. Apparently there are clients that generate
481 overlong encoded strings. Why am I not surprised?
483 PH/15 If the first argument of "${if match_address" was not empty, but did not
484 contain an "@" character, Exim crashed. Now it writes a panic log message
485 and treats the condition as false.
487 PH/16 In autoreply, treat an empty string for "once" the same as unset.
489 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
490 extension "envelope-auth". The code is finished and in agreement with
491 other implementations, but there is no documentation so far and in fact,
492 nobody wrote the draft yet. This extension is currently #undef'ed, thus
493 not changing the active code.
495 Print executed "if" and "elsif" statements when debugging is used. This
496 helps a great deal to understand what a filter does.
498 Document more things not specified clearly in RFC3028. I had all this
499 sorted out, when out of a sudden new issues came to my mind. Oops."
501 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
504 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
505 canonical form (as documented). However, after a host name lookup from
506 the IP address, check_host() was doing a simple string comparison with
507 addresses acquired from the DNS when checking that the found name did
508 have the original IP as one of its addresses. Since any found IPv6
509 addresses are likely to be in abbreviated form, the comparison could
510 fail. Luckily, there already exists a function for doing the comparison
511 by converting both addresses to binary, so now that is used instead of
514 PH/20 There was another similar case to PH/19, when a complete host name was
515 given in a host list; looking up its IP address could give an abbreviated
516 form, whereas the current host's name might or might not be abbreviated.
517 The same fix has been applied.
523 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
524 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
527 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
528 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
530 PH/03 Typo: missing ".o" in src/pcre/Makefile.
532 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
533 header line, restrict the check to what is listed in RFCs 2369 and 2929.
534 Also, for "Auto-Submitted", treat anything other than "no" as
535 non-personal, in accordance with RFC 3834. (Previously it treated
536 anything starting "auto-" as non-personal.)
538 TF/01 The control=submission/name=... option had a problem with syntax
539 errors if the name included a slash character. The /name= option
540 now slurps the rest of the string, so it can include any characters
541 but it must come last in the list of options (after /sender_retain
544 PH/05 Some modifications to the interface to the fake nameserver for the new
552 TK/01 Added the "success_on_redirect" address verification option. See
553 NewStuff for rationale and an example.
555 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
557 PH/02 Patch to exigrep to allow it to work on syslog lines.
559 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
560 fread() to skip over the body file's header line, because in Cygwin the
561 header line is locked and is inaccessible.
563 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
564 co-exist for some time) to make it clear that it is the Exim ID that is
565 referenced, not the Message-ID: header line.
567 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
568 string_format() function, because snprintf() does not exist on all
571 PH/06 The use of forbid_filter_existstest now also locks out the use of the
572 ${stat: expansion item.
574 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
575 protocol synchronization error", to keep the pedants happy.
577 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
578 well as for IRIX systems, when gcc is being used. See the host.c source
581 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
583 PH/10 Named domain lists were not working if used in a queue_smtp_domains
586 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
587 transport and to the smtp transport in LMTP mode.
589 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
591 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
592 run a filter in a subprocess. This could lead to confusion in subsequent
593 lookups in the parent process. There should also be a search_tidyup() at
594 the end of the subprocess.
596 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
597 only if the host matched helo_try_verify_hosts, which caused the
598 verification to occur when the EHLO/HELO command was issued. The ACL just
599 tested the remembered result. Now, if a previous verification attempt has
600 not happened, "verify = helo" does it there and then.
602 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
604 TK/03 Fix log output including CR from clamd.
606 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
607 exist provoked a memory error which could cause a segfault.
609 PH/15 Installed PCRE 6.2
611 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
613 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
614 of the problem. Specifically, suggested +O2 rather than +O1 for the
617 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
619 PH/20 If a delivery was routed to a non-standard port by means of an SRV
620 record, the port was not correctly logged when the outgoing_port log
621 selector was set (it logged the transort's default port).
623 PH/21 Added support for host-specific ports to manualroute, queryprogram,
624 fallback_hosts, and "hosts" in the smtp transport.
626 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
627 host errors such as "Connection refused".
629 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
630 authentication with radiusclient 0.4.9:
632 - Error returned from rc_read_config was caught wrongly
633 - Username/password not passed on to radius server due to wrong length.
635 The presumption is that some radiusclient API changes for 4.51/PH/17
636 were not taken care of correctly. The code is still untested by me (my
637 Linux distribution still has 0.3.2 of radiusclient), but it was
638 contributed by a Radius user.
640 PH/24 When doing a callout, the value of $domain wasn't set correctly when
641 expanding the "port" option of the smtp transport.
643 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
644 while reading a MIME header. Thanks to Tom Hughes for a patch.
646 PH/24 Include config.h inside local_scan.h so that configuration settings are
649 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
650 that in an ACL for RCPT (for example), you can examine exactly what was
653 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
654 commands, but it was not correctly comparing the address with the actual
655 client host address. Thus, it would show the EHLO address in Received:
656 header lines when this was not necessary.
658 PH/27 Added the % operator to ${eval:}.
660 PH/28 Exim tries to create and chdir to its spool directory when it starts;
661 it should be ignoring failures (because with -C, for example, it has lost
662 privilege). It wasn't ignoring creation failures other than "already
665 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
668 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
669 because an input line was too long, either on its own, or by virtue of
670 too many continuations, the temporary file was not being removed, and the
671 return code was incorrect.
673 PH/31 Missing "BOOL" in function definition in filtertest.c.
675 PH/32 Applied Sieve patches from the maintainer.
677 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
679 PH/33 Added "verify = not_blind".
681 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
682 Local/Makefile (with some defaults set). These are used in built scripts
683 such as exicyclog, but they have never been used in the exim_install
684 script (though there are many overriding facilities there). I have
685 arranged that the exim_install script now takes note of these two
688 PH/35 Installed configuration files for Dragonfly.
690 PH/36 When a locally submitted message by a trusted user did not contain a
691 From: header, and the sender address was obtained from -f or from an SMTP
692 MAIL command, and the trusted user did not use -F to supply a sender
693 name, $originator_name was incorrectly used when constructing a From:
694 header. Furthermore, $originator_name was used for submission mode
695 messages from external hosts without From: headers in a similar way,
696 which is clearly wrong.
698 PH/37 Added control=suppress_local_fixups.
700 PH/38 When log_selector = +received_sender was set, and the addition of the
701 sender made the log line's construction buffer exactly full, or one byte
702 less than full, an overflow happened when the terminating "\n" was
705 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
706 when the result of a list match is failure because a DNS lookup failed.
708 PH/40 RM_COMMAND is now used in the building process.
710 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
711 the "build-* directories that it finds.
713 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
714 address in a domain literal was a prefix of an interface address.
716 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
717 when verifying a sender address, unless rewrite_headers is false.
719 PH/44 Wrote a long comment about why errors_to addresses are verified as
720 recipients, not senders.
722 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
723 the ratelimit ACL was added.
725 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
727 PH/46 Added extra information about PostgreSQL errors to the error string.
729 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
730 avoiding the need to install special zones in a real server. This is
731 backwards compatible; if it can't find the fake resolver, it drops back.
732 Thus, both old and new test suites can be run.
734 TF/02 Added util/ratelimit.pl
736 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
737 clock is set back in time.
739 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
740 Candler <B.Candler@pobox.com>.
742 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
743 OK for addresses that are the result of redirections.
745 PH/48 A number of further additions for the benefit of the new test suite,
746 including a fake gethostbyname() that interfaces to the fake DNS resolver
749 TF/06 The fix for widen_domains has also been applied to qualify_single and
750 search_parents which are the other dnslookup options that can cause
753 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
756 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
758 TF/07 Exim produced the error message "an SRV record indicated no SMTP
759 service" if it encountered an MX record with an empty target hostname.
760 The message is now "an MX or SRV record indicated no SMTP service".
762 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
763 if the DNS of the sending site is misconfigured. This is quite a
764 common situation. This change restores the behaviour of treating a
765 helo verification defer as a failure.
767 PH/51 If self=fail was set on a router, the bounce message did not include the
768 actual error message.
774 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
776 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
777 command itself ended in error, the underlying message about the transport
778 filter timeout was being overwritten with the pipe command error. Now the
779 underlying error message should be appended to the second error message.
781 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
783 PH/02 Reduce the amount of output that "make" produces by default. Full output
784 can still be requested.
786 PH/03 The warning log line about a condition test deferring for a "warn" verb
787 was being output only once per connection, rather than after each
788 occurrence (because it was using the same function as for successful
789 "warn" verbs). This seems wrong, so I have changed it.
791 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
792 it should not have, which might have caused a crash in the right
793 circumstances, but probably never did.
795 PH/04 Installed a modified version of Tony Finch's patch to make submission
796 mode fix the return path as well as the Sender: header line, and to
797 add a /name= option so that you can make the user's friendly name appear
800 TF/03 Added the control = fakedefer ACL modifier.
802 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
803 Mark Lowes for thorough testing.
805 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
807 TK/03 Merged latest SRS patch from Miles Wilton.
809 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
810 with the definition in sysexits.h (which is #included earlier).
811 Fortunately, Exim does not actually use EX_OK. The code used to try to
812 preserve the sysexits.h value, by assumimg that macro definitions were
813 scanned for macro replacements. I have been disabused of this notion,
814 so now the code just undefines EX_OK before #including unistd.h.
816 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
817 in the smtp transport. When a block could not be written in a single
818 write() function, the timeout was being re-applied to each part-write.
819 This seems wrong - if the receiver was accepting one byte at a time it
820 would take for ever. The timeout is now adjusted when this happens. It
821 doesn't have to be particularly precise.
823 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
824 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
826 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
827 without a domain if the check to <postmaster@domain> fails.
829 SC/01 Eximstats: added -xls and the ability to specify output files
830 (patch written by Frank Heydlauf).
832 SC/02 Eximstats: use FileHandles for outputing results.
834 SC/03 Eximstats: allow any combination of xls, txt, and html output.
836 SC/04 Eximstats: fixed display of large numbers with -nvr option
838 SC/05 Eximstats: fixed merging of reports with empty tables.
840 SC/06 Eximstats: added the -include_original_destination flag
842 SC/07 Eximstats: removed tabs and trailing whitespace.
844 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
846 TK/06 MBOX spool code: Add real "From " MBOX separator line
847 so the .eml file is really in mbox format (even though
848 most programs do not really care). Patch from Alex Miller.
850 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
851 The latter is generated from $received_to and is only set if the
852 message has one envelope recipient. SA can use these headers,
853 obviously out-of-the-box. Patch from Alex Miller.
855 PH/08 The ${def test on a variable was returning false if the variable's
856 value was "0", contrary to what the specification has always said!
857 The result should be true unless the variable is empty.
859 PH/09 The syntax error of a character other than { following "${if
860 def:variable_name" (after optional whitespace) was not being diagnosed.
861 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
862 accidental colon was present, for example, could give incorrect results.
864 PH/10 Tidied the code in a number of places where the st_size field of a stat()
865 result is used (not including appendfile, where other changes are about
868 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
869 This involved changing a lot of size variables from int to off_t. It
870 should work with maildirs and everything.
872 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
873 spamd dying while we are connected to it.
875 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
878 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
879 function for BDB 4.3.
881 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
883 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
884 the sources for PCRE have been re-arranged and more files are now
887 PH/15 The code I had for printing potentially long long variables in PH/11
888 above was not the best (it lost precision). The length of off_t variables
889 is now inspected at build time, and an appropriate printing format (%ld
890 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
891 to be "long long int" or "long int". This is needed for the internal
892 formatting function string_vformat().
894 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
895 the configuration file to be ":syslog", then the script "guesses" where
896 the logs files are, rather than using the compiled in default. In our
897 case the guess is not the same as the compiled default, so the script
898 suddenly stopped working when I started to use syslog. The patch checks
899 to see if log_file_path is "". If so, it attempts to read it from exim
900 with no configuration file to get the compiled in version, before it
901 falls back to the previous guessing code."
903 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
904 implementing BATV in an Exim configuration. See NewStuff for the gory
907 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
908 Makefile that are specific to HP-UX.
910 PH/18 If the "use_postmaster" option was set for a recipient callout together
911 with the "random" option, the postmaster address was used as the MAIL
912 FROM address for the random test, but not for the subsequent recipient
913 test. It is now used for both.
915 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
916 patch removes a few documentation additions to RFC 3028, because the
917 latest draft now contains them. It adds the new en;ascii-case comparator
918 and a new error check for 8bit text in MIME parts. Comparator and
919 require names are now matched exactly. I enabled the subaddress
920 extension, but it is not well tested yet (read: it works for me)."
922 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
923 rework some of the code of TK/09 above to avoid the hardwired use of
924 "%lld" and "long long". Replaced the call to snprintf() with a call to
927 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
928 records point to non-existent hosts", "retry timeout exceeded", and
929 "retry time not reached for any host after a long failure period".
931 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
932 experimental DomainKeys support:
934 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
935 (2) On an error such as an illegally used "control", the wrong name for
936 the control was given.
938 These problems did NOT occur unless DomainKeys support was compiled.
940 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
942 PH/24 Added ${if match_ip condition.
944 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
945 databases so that it will be absolutely obvious if a crash occurs in the
946 DB library. This is a regular occurrence (often caused by mis-matched
949 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
950 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
951 picked up on a user's system that detects such things. There doesn't seem
952 to be a gcc warning option for this - only an attribute that has to be
953 put on the function's prototype. It seems that in Fedora Core 4 they have
954 set this on a number of new functions. No doubt there will be more in due
957 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
958 specify a transport. However, if an address that was verified by such a
959 router was the subject of a callout, Exim crashed because it tried to
960 read the rcpt_include_affixes from the non-existent transport. Now it
961 just assumes that the setting of that option is false. This bug was
962 introduced by 4.51/PH/31.
964 PH/28 Changed -d+all to exclude +memory, because that information is very
965 rarely of interest, but it makes the output a lot bigger. People tend to
966 do -d+all out of habit.
968 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
969 code in os-type was giving problems when libc.so lives in lib64, like on
972 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
973 aren't the modern standard, and it seems that some systems' include files
974 don't always have them. Exim was already checking for some of the newer
975 ones like T_AAAA, and defining it itself. I've added checks for all the
976 record types that Exim uses.
978 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
979 not automatically generating a new one, as it is supposed to. This
980 prevented TLS from working. If the file did exist, but contained invalid
981 data, a new version was generated, as expected. It was only the case of a
982 non-existent file that was broken.
984 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
985 with a change in libdomainkeys > 0.64.
987 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
988 from DNS. If the selector record carries the flag, it now has
989 precedence over the domain-wide flag.
991 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
993 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
994 the use of a port name that isn't defined in /etc/services) occurred, the
995 message was deferred as in a normal delivery, and thus remained on the
996 spool, instead of being failed because of the mua_wrapper setting. This
997 is now fixed, and I tidied up some of the mua_wrapper messages at the
1000 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1001 the messages in a hash of arrays rather than using individual hashes.
1002 This is a bit cleaner and results in dramatic memory savings, albeit
1003 at a slight CPU cost.
1005 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1006 as requested by Marc Sherman.
1008 SC/10 Eximstats: added histograms for user specified patterns as requested
1011 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1013 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1014 fopen() in the content-scanning modules that did not already have it.
1020 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1021 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1023 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1025 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1027 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1028 to test Sieve filters that use "vacation".
1030 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1031 that changes the way the GnuTLS parameters are stored in the cache file.
1032 The new format can be generated externally. For backward compatibility,
1033 if the data in the cache doesn't make sense, Exim assumes it has read an
1034 old-format file, and it generates new data and writes a new file. This
1035 means that you can't go back to an older release without removing the
1038 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1039 work if there are any delivery delays because "one_time" forces the
1040 parent to be marked "delivered", so its unseen clone is never tried
1041 again. For this reason, Exim now forbids the simultaneous setting of
1044 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1045 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1046 entry if you want to know the details. The fix, however, neglected to
1047 consider the case when local delivery batching is involved. The test for
1048 "previously delivered" was not happening when checking to see if an
1049 address could be batched with a previous (undelivered) one; under
1050 certain circumstances this could lead to multiple deliveries to the same
1053 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1054 in its include files, and this causes problems building Exim.
1056 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1057 header_syntax) but Exim was just ignoring anything given after a slash.
1058 In particular, this caused confusion with an attempt to use "verify =
1059 reverse_host_lookup/defer_ok". An error is now given when options are
1060 supplied for verify items that do not have them. (Maybe reverse_host_
1061 lookup should have a defer_ok option, but that's a different point.)
1063 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1064 defined by RFC 821) to 2048, because there were problems with some AUTH
1065 commands, and RFC 1869 says the size should be increased for extended
1066 SMTP commands that take arguments.
1068 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1071 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1072 "unknown" error; now it says that the functionality isn't in the binary.
1074 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1075 an address' error message when a string expansion fails (syntax or
1076 whatever). Otherwise the password may appear in the log. Following change
1077 PH/42 below, there is no longer a chance of it appearing in a bounce
1080 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1082 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1083 of its addresses was ever tried. (Bugzilla bug #2.)
1085 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1086 the result incorrectly in the debug output. (It correctly added a newline
1087 to what was transported.)
1089 TF/01 Added $received_time.
1091 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1092 commented out examples of how to interface to a virus scanner and to
1093 SpamAssassin. Also added commented examples of av_scanner and
1094 spamd_address settings.
1096 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1097 and controls are allowed in which ACLs. There were a couple of minor
1098 errors. Some of the entries in the conditions table (which is a table of
1099 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1100 negation of where the condition IS allowed.
1102 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1104 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1105 header file does not have a version number, so I've had to invent a new
1106 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1107 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1108 radiusclient), but it was contributed by a Radius user.
1110 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1111 files or directories, for OpenSSL.
1113 PH/19 When an Exim process that is running as root has to create an Exim log
1114 file, it does so in a subprocess that runs as exim:exim so as to get the
1115 ownership right at creation (otherwise, other Exim processes might see
1116 the file with the wrong ownership). There was no test for failure of this
1117 fork() call, which would lead to the process getting stuck as it waited
1118 for a non-existent subprocess. Forks do occasionally fail when resources
1119 run out. I reviewed all the other calls to fork(); they all seem to check
1122 PH/20 When checking for unexpected SMTP input at connect time (before writing
1123 the banner), Exim was not dealing correctly with a non-positive return
1124 from the read() function. If the client had disconnected by this time,
1125 the result was a log entry for a synchronization error with an empty
1126 string after "input=" when read() returned zero. If read() returned -1
1127 (an event I could not check), uninitialized data bytes were printed.
1128 There were reports of junk text (parts of files, etc) appearing after
1131 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1133 PH/22 Added support for macro redefinition, and (re)definition in between
1134 driver and ACL definitions.
1136 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1137 forgetting to use the resulting value; it was using the unexpanded value.
1139 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1140 hadn't been configured. The fix is from Juergen Kreileder, who
1141 understands it better than I do:
1143 "Here's what I see happening with three configured cyrus_sasl
1144 authenticators configured (plain, login, cram-md5):
1146 On startup auth_cyrus_sasl_init() gets called for each of these.
1147 This means three calls to sasl_listmech() without a specified mech_list.
1148 => SASL tests which mechs of all available mechs actually work
1149 => three warnings about OTP not working
1150 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1152 With the patch, sasl_listmech() also gets called three times. But now
1153 SASL's mech_list option is set to the server_mech specified in the the
1154 authenticator. Or in other words, the answer from sasl_listmech()
1155 gets limited to just the mech you're testing for (which is different
1157 => the return list contains just 'plain' or 'login', 'cram-md5' or
1158 nothing depending on the value of ob->server_mech.
1160 I've just tested the patch: Authentication still works fine,
1161 unavailable mechs specified in the exim configuration are still
1162 caught, and the auth.log warnings about OTP are gone."
1164 PH/25 When debugging is enabled, the contents of the command line are added
1165 to the debugging output, even when log_selector=+arguments is not
1168 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1169 answer is "GNU", and only if the return is "GNU/something" is the answer
1172 PH/27 $acl_verify_message is now set immediately after the failure of a
1173 verification in an ACL, and so is available in subsequent modifiers. In
1174 particular, the message can be preserved by coding like this:
1176 warn !verify = sender
1177 set acl_m0 = $acl_verify_message
1179 Previously, $acl_verify_message was set only while expanding "message"
1180 and "log_message" when a very denied access.
1182 PH/28 Modified OS/os.c-Linux with
1184 -#ifndef OS_LOAD_AVERAGE
1185 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1187 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1188 nomenclature these days.)
1190 PH/29 Installed patch from the Sieve maintainer that adds the options
1191 sieve_useraddress and sieve_subaddress to the redirect router.
1193 PH/30 In these circumstances:
1194 . Two addresses routed to the same list of hosts;
1195 . First host does not offer TLS;
1196 . First host accepts first address;
1197 . First host gives temporary error to second address;
1198 . Second host offers TLS and a TLS session is established;
1199 . Second host accepts second address.
1200 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1201 and peerdn, if requested) that were in fact used only for the second
1204 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1205 attention to any local part prefix or suffix that was matched by the
1206 router that accepted the address. It now behaves in the same way as it
1207 does for delivery: the affixes are removed from the local part unless
1208 rcpt_include_affixes is set on the transport.
1210 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1211 timeout during the DATA phase of an incoming message.
1213 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1214 applied a patch sanctioned by the Sieve maintainer.
1216 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1217 the uid or gid is negative. A case of a negative gid caused this to be
1218 noticed. The fix allows for either to be negative.
1220 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1221 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1224 PH/36 The change PH/12 above was broken. Fixed it.
1226 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1227 the grounds that routing the same address twice would always produce the
1228 same answer. This might have been true once, but it is certainly no
1229 longer true now. Routing a child address may depend on the previous
1230 routing that produced that child. Some complicated redirection strategies
1231 went wrong when messages had multiple recipients, and made Exim's
1232 behaviour dependent on the order in which the addresses were given.
1234 I have moved the duplicate checking until after the routing is complete.
1235 Exim scans the addresses that are assigned to local and remote
1236 transports, and removes any duplicates. This means that more work will be
1237 done, as duplicates will always all be routed, but duplicates are
1238 presumably rare, so I don't expect this is of any significance.
1240 For deliveries to pipes, files, and autoreplies, the duplicate checking
1241 still happens during the routing process, since they are not going to be
1244 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1245 It corrects a timeout issue with spamd. This is Ian's comment: "The
1246 background is that sometimes spamd either never reads data from a
1247 connection it has accepted, or it never writes response data. The exiscan
1248 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1249 blindly assumes that writes won't block so it may never time out."
1251 PH/39 Allow G after quota size as well as K and M.
1253 PH/40 The value set for $authenticated_id in an authenticator may not contain
1254 binary zeroes or newlines because the value is written to log lines and
1255 to spool files. There was no check on this. Now the value is run through
1256 the string_printing() function so that such characters are converted to
1257 printable escape sequences.
1259 PH/41 $message_linecount is a new variable that contains the total number of
1260 lines in the message. Compare $body_linecount, which is the count for the
1263 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1264 bounce and delay warning messages, except in certain special cases, which
1267 (a) An SMTP error message from a remote host;
1268 (b) A message specified in a :fail: redirection;
1269 (c) A message specified in a "fail" command in a system filter;
1270 (d) A message specified in a FAIL return from the queryprogram router;
1271 (e) A message specified by the cannot_route_message router option.
1273 In these cases only, Exim does include the error details in bounce and
1274 warning messages. There are also a few cases where bland messages such
1275 as "unrouteable address" or "local delivery error" are given.
1277 PH/43 $value is now also set for the "else" part of a ${run expansion.
1279 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1280 being worked on, but at least Exim now implements the latest version to
1283 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1284 process to complete was treated as a delivery failure, a timeout while
1285 writing the message to the pipe was logged, but erroneously treated as a
1286 successful delivery. Such timeouts include transport filter timeouts. For
1287 consistency with the overall process timeout, these timeouts are now
1288 treated as errors, giving rise to delivery failures by default. However,
1289 there is now a new Boolean option for the pipe transport called
1290 timeout_defer, which, if set TRUE, converts the failures into defers for
1291 both kinds of timeout. A transport filter timeout is now identified in
1294 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1295 systems where "make" and "gmake" are different, calling "gmake" at top
1296 level broke things. I've arranged for the value of $(MAKE) to be passed
1297 from the Makefile to this script so that it can call the same version of
1301 A note about Exim versions 4.44 and 4.50
1302 ----------------------------------------
1304 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1305 changes of various kinds. As a consequence, a big documentation update was
1306 needed. This delayed the release for rather longer than seemed good, especially
1307 in the light of a couple of (minor) security issues. Therefore, the changes
1308 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1309 release. So 4.44 and 4.50 are in effect two different branches that both start
1312 I have left the 4.50 change log unchanged; it contains all the changes since
1313 4.43. The change log for 4.44 is below; many of its items are identical to
1314 those for 4.50. This seems to be the most sensible way to preserve the
1315 historical information.
1321 1. Minor wording change to the doc/README.SIEVE file.
1323 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1324 computation of the current number of files was incorrect.
1326 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1327 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1330 4. Give more explanation in the error message when the command for a transport
1331 filter fails to execute.
1333 5. There are several places where Exim runs a non-Exim command in a
1334 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1335 was being done only for the command run by the queryprogram router. It is
1336 now done for all such subprocesses. The other cases are: ${run, transport
1337 filters, and the commands run by the lmtp and pipe transports.
1339 6. Added CONFIGURE_GROUP build-time option.
1341 7. Some older OS have a limit of 256 on the maximum number of file
1342 descriptors. Exim was using setrlimit() to set 1000 as a large value
1343 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1344 systems. I've change it so that if it can't get 1000, it tries for 256.
1346 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1347 was an oversight, and furthermore, ever since the addition of extra
1348 controls (e.g. 4.43/32), the checks on when to allow different forms of
1349 "control" were broken. There should now be diagnostics for all cases when a
1350 control that does not make sense is encountered.
1352 9. Added the /retain_sender option to "control=submission".
1354 10. $recipients is now available in the predata ACL (oversight).
1356 11. Tidy the search cache before the fork to do a delivery from a message
1357 received from the command line. Otherwise the child will trigger a lookup
1358 failure and thereby defer the delivery if it tries to use (for example) a
1359 cached ldap connection that the parent has called unbind on.
1361 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1362 of $address_data from the recipient verification was clobbered by the
1363 sender verification.
1365 13. The value of address_data from a sender verification is now available in
1366 $sender_address_data in subsequent conditions in the ACL statement.
1368 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1370 15. Added a new option "connect=<time>" to callout options, to set a different
1373 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1374 was its contents. (It was OK if the option was not defined at all.)
1376 17. A "Completed" log line is now written for messages that are removed from
1377 the spool by the -Mrm option.
1379 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1380 information about exactly what failed.
1382 19. Added -dd to debug only the daemon process.
1384 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1385 handles timeouts, both on the server side and network timeouts. Renamed the
1386 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1388 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1389 It is now set to "smtps".
1391 22. $host_address is now set to the target address during the checking of
1392 ignore_target_hosts.
1394 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1395 being passed; this would have caused $sender_host_name to have been used if
1396 matching the list had actually called for a host name (not very likely,
1397 since this list is usually IP addresses). A host name is now passed as
1400 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1401 code that specifies a non-restarting handler (typically sigaction() in
1402 modern systems) in an attempt to fix a rare and obscure crash bug.
1404 25. Narrowed the window for a race in the daemon that could cause it to ignore
1405 SIGCHLD signals. This is not a major problem, because they are used only to
1406 wake it up if nothing else does.
1408 26. A malformed maildirsize file could cause Exim to calculate negative values
1409 for the mailbox size or file count. Odd effects could occur as a result.
1410 The maildirsize information is now recalculated if the size or filecount
1413 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1414 support for a long time. Removed HAVE_SYS_VFS_H.
1416 28. Installed the latest version of exipick from John Jetmore.
1418 29. In an address list, if the pattern was not a regular expression, an empty
1419 subject address (from a bounce message) matched only if the pattern was an
1420 empty string. Non-empty patterns were not even tested. This was the wrong
1421 because it is perfectly reasonable to use an empty address as part of a
1422 database query. An empty address is now tested by patterns that are
1423 lookups. However, all the other forms of pattern expect the subject to
1424 contain a local part and a domain, and therefore, for them, an empty
1425 address still always fails if the pattern is not itself empty.
1427 30. Exim went into a mad DNS loop when attempting to do a callout where the
1428 host was specified on an smtp transport, and looking it up yielded more
1429 than one IP address.
1431 31. Re-factored the code for checking spool and log partition space into a
1432 function that finds that data and another that does the check. The former
1433 is then used to implement four new variables: $spool_space, $log_space,
1434 $spool_inodes, and $log_inodes.
1436 32. The RFC2047 encoding function was originally intended for short strings
1437 such as real names; it was not keeping to the 75-character limit for
1438 encoded words that the RFC imposes. It now respects the limit, and
1439 generates multiple encoded words if necessary. To be on the safe side, I
1440 have increased the buffer size for the ${rfc2047: expansion operator from
1443 33. It is now permitted to omit both strings after an "if" condition; if the
1444 condition is true, the result is "true". As before, when the second string
1445 is omitted, a false condition yields an empty string. This makes it less
1446 cumbersome to write custom ACL and router conditions.
1448 34. Failure to deliver a bounce message always caused it to be frozen, even if
1449 there was an errors_to setting on the router. The errors_to setting is now
1452 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1453 canonical form (fully expanded) before being placed in
1454 $sender_host_address.
1456 36. The table in the code that translates DNS record types into text (T_A to
1457 "A" for instance) was missing entries for NS and CNAME. It is just possible
1458 that this could have caused confusion if both these types were looked up
1459 for the same domain, because the text type is used as part of Exim's
1460 per-process caching. But the chance of anyone hitting this buglet seems
1463 37. The dnsdb lookup has been extended in a number of ways.
1465 (1) There is a new type, "zns", which walks up the domain tree until it
1466 finds some nameserver records. It should be used with care.
1468 (2) There is a new type, "mxh", which is like "mx" except that it returns
1469 just the host names, not the priorities.
1471 (3) It is now possible to give a list of domains (or IP addresses) to be
1472 looked up. The behaviour when one of the lookups defers can be
1473 controlled by a keyword.
1475 (4) It is now possible to specify the separator character for use when
1476 multiple records are returned.
1478 38. The dnslists ACL condition has been extended: it is now possible to supply
1479 a list of IP addresses and/or domains to be looked up in a particular DNS
1482 39. Added log_selector=+queue_time_overall.
1484 40. When running the queue in the test harness, wait just a tad after forking a
1485 delivery process, to get repeatability of debugging output.
1487 41. Include certificate and key file names in error message when GnuTLS fails
1488 to set them up, because the GnuTLS error message doesn't include the name
1489 of the failing file when there is a problem reading it.
1491 42. Allow both -bf and -bF in the same test run.
1493 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1495 44. The "Exiscan patch" is now merged into the mainline Exim source.
1497 45. Sometimes the final signoff response after QUIT could fail to get
1498 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1499 < 0 before doing a fflush(). This bug looks as though it goes back to the
1500 introduction of TLS in release 3.20, but "sometimes" must have been rare
1501 because the tests only now provoked it.
1503 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1504 (this can affect the format of dates).
1506 47. exim_tidydb, when checking for the continued existence of a message for
1507 which it has found a message-specific retry record, was not finding
1508 messages that were in split spool directories. Consequently, it was
1509 deleting retry records that should have stayed in existence.
1511 48. Steve fixed some bugs in eximstats.
1513 49. The SPA authentication driver was not abandoning authentication and moving
1514 on to the next authenticator when an expansion was forced to fail,
1515 contradicting the general specification for all authenticators. Instead it
1516 was generating a temporary error. It now behaves as specified.
1518 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1519 (the order specifies the preference for clients). The order is now AES256,
1520 AES128, 3DES, ARCFOUR128.
1522 51. Small patch to Sieve code - explicitly set From: when generating an
1525 52. Exim crashed if a remote delivery caused a very long error message to be
1526 recorded - for instance if somebody sent an entire SpamAssassin report back
1527 as a large number of 550 error lines. This bug was coincidentally fixed by
1528 increasing the size of one of Exim's internal buffers (big_buffer) that
1529 happened as part of the Exiscan merge. However, to be on the safe side, I
1530 have made the code more robust (and fixed the comments that describe what
1533 53. Now that there can be additional text after "Completed" in log lines (if
1534 the queue_time_overall log selector is set), a one-byte patch to exigrep
1535 was needed to allow it to recognize "Completed" as not the last thing in
1538 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1539 patch that reportedly fixes this has been added. I am not expert enough to
1540 create a test for it. This is what the patch creator wrote:
1542 "I found a little strange behaviour of ldap code when working with
1543 Windows 2003 AD Domain, where users was placed in more than one
1544 Organization Units. When I tried to give exim partial DN, the exit code
1545 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1546 But simultaneously result of request was absolutely normal ldap result,
1547 so I produce this patch..."
1549 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1550 REFERENCE, so I have modified the code to exclude the patch when that macro
1553 55. Some experimental protocols are using DNS PTR records for new purposes. The
1554 keys for these records are domain names, not reversed IP addresses. The
1555 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1556 leaves it alone. Component reversal etc. now happens only for IP addresses.
1559 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1561 57. Double the size of the debug message buffer (to 2048) so that more of very
1562 long debug lines gets shown.
1564 58. The exicyclog utility now does better if the number of log files to keep
1565 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1567 59. Two changes related to the smtp_active_hostname option:
1569 (1) $smtp_active_hostname is now available as a variable.
1570 (2) The default for smtp_banner uses $smtp_active_hostname instead
1571 of $primary_hostname.
1573 60. The host_aton() function is supposed to be passed a string that is known
1574 to be a valid IP address. However, in the case of IPv6 addresses, it was
1575 not checking this. This is a hostage to fortune. Exim now panics and dies
1576 if the condition is not met. A case was found where this could be provoked
1577 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1578 components; fortuitously, this particular loophole had already been fixed
1579 by change 4.50/55 above.
1581 If there are any other similar loopholes, the new check in host_aton()
1582 itself should stop them being exploited. The report I received stated that
1583 data on the command line could provoke the exploit when Exim was running as
1584 exim, but did not say which command line option was involved. All I could
1585 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1586 running as the user.
1589 61. There was a buffer overflow vulnerability in the SPA authentication code
1590 (which came originally from the Samba project). I have added a test to the
1591 spa_base64_to_bits() function which I hope fixes it.
1594 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1595 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1597 63. The daemon start-up calls getloadavg() while still root for those OS that
1598 need the first call to be done as root, but it missed one case: when
1599 deliver_queue_load_max is set with deliver_drop_privilege. This is
1600 necessary for the benefit of the queue runner, because there is no re-exec
1601 when deliver_drop_privilege is set.
1603 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1604 This has been fixed.
1606 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1607 was in use, was not putting the data itself into the right store pool;
1608 consequently, it could be overwritten for a subsequent message in the same
1609 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1612 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1614 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1615 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1616 the function were treating the return as a boolean value, which happened to
1617 work because 0=false and not-0=true, but is not correct code.
1619 68. The host_aton() function was not handling scoped IPv6 addresses (those
1620 with, for example, "%eth0" on the end) correctly.
1622 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1623 negated items (that is, ~something) in unsigned ints. Some compilers
1624 apparently mutter when there is no cast.
1626 70. If an address verification called from an ACL failed, and did not produce a
1627 user-specific message (i.e. there was only a "system" message), nothing was
1628 put in $acl_verify_message. In this situation, it now puts the system
1631 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1632 session; change 4.31/43 added the unwanted input to the log line - except
1633 that it did not do this in the start of session case. It now does.
1635 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1636 This is wrong and can cause the other end to generate a synchronization
1637 error if it is another Exim or anything else that does the synchronization
1638 check. A QUIT command is no longer sent after a timeout.
1640 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1641 during host lookups.
1643 74. The defer_ok option of callout verification was not working if it was used
1644 when verifying addresses in header lines, that is, for this case:
1646 verify = header_sender/callout=defer_ok
1648 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1649 those file descriptors could be used for SMTP connections. If anything
1650 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1651 could be sent to the SMTP client, causing chaos. The daemon now opens
1652 stdin, stdout, and stderr to /dev/null when it puts itself into the
1655 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1656 log by default. The user can override this with suitable Perl magic.
1658 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1659 the log message when discard triggers, was not working for the DATA ACL or
1660 for the non-SMTP ACL.
1662 78. Error message wording change in sieve.c.
1664 79. If smtp_accept_max_per_host was set, the number of connections could be
1665 restricted to fewer than expected, because the daemon was trying to set up
1666 a new connection before checking whether the processes handling previous
1667 connections had finished. The check for completed processes is now done
1668 earlier. On busy systems, this bug wouldn't be noticed because something
1669 else would have woken the daemon, and it would have reaped the completed
1672 80. If a message was submitted locally by a user whose login name contained one
1673 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1674 It caused a spool format error. I have fixed the spool reading code. A
1675 related problem was that the "from" clause in the Received: line became
1676 illegal because of the space(s). It is now covered by ${quote_local_part.
1678 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1681 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1683 83. Patch from Sieve maintainer to fix unterminated string problem in
1684 "vacation" handling.
1686 84. Some minor changes to the Linux configuration files to help with other
1687 OS variants using glibc.
1689 85. One more patch for Sieve to update vacation handling to latest spec.
1692 ----------------------------------------------------
1693 See the note above about the 4.44 and 4.50 releases.
1694 ----------------------------------------------------
1700 1. Change 4.43/35 introduced a bug that caused file counts to be
1701 incorrectly computed when quota_filecount was set in an appendfile
1704 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1705 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1708 3. Give more explanation in the error message when the command for a transport
1709 filter fails to execute.
1711 4. There are several places where Exim runs a non-Exim command in a
1712 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1713 was being done only for the command run by the queryprogram router. It is
1714 now done for all such subprocesses. The other cases are: ${run, transport
1715 filters, and the commands run by the lmtp and pipe transports.
1717 5. Some older OS have a limit of 256 on the maximum number of file
1718 descriptors. Exim was using setrlimit() to set 1000 as a large value
1719 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1720 systems. I've change it so that if it can't get 1000, it tries for 256.
1722 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1723 was an oversight, and furthermore, ever since the addition of extra
1724 controls (e.g. 4.43/32), the checks on when to allow different forms of
1725 "control" were broken. There should now be diagnostics for all cases when a
1726 control that does not make sense is encountered.
1728 7. $recipients is now available in the predata ACL (oversight).
1730 8. Tidy the search cache before the fork to do a delivery from a message
1731 received from the command line. Otherwise the child will trigger a lookup
1732 failure and thereby defer the delivery if it tries to use (for example) a
1733 cached ldap connection that the parent has called unbind on.
1735 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1736 of $address_data from the recipient verification was clobbered by the
1737 sender verification.
1739 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1740 was its contents. (It was OK if the option was not defined at all.)
1742 11. A "Completed" log line is now written for messages that are removed from
1743 the spool by the -Mrm option.
1745 12. $host_address is now set to the target address during the checking of
1746 ignore_target_hosts.
1748 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1749 being passed; this would have caused $sender_host_name to have been used if
1750 matching the list had actually called for a host name (not very likely,
1751 since this list is usually IP addresses). A host name is now passed as
1754 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1755 code that specifies a non-restarting handler (typically sigaction() in
1756 modern systems) in an attempt to fix a rare and obscure crash bug.
1758 15. Narrowed the window for a race in the daemon that could cause it to ignore
1759 SIGCHLD signals. This is not a major problem, because they are used only to
1760 wake it up if nothing else does.
1762 16. A malformed maildirsize file could cause Exim to calculate negative values
1763 for the mailbox size or file count. Odd effects could occur as a result.
1764 The maildirsize information is now recalculated if the size or filecount
1767 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1768 support for a long time. Removed HAVE_SYS_VFS_H.
1770 18. Updated exipick to current release from John Jetmore.
1772 19. Allow an empty sender to be matched against a lookup in an address list.
1773 Previously the only cases considered were a regular expression, or an
1776 20. Exim went into a mad DNS lookup loop when doing a callout where the
1777 host was specified on the transport, if the DNS lookup yielded more than
1780 21. The RFC2047 encoding function was originally intended for short strings
1781 such as real names; it was not keeping to the 75-character limit for
1782 encoded words that the RFC imposes. It now respects the limit, and
1783 generates multiple encoded words if necessary. To be on the safe side, I
1784 have increased the buffer size for the ${rfc2047: expansion operator from
1787 22. Failure to deliver a bounce message always caused it to be frozen, even if
1788 there was an errors_to setting on the router. The errors_to setting is now
1791 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1792 canonical form (fully expanded) before being placed in
1793 $sender_host_address.
1795 24. Updated eximstats to version 1.33
1797 25. Include certificate and key file names in error message when GnuTLS fails
1798 to set them up, because the GnuTLS error message doesn't include the name
1799 of the failing file when there is a problem reading it.
1801 26. Expand error message when OpenSSL has problems setting up cert/key files.
1804 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1805 (this can affect the format of dates).
1807 28. exim_tidydb, when checking for the continued existence of a message for
1808 which it has found a message-specific retry record, was not finding
1809 messages that were in split spool directories. Consequently, it was
1810 deleting retry records that should have stayed in existence.
1812 29. eximstats updated to version 1.35
1813 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1814 1.35 - bugfix such that pie charts by volume are generated correctly
1816 30. The SPA authentication driver was not abandoning authentication and moving
1817 on to the next authenticator when an expansion was forced to fail,
1818 contradicting the general specification for all authenticators. Instead it
1819 was generating a temporary error. It now behaves as specified.
1821 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1822 (the order specifies the preference for clients). The order is now AES256,
1823 AES128, 3DES, ARCFOUR128.
1825 31. Small patch to Sieve code - explicitly set From: when generating an
1828 32. Exim crashed if a remote delivery caused a very long error message to be
1829 recorded - for instance if somebody sent an entire SpamAssassin report back
1830 as a large number of 550 error lines. This bug was coincidentally fixed by
1831 increasing the size of one of Exim's internal buffers (big_buffer) that
1832 happened as part of the Exiscan merge. However, to be on the safe side, I
1833 have made the code more robust (and fixed the comments that describe what
1836 33. Some experimental protocols are using DNS PTR records for new purposes. The
1837 keys for these records are domain names, not reversed IP addresses. The
1838 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1839 leaves it alone. Component reversal etc. now happens only for IP addresses.
1842 34. The host_aton() function is supposed to be passed a string that is known
1843 to be a valid IP address. However, in the case of IPv6 addresses, it was
1844 not checking this. This is a hostage to fortune. Exim now panics and dies
1845 if the condition is not met. A case was found where this could be provoked
1846 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1847 components; fortuitously, this particular loophole had already been fixed
1848 by change 4.50/55 or 4.44/33 above.
1850 If there are any other similar loopholes, the new check in host_aton()
1851 itself should stop them being exploited. The report I received stated that
1852 data on the command line could provoke the exploit when Exim was running as
1853 exim, but did not say which command line option was involved. All I could
1854 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1855 running as the user.
1858 35. There was a buffer overflow vulnerability in the SPA authentication code
1859 (which came originally from the Samba project). I have added a test to the
1860 spa_base64_to_bits() function which I hope fixes it.
1863 36. The daemon start-up calls getloadavg() while still root for those OS that
1864 need the first call to be done as root, but it missed one case: when
1865 deliver_queue_load_max is set with deliver_drop_privilege. This is
1866 necessary for the benefit of the queue runner, because there is no re-exec
1867 when deliver_drop_privilege is set.
1869 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1870 was in use, was not putting the data itself into the right store pool;
1871 consequently, it could be overwritten for a subsequent message in the same
1872 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1875 38. Sometimes the final signoff response after QUIT could fail to get
1876 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1877 < 0 before doing a fflush(). This bug looks as though it goes back to the
1878 introduction of TLS in release 3.20, but "sometimes" must have been rare
1879 because the tests only now provoked it.
1885 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1886 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1887 Now it returns an integer. A number of calls were still expecting a T/F
1888 return. Fortuitously, in all cases, the tests worked in OK situations,
1889 which is the norm. However, things would have gone wrong on any write
1890 failures on the smtp file descriptor. This function is used when sending
1891 messages over SMTP and also when doing verify callouts.
1893 2. When Exim is called to do synchronous delivery of a locally submitted
1894 message (the -odf or -odi options), it no longer closes stderr before doing
1897 3. Implemented the mua_wrapper option.
1899 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1901 5. Implemented the functions header_remove(), header_testname(),
1902 header_add_at_position(), and receive_remove_recipient(), and exported them
1905 6. If an ACL "warn" statement specified the addition of headers, Exim already
1906 inserted X-ACL-Warn: at the start if there was no header name. However, it
1907 was not making this test for the second and subsequent header lines if
1908 there were newlines in the string. This meant that an invalid header could
1909 be inserted if Exim was badly configured.
1911 7. Allow an ACL "warn" statement to add header lines at the start or after all
1912 the Received: headers, as well as at the end.
1914 8. Added the rcpt_4xx retry error code.
1916 9. Added postmaster_mailfrom=xxx to callout verification option.
1918 10. Added mailfrom=xxxx to the callout verification option, for verify=
1921 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1922 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1924 12. Inserted some casts to stop certain compilers warning when using pointer
1925 differences as field lengths or precisions in printf-type calls (mostly
1926 affecting debugging statements).
1928 13. Added optional readline() support for -be (dynamically loaded).
1930 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1931 same clock tick as a message's arrival, so that its received time was the
1932 same as the "first fail" time on the retry record, and that message
1933 remained on the queue past the ultimate address timeout, every queue runner
1934 would try a delivery (because it was past the ultimate address timeout) but
1935 after another failure, the ultimate address timeout, which should have then
1936 bounced the address, did not kick in. This was a "< instead of <=" error;
1937 in most cases the first failure would have been in the next clock tick
1938 after the received time, and all would be well.
1940 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1941 being recognized when the domain list was tested by the match_domain
1942 condition in an expansion string.
1944 16. Added the ${str2b64: operator.
1946 17. Exim was always calling setrlimit() to set a large limit for the number of
1947 processes, without checking whether the existing limit was already
1948 adequate. (It did check for the limit on file descriptors.) Furthermore,
1949 errors from getrlimit() and setrlimit() were being ignored. Now they are
1950 logged to the main and panic logs, but Exim does carry on, to try to do its
1951 job under whatever limits there are.
1953 18. Imported PCRE 5.0.
1955 19. Trivial typo in log message " temporarily refused connection" (the leading
1958 20. If the log selector return_path_on_delivery was set and an address was
1959 redirected to /dev/null, the delivery process crashed because it assumed
1960 that a return path would always be set for a "successful" delivery. In this
1961 case, the whole delivery is bypassed as an optimization, and therefore no
1964 21. Internal re-arrangement: the function for sending a challenge and reading
1965 a response while authentication was assuming a zero-terminated challenge
1966 string. It's now changed to take a pointer and a length, to allow for
1967 binary data in such strings.
1969 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1971 23. Exim was not respecting finduser_retries when seeking the login of the
1972 uid under which it was called; it was always trying 10 times. (The default
1973 setting of finduser_retries is zero.) Also, it was sleeping after the final
1974 failure, which is pointless.
1976 24. Implemented tls_on_connect_ports.
1978 25. Implemented acl_smtp_predata.
1980 26. If the domain in control=submission is set empty, Exim assumes that the
1981 authenticated id is a complete email address when it generates From: or
1982 Sender: header lines.
1984 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1985 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1986 chown and chgrp in /bin and hostname in /usr/bin.
1988 28. Exim was keeping the "process log" file open after each use, just as it
1989 does for the main log. This opens the possibility of it remaining open for
1990 long periods when the USR1 signal hits a daemon. Occasional processlog
1991 errors were reported, that could have been caused by this. Anyway, it seems
1992 much more sensible not to leave this file open at all, so that is what now
1995 29. The long-running daemon process does not normally write to the log once it
1996 has entered its main loop, and it closes the log before doing so. This is
1997 so that log files can straightforwardly be renamed and moved. However,
1998 there are a couple of unusual error situations where the daemon does write
1999 log entries, and I had neglected to close the log afterwards.
2001 30. The text of an SMTP error response that was received during a remote
2002 delivery was being truncated at 512 bytes. This is too short for some of
2003 the long messages that one sometimes sees. I've increased the limit to
2006 31. It is now possible to make retry rules that apply only when a message has a
2007 specific sender, in particular, an empty sender.
2009 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2010 it possible to be selective about when SMTP synchronization is enforced.
2012 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2014 32. Implemented hosts_connection_nolog.
2016 33. Added an ACL for QUIT.
2018 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2021 35. Added mailbox_size and mailbox_filecount to appendfile.
2023 36. Added control = no_multiline_responses to ACLs.
2025 37. There was a bug in the logic of the code that waits for the clock to tick
2026 in the case where the clock went backwards by a substantial amount such
2027 that the microsecond fraction of "now" was more than the microsecond
2028 fraction of "then" (but the whole seconds number was less).
2030 38. Added support for the libradius Radius client library this is found on
2031 FreeBSD (previously only the radiusclient library was supported).
2037 1. When certain lookups returned multiple values in the form name=value, the
2038 quoting of the values was not always being done properly. Specifically:
2039 (a) If the value started with a double quote, but contained no whitespace,
2041 (b) If the value contained whitespace other than a space character (i.e.
2042 tabs or newlines or carriage returns) it was not quoted.
2043 This fix has been applied to the mysql and pgsql lookups by writing a
2044 separate quoting function and calling it from the lookup code. The fix
2045 should probably also be applied to nisplus, ibase and oracle lookups, but
2046 since I cannot test any of those, I have not disturbed their existing code.
2048 2. A hit in the callout cache for a specific address caused a log line with no
2049 reason for rejecting RCPT. Now it says "Previous (cached) callout
2050 verification failure".
2052 3. There was an off-by-one bug in the queryprogram router. An over-long
2053 return line was truncated at 256 instead of 255 characters, thereby
2054 overflowing its buffer with the terminating zero. As well as fixing this, I
2055 have increased the buffer size to 1024 (and made a note to document this).
2057 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2058 when Exim is waiting for an SMTP response from a remote server, Exim
2059 restarts its select() call on the socket, thereby resetting its timeout.
2060 This is not a problem when such interrupts are rare. Somebody set up a cron
2061 job to run exiwhat every 2 minutes, which is less than the normal select()
2062 timeout (5 or 10 minutes). This meant that the select() timeout never
2063 kicked in because it was always reset. I have fixed this by comparing the
2064 time when an interrupt arrives with the time at the start of the first call
2065 to select(). If more time than the timeout has elapsed, the interrupt is
2066 treated as a timeout.
2068 5. Some internal re-factoring in preparation for the addition of Sieve
2069 extensions (by MH). In particular, the "personal" test is moved to a
2070 separate function, and given an option for scanning Cc: and Bcc: (which is
2071 not set for Exim filters).
2073 6. When Exim created an email address using the login of the caller as the
2074 local part (e.g. when creating a From: or Sender: header line), it was not
2075 quoting the local part when it contained special characters such as @.
2077 7. Installed new OpenBSD configuration files.
2079 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2080 try to make them clearer.
2082 9. Callout options, other than the timeout value, were being ignored when
2083 verifying sender addresses in header lines. For example, when using
2085 verify = header_sender/callout=no_cache
2087 the cache was (incorrectly) being used.
2089 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2090 only the Cygwin environment.
2092 11. When return_path_on_delivery was set as a log selector, if different remote
2093 addresses in the same message used different return paths and parallel
2094 remote delivery occurred, the wrong values would sometimes be logged.
2095 (Whenever a remote delivery process finished, the return path value from
2096 the most recently started remote delivery process was logged.)
2098 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2099 lines when AUTH and/or TLS are in use. This is the "received protocol"
2100 field. Exim used to use "asmtp" for authenticated SMTP, without any
2101 indication (in the protocol name) for TLS use. Now it follows the RFC and
2102 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2103 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2104 names appear in log lines as well as in Received: header lines.
2106 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2107 extensions, and comparison tests, and to fix some bugs.
2109 14. Changes to the "personal" filter test:
2111 (1) The test was buggy in that it was just doing the equivalent of
2112 "contains" tests on header lines. For example, if a user's address was
2113 anne@some.where, the "personal" test would incorrectly be true for
2115 To: susanne@some.where
2117 This test is now done by extracting each address from the header in turn,
2118 and checking the entire address. Other tests that are part of "personal"
2119 are now done using regular expressions (for example, to check local parts
2120 of addresses in From: header lines).
2122 (2) The list of non-personal local parts in From: addresses has been
2123 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2124 taken from the Sieve specification recommendations.
2126 (3) If the message contains any header line starting with "List-" it is
2127 treated as non-personal.
2129 (4) The test for "circular" in the Subject: header line has been removed
2130 because it now seems ill-conceived.
2132 15. Minor typos in src/EDITME comments corrected.
2134 16. Installed latest exipick from John Jetmore.
2136 17. If headers_add on a router specified a text string that was too long for
2137 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2138 of string_sprintf() is now avoided.
2140 18. $message_body_size was not set (it was always zero) when running the DATA
2141 ACL and the local_scan() function.
2143 19. For the "mail" command in an Exim filter, no default was being set for
2144 the once_repeat time, causing a random time value to be used if "once" was
2145 specified. (If the value happened to be <= 0, no repeat happened.) The
2146 default is now 0s, meaning "never repeat". The "vacation" command was OK
2147 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2148 (I found it when inspecting the code).
2150 20. There is now an overall timeout for performing a callout verification. It
2151 defaults to 4 times the callout timeout, which applies to individual SMTP
2152 commands during the callout. The overall timeout applies when there is more
2153 than one host that can be tried. The timeout is checked before trying the
2154 next host. This prevents very long delays if there are a large number of
2155 hosts and all are timing out (e.g. when the network connections are timing
2156 out). The value of the overall timeout can be changed by specifying an
2157 additional sub-option for "callout", called "maxwait". For example:
2159 verify = sender/callout=5s,maxwait=20s
2161 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2162 to the end before writing, but this should make it even safer).
2164 22. Exim was forgetting that it had advertised PIPELINING for the second and
2165 subsequent messages on an SMTP connection. It was also not resetting its
2166 memory on STARTTLS and an internal HELO.
2168 23. When Exim logs an SMTP synchronization error within a session, it now
2169 records whether PIPELINING has been advertised or not.
2171 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2172 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2173 rather than long int.
2175 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2177 26. Added the never_mail option to autoreply.
2183 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2184 crash if the getsockname() call failed; this can happen if a connection is
2185 closed very soon after it is established. The problem was simply in the
2186 order in which certain operations were done, causing Exim to try to write
2187 to the SMTP stream before it had set up the file descriptor. The bug has
2188 been fixed by making things happen in the correct order.
2194 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2195 before the connection was closed, thus losing the rejection response.
2197 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2198 some early Solaris releases, but causes trouble in current releases where
2199 socklen_t is defined.
2201 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2204 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2207 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2208 that releasing the top of it at the end releases what was used for sub-
2209 expansions (unless the block got too big). However, discard this block if
2210 the first thing is a variable or header, so that we can use its block when
2211 it is dynamic (useful for very large $message_headers, for example).
2213 6. Lookups now cache *every* query, not just the most recent. A new, separate
2214 store pool is used for this. It can be recovered when all lookup caches are
2215 flushed. Lookups now release memory at the end of their result strings.
2216 This has involved some general refactoring of the lookup sources.
2218 7. Some code has been added to the store_xxx() functions to reduce the amount
2219 of flapping under certain conditions.
2221 8. log_incoming_interface used to affect only the <= reception log lines. Now
2222 it causes the local interface and port to be added to several more SMTP log
2223 lines, for example "SMTP connection from", and rejection lines.
2225 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2227 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2229 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2230 could be overwritten at the end of the current message (or the start of a
2231 new message if it was set in a HELO ACL). The value is now preserved for
2232 the duration of the SMTP connection.
2234 12. If a transport had a headers_rewrite setting, and a matching header line
2235 contained an unqualified address, that address was qualified, even if it
2236 did not match any rewriting rules. The underlying bug was that the values
2237 of the flags that permit the existence of unqualified sender and recipient
2238 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2239 non-local messages, and by -bnq for local messages) were not being
2240 preserved with the message after it was received.
2242 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2243 "next input=" as part of the text comprising the host identity instead of
2244 the correct text. The code was using the same buffer for two different
2245 strings. However, depending on which order the printing function evaluated
2246 its arguments, the bug did not always show up. Under Linux, for example, my
2247 test suite worked just fine.
2249 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2250 doesn't work with some older versions of Perl. It has been changed to "my",
2251 which in any case is probably the better facility to use.
2253 15. A really picky compiler found some instances of statements for creating
2254 error messages that either had too many or two few arguments for the format
2257 16. The size of the buffer for calls to the DNS resolver has been increased
2258 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2259 for addresses that have a lot of PTR records. This alleviates a problem; it
2260 does not fully solve it.
2262 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2263 buffer now truncates the list and logs the incident, which is the same
2264 action as happens when Exim is looking up a host name and its aliases.
2265 Previously in this situation something unpredictable would happen;
2266 sometimes it was "internal error: store_reset failed".
2268 18. If a server dropped the connection unexpectedly when an Exim client was
2269 using GnuTLS and trying to read a response, the client delivery process
2270 crashed while trying to generate an error log message.
2272 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2273 string, for example:
2275 warn message = H1: something\nH2: something
2277 the text was added as a single header line from Exim's point of view
2278 though it ended up OK in the delivered message. However, searching for the
2279 second and subsequent header lines using $h_h2: did not work. This has been
2280 fixed. Similarly, if a system filter added multiple headers in this way,
2281 the routers could not see them.
2283 20. Expanded the error message when iplsearch is called with an invalid key to
2284 suggest using net-iplsearch in a host list.
2286 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2287 ACLs are no longer actually imposed (and a message to that effect is
2290 22. If a "gecos" field in a passwd entry contained escaped characters, in
2291 particular, if it contained a \" sequence, Exim got it wrong when building
2292 a From: or a Sender: header from that name. A second bug also caused
2293 incorrect handling when an unquoted " was present following a character
2294 that needed quoting.
2296 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2297 was not being matched caselessly.
2299 24. Arranged for all hyphens in the exim.8 source to be escaped with
2302 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2303 properly. Recipient callout cache records were still being keyed to include
2304 the sender, even when use_sender was set false. This led to far more
2305 callouts that were necessary. The sender is no longer included in the key
2306 when use_sender is false.
2308 26. Added "control = submission" modifier to ACLs.
2310 27. Added the ${base62d: operator to decode base 62 numbers.
2312 28. dnsdb lookups can now access SRV records.
2314 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2315 the configuration file.
2317 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2318 (-v) mode. This makes the output for a verbose queue run more intelligible.
2320 31. Added a use_postmaster feature to recipient callouts.
2322 32. Added the $body_zerocount variable, containing the number of binary zero
2323 bytes in the message body.
2325 33. The time of last modification of the "new" subdirectory is now used as the
2326 "mailbox time last read" when there is a quota error for a maildir
2329 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2331 35. Added +ignore_unknown as a special item in host lists.
2333 36. Code for decoding IPv6 addresses in host lists is now included, even if
2334 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2335 address was recognized as an IP address, but was then not correctly decoded
2336 into binary, causing unexpected and incorrect effects when compared with
2343 1. Very minor rewording of debugging text in manualroute to say "list of
2344 hosts" instead of "hostlist".
2346 2. If verify=header_syntax was set, and a header line with an unqualified
2347 address (no domain) and a large number of spaces between the end of the
2348 name and the colon was received, the reception process suffered a buffer
2349 overflow, and (when I tested it) crashed. This was caused by some obsolete
2350 code that should have been removed. The fix is to remove it!
2352 3. When running in the test harness, delay a bit after writing a bounce
2353 message to get a bit more predictability in the log output.
2355 4. Added a call to search_tidyup() just before forking a reception process. In
2356 theory, someone could use a lookup in the expansion of smtp_accept_max_
2357 per_host which, without the tidyup, could leave open a database connection.
2359 5. Added the variables $recipient_data and $sender_data which get set from a
2360 lookup success in an ACL "recipients" or "senders" condition, or a router
2361 "senders" option, similar to $domain_data and $local_part_data.
2363 6. Moved the writing of debug_print from before to after the "senders" test
2366 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2367 problems for message scanning, either using a data ACL, or using
2368 local_scan() because the Received: header was not generated till after they
2369 were called (in order to set the time as the time of reception completion).
2370 I have revised the way this works. The header is now generated after the
2371 body is received, but before the ACL or local_scan() are called. After they
2372 are run, the timestamp in the header is updated.
2378 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2379 before starting a queue runner without re-exec. This happened only when
2380 deliver_drop_privilege was set or when the Exim user was set to root. The
2381 effect of the bug was that timeouts during subsequent deliveries caused
2382 crashes instead of being properly handled. The handler is now left at its
2383 default (and expected) setting.
2385 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2386 message, again when deliver_drop_privilege is set or Exim is run as root.
2387 The bug described in (1) was not present in this case, but the tidying up
2388 of the other signals was missing. I have made the two cases consistent.
2390 3. The ignore_target_hosts setting on a manualroute router was being ignored
2391 for hosts that were looked up using the /MX notation.
2393 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2396 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2397 operated on the sender address. After changing the $sender_address to <>
2398 for the sender address verify, Exim was re-instated it as the original
2399 (before rewriting) address, but remembering that it had rewritten it, so it
2400 wasn't rewriting it again. This bug also had the effect of breaking the
2401 sender address verification caching when the sender address was rewritten.
2403 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2404 This has been changed so that if the ip literal address matches
2405 ignore_target_hosts, the router declines.
2407 7. Added expansion conditions match_domain, match_address, and match_local_
2408 part (NOT match_host).
2410 8. The placeholder for the Received: header didn't have a length field set.
2412 9. Added code to Exim itself and to exim_lock to test for a specific race
2413 condition that could lead to file corruption when using MBX delivery. The
2414 issue is with the lockfile that is created in /tmp. If this file is removed
2415 after a process has opened it but before that process has acquired a lock,
2416 there is the potential for a second process to recreate the file and also
2417 acquire a lock. This could lead to two Exim processes writing to the file
2418 at the same time. The added code performs the same test as UW imapd; it
2419 checks after acquiring the lock that its file descriptor still refers to
2420 the same named file.
2422 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2423 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2426 11. Added the smtp_active_hostname option. If used, this will typically be made
2427 to depend on the incoming interface address. Because $interface_address is
2428 not set up until the daemon has forked a reception process, error responses
2429 that can happen earlier (such as "too many connections") no longer contain
2432 12. If an expansion in a condition on a "warn" statement fails because a lookup
2433 defers, the "warn" statement is abandoned, and the next ACL statement is
2434 processed. Previously this caused the whole ACL to be aborted.
2436 13. Added the iplsearch lookup type.
2438 14. Added ident_timeout as a log selector.
2440 15. Added tls_certificate_verified as a log selector.
2442 16. Added a global option tls_require_ciphers (compare the smtp transport
2443 option of the same name). This controls incoming TLS connections.
2445 17. I finally figured out how to make tls_require_ciphers do a similar thing
2446 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2447 before starting the TLS session.
2449 18. Tabs are now shown as \t in -bP output.
2451 19. If the log selector return_path_on_delivery was set, Exim crashed when
2452 bouncing a message because it had too many Received: header lines.
2454 20. If two routers both had headers_remove settings, and the first one included
2455 a superfluous trailing colon, the final name in the first list and the
2456 first name in the second list were incorrectly joined into one item (with a
2457 colon in the middle).
2463 1. Added -C and -D options to the exinext utility, mainly to make it easier
2464 to include in the automated testing, but these could be helpful when
2465 multiple configurations are in use.
2467 2. The exinext utility was not formatting the output nicely when there was
2468 an alternate port involved in the retry record key, nor when there was a
2469 message id as well (for retries that were specific to a specific message
2470 and a specific host). It was also confused by IPv6 addresses, because of
2471 the additional colons they contain. I have fixed the IPv4 problem, and
2472 patched it up to do a reasonable job for IPv6.
2474 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2475 delivery, the log line now contains "pipelined" if PIPELINING was used.
2477 4. An SMTP transport process used to panic and die if the bind() call to set
2478 an explicit outgoing interface failed. This has been changed; it is now
2479 treated in the same way as a connect() failure.
2481 5. A reference to $sender_host_name in the part of a conditional expansion
2482 that was being skipped was still causing a DNS lookup. This no longer
2485 6. The def: expansion condition was not recognizing references to header lines
2486 that used bh_ and bheader_.
2488 7. Added the _cache feature to named lists.
2490 8. The code for checking quota_filecount in the appendfile transport was
2491 allowing one more file than it should have been.
2493 9. For compatibility with Sendmail, the command line option
2501 and sets the incoming protocol and host name (for trusted callers). The
2502 host name and its colon can be omitted when only the protocol is to be set.
2503 Note the Exim already has two private options, -pd and -ps, that refer to
2504 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2505 "s", but I don't think that's a major issue.
2507 10. A number of refactoring changes to the code, none of which should affect
2510 (a) The number of logging options was getting close to filling up the
2511 32-bit word that was used as a bit map. I have split them into two classes:
2512 those that are passed in the argument to log_write(), and those that are
2513 only ever tested independently outside of that function. These are now in
2514 separate 32-bit words, so there is plenty of room for expansion again.
2515 There is no change in the user interface or the logging behaviour.
2517 (b) When building, for example, log lines, the code previously used a
2518 macro that called string_cat() twice, in order to add two strings. This is
2519 not really sufficiently general. Furthermore, there was one instance where
2520 it was actually wrong because one of the argument was used twice, and in
2521 one call a function was used. (As it happened, calling the function twice
2522 did not affect the overall behaviour.) The macro has been replaced by a
2523 function that can join an arbitrary number of extra strings onto a growing
2526 (c) The code for expansion conditions now uses a table and a binary chop
2527 instead of a serial search (which was left over from when there were very
2528 few conditions). Also, it now recognizes conditions like "pam" even when
2529 the relevant support is not compiled in: a suitably worded error message is
2530 given if an attempt is made to use such a condition.
2532 11. Added ${time_interval:xxxxx}.
2534 12. A bug was causing one of the ddress fields not to be passed back correctly
2535 from remote delivery subprocesses. The field in question was not being
2536 subsequently used, so this caused to problems in practice.
2538 13. Added new log selectors queue_time and deliver_time.
2540 14. Might have fixed a bug in maildirsizefile handling that threw up
2541 "unexpected character" debug warnings, and recalculated the data
2542 unnecessarily. In any case, I expanded the warning message to give more
2545 15. Added the message "Restricted characters in address" to the statements in
2546 the default ACL that block characters like @ and % in local parts.
2548 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2549 Three changes have been made:
2551 (a) There was a serious bug; a negative response to MAIL caused the whole
2552 recipient domain to be cached as invalid, thereby blocking all messages
2553 to all local parts at the same domain, from all senders. This bug has
2554 been fixed. The domain is no longer cached after a negative response to
2555 MAIL if the sender used is not empty.
2557 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2560 (c) A new callout option, "use_sender" has been added for people who want
2561 the modified behaviour.
2567 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2570 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2571 indeed breaks things for older releases.
2573 3. Added additional logging to the case where there is a problem reading data
2574 from a filter that is running in a subprocess using a pipe, in order to
2575 try to track down a specific problem.
2577 4. Testing facility fudge: when running in the test harness and attempting
2578 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2579 getting "No route to host". Convert this to a timeout.
2581 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2584 6. Some OS don't have socklen_t but use size_t instead. This affects the
2585 fifth argument of getsockopt() amongst other things. This is now
2586 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2587 can be set for individual OS. I have set it for SunOS5, OSF1, and
2588 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2589 some earlier ones do not.
2591 7. Change 4.30/15 was not doing the test caselessly.
2593 8. The standard form for an IPv6 address literal was being rejected by address
2594 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2595 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2596 this, as well as the form without the "IPv6" on the front (but only when
2597 address literals are enabled, of course).
2599 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2601 10. Exim crashed if a message with an empty sender address specified by -f
2602 encountered a router with an errors_to setting. This could be provoked only
2603 by a command such as
2607 where an empty string was supplied; "<>" did not hit this bug.
2609 11. Installed PCRE release 4.5.
2611 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2612 remained set. It is now erased.
2614 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2615 times from message ids (which are base 36 rather than the normal 62).
2617 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2618 were being counted as actual protocol errors, and logged if the log
2619 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2620 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2621 and DATA following a set of rejected RCPTs do not count as protocol errors.
2622 In other words, Exim assumes they were pipelined, though this may not
2623 actually be the case. Of course, in all cases the client gets an
2624 appropriate error code.
2626 15. If a lookup fails in an ACL condition, a message about the failure may
2627 be available; it is used if testing the ACL cannot continue, because most
2628 such messages specify what the cause of the deferral is. However, some
2629 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2630 that caused an old message to be retained and used if a later statement
2631 caused a defer, replacing the real cause of the deferral.
2633 16. If an IP address had so many PTR records that the DNS lookup buffer
2634 was not large enough to hold them, Exim could crash while trying to process
2635 the truncated data. It now detects and logs this case.
2637 17. Further to 4.21/58, another change has been made: if (and only if) the
2638 first line of a message (the first header line) ends with CRLF, a bare LF
2639 in a subsequent header line has a space inserted after it, so as not to
2640 terminate the header.
2642 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2643 unnecessarily, used atoi() instead of strtol(), and didn't check the
2644 termination when getting file sizes from file names by regex.
2646 19. Completely re-implemented the support for maildirsize files, in the light
2647 of a number of problems with the previous contributed implementation
2648 (4.30/29). In particular:
2650 . If the quota is zero, the maildirsize file is maintained, but no quota is
2653 . If the maildir directory does not exist, it is created before any attempt
2654 to write a maildirsize file.
2656 . The quota value in the file is just a cache; if the quota is changed in
2657 the transport, the new value overrides.
2659 . A regular expression is available for excluding directories from the
2662 20. The autoreply transport checks the characters in options that define the
2663 message's headers; it allows continued headers, but it was checking with
2664 isspace() after an embedded newline instead of explicitly looking for a
2667 21. If all the "regular" hosts to which an address was routed had passed their
2668 expiry times, and had not reached their retry times, the address was
2669 bounced, even if fallback hosts were defined. Now Exim should go on to try
2672 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2673 equivalent code in the SMTP transport. Some hosts send humungous responses
2674 to HELO/EHLO, more than 1024 it seems.
2676 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2677 gives compiler warnings in some environments. I've now done it "properly",
2680 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2681 (because of problems with the built-in one) was declared to return uschar *
2682 instead of char *, causing compiler failure.
2684 25. Fixed a file descriptor leak when processing alias/forward files.
2686 26. Fixed a minor format string issue in dbfn.c.
2688 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2690 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2691 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2694 29. When a sender address is verified, it is cached, to save repeating the test
2695 when there is more than one recipient in a message. However, when the
2696 verification involves a callout, it is possible for different callout
2697 options to be set for different recipients. It is too complicated to keep
2698 track of this in the cache, so now Exim always runs a verification when a
2699 callout is required, relying on the callout cache for the optimization.
2700 The overhead is duplication of the address routing, but this should not be
2703 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2704 to be verified with callout=postmaster, and the main callout worked but the
2705 postmaster check failed, the verification correctly failed. However, if a
2706 subsequent RCPT command asked for sender verification *without* the
2707 postmaster check, incorrect caching caused this verification also to fail,
2710 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2711 it was not caching the DNS options (qualify_single, search_parents) that
2712 were used when the lookup failed. A subsequent lookup with different
2713 options therefore always gave the same answer, though there were cases
2714 where it should not have. (Example: a "domains = !$mx_any" option on a
2715 dnslookup router: the "domains" option is always processed without any
2716 widening, but the router might have qualify_single set.) Now Exim uses the
2717 cached value only when the same options are set.
2719 32. Added John Jetmore's "exipick" utility to the distribution.
2721 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2722 than a timeout (e.g. a certificate is required, and is not provided), an
2723 Exim server now closes the connection immediately. Previously it waited for
2724 the client to close - but if the client is SSL, it seems that they each
2725 wait for each other, leading to a delay before one of them times out.
2727 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2728 maintained 0.8.x compatibility because I don't think many are using it, and
2729 it is clearly obsolete.
2731 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2734 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2735 client certificate was expired. A simple patch fixes this, though I don't
2736 understand the full logic of why the verify callback is called multiple
2739 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2740 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2741 which causes problems with some clients (such as the Certicom SSL Plus
2742 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2743 disables the coutermeasure allowing Eudora to connect."
2745 38. Exim was not checking that a write() to a log file succeeded. This could
2746 lead to Bad Things if a log got too big, in particular if it hit a file
2747 size limit. Exim now panics and dies if it cannot write to a log file, just
2748 as it does if it cannot open a log file.
2750 39. Modified OS/Makefile-Linux so that it now contains
2752 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2754 The two -D definitions ensure that Exim is compiled with large file
2755 support, which makes it possible to handle log files that are bigger than
2758 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2759 instance) a domain was checked against a named list that involved a lookup,
2760 causing $domain_data to be set, then another domain was checked against the
2761 same list, then the first domain was re-checked, the value of $domain_data
2762 after the final check could be wrong. In particular, if the second check
2763 failed, it could be set empty. This bug probably also applied to
2766 41. The strip_trailing_dot option was not being applied to the address given
2767 with the -f command-line option.
2769 42. The code for reading a message's header from the spool was incrementing
2770 $received_count, but never initializing it. This meant that the value was
2771 incorrect (doubled) while delivering a message in the same process in which
2772 it was received. In the most common configuration of Exim, this never
2773 happens - a fresh exec is done - but it can happen when
2774 deliver_drop_privilege is set.
2776 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2777 it now includes up to 150 characters of the unexpected data in the log
2780 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2781 and building data strings. The size of both of these buffers was 10 000
2782 bytes - far larger than anybody would *ever* want, thought I. Needless to
2783 say, somebody hit the limit. I have increased the maximum line length to
2784 20 000 and the maximum data length of concatenated lines to 100 000. I have
2785 also fixed two bugs, because there was no checking on these buffers. Tsk,
2786 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2787 buffer is too small.
2789 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2790 lsearch lookups. Now it does.
2792 46. When parsing a route_list item in a manualroute router, a fixed-length
2793 buffer was used for the list of hosts. I made this 1024 bytes long,
2794 thinking that nobody would ever have a list of hosts that long. Wrong.
2795 Somebody had a whole pile of complicated expansion conditions, and the
2796 string was silently truncated, leading to an expansion error. It turns out
2797 that it is easier to change to an unlimited length (owing to other changes
2798 that have happened since this code was originally written) than to build
2799 structure for giving a limitation error. The length of the item that
2800 expands into the list of hosts is now unlimited.
2802 47. The lsearch lookup could not handle data where the length of text line was
2803 more than 4095 characters. Such lines were truncated, leading to shortened
2804 data being returned. It should now handle lines of any length.
2806 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2807 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2810 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2811 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2812 no longer happens in comment lines. A list of replacements is now placed
2813 at the head of all of the source files, except those whose only change is
2814 to replace PERL_COMMAND in the very first #! line.
2816 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2817 messages on the queue, with a bottom-up merge sort, using code contributed
2818 by Michael Haardt. This should make operations like -bp somewhat faster on
2819 large queues. It won't affect queue runners, except when queue_run_in_order
2822 51. Installed eximstats 1.31 in the distribution.
2824 52. Added support for SRV lookups to the dnslookup router.
2826 53. If an ACL referred to $message_body or $message_body_end, the value was not
2827 reset for any messages that followed in the same SMTP session.
2829 54. The store-handling optimization for building very long strings was not
2830 differentiating between the different store pools. I don't think this
2831 actually made any difference in practice, but I've tidied it.
2833 55. While running the routers to verify a sender address, $sender_address
2834 was still set to the sender address. This is wrong, because when routing to
2835 send a bounce to the sender, it would be empty. Therefore, I have changed
2836 it so that, while verifying a sender address, $sender_address is set to <>.
2837 (There is no change to what happens when verifying a recipient address.)
2839 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2840 target A or AAAA records (if not already returned) without resetting the
2841 qualify_single or search_parents options of the DNS resolver. These are
2842 inappropriate in this case because the targets of MX and SRV records must
2843 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2844 target that, when qualified, matched something in the local domain. These
2845 two options are now turned off when doing these lookups.
2847 57. It seems that at least some releases of Reiserfs (which does not have the
2848 concept of a fixed number of inodes) returns zero and not -1 for the
2849 number of available inodes. This interacted badly with check_spool_inodes,
2850 which assumed that -1 was the "no such thing" setting. What I have done is
2851 to check that the total number of inodes is greater than zero before doing
2852 the test of how many are available.
2854 58. When a "warn" ACL statement has a log_message modifier, the message is
2855 remembered, and not repeated. This is to avoid a lot of repetition when a
2856 message has many recipients that cause the same warning to be written.
2857 Howewer, Exim was preserving the list of already written lines for an
2858 entire SMTP session, which doesn't seem right. The memory is now reset if a
2859 new message is started.
2861 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2862 debugging output unless log_rewrite was also set.
2864 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2865 of (int)(handle) when we know that handle contains (void *)(-1).
2867 61. The Exim daemon panic-logs an error return when it closes the incoming
2868 connection. However "connection reset by peer" seems to be common, and
2869 isn't really an error worthy of noting specially, so that particular error
2872 62. When Exim is trying to find all the local interfaces, it used to panic and
2873 die if the ioctl to get the interface flags failed. However, it seems that
2874 on at least one OS (Solaris 9) it is possible to have an interface that is
2875 included in the list of interfaces, but for which you get a failure error
2876 for this call. This happens when the interface is not "plumbed" into a
2877 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2878 failure of the "get flags" call assumes that the interface is down.
2880 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2881 makes life easier for people who are doing arithmetic on fields extracted
2882 from dates, where you often get leading zeros that should not be
2883 interpreted as octal.
2885 64. Added qualify_domain to the redirect router, to override the global
2888 65. If a pathologically long header line contained very many addresses (the
2889 report of this problem mentioned 10 000) and each of them was rewritten,
2890 Exim could use up a very large amount of memory. (It kept on making new
2891 copies of the header line as it rewrote, and never released the old ones.)
2892 At the expense of a bit more processing, the header rewriting function has
2893 been changed so that it no longer eats memory in this way.
2895 66. The generation of the Received: header has been moved from the time that a
2896 message starts to be received, to the time that it finishes. The timestamp
2897 in the Received: header should now be very close to that of the <= log
2898 line. There are two side-effects of this change:
2900 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2901 logged header lines no longer include the local Received: line, because
2902 it has not yet been created. The same applies to a copy of the message
2903 that is returned to a non-SMTP sender when a message is rejected.
2905 (b) When a filter file is tested using -bf, no additional Received: header
2906 is added to the test message. After some thought, I decided that this
2909 This change does not affect the value of $received_for. It is still set
2910 after address rewriting, but before local_scan() is called.
2912 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2914 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2915 gave an unhelpful panic error message, and a defer error. I have managed to
2916 change this behaviour so that it now rejects any supplied certificate,
2917 which seems right, as the list of acceptable certificates is empty.
2919 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2920 gave an unhelpful defer error. I have not managed to make this reject any
2921 supplied certificates, but the error message it gives is "no certificate
2922 supplied", which is not helpful.
2924 70. exigrep's output now also includes lines that are not associated with any
2925 message, but which match the given pattern. Implemented by a patch from
2926 Martin Sluka, which also tidied up the Perl a bit.
2928 71. Recipient callout verification, like sender verification, was using <> in
2929 the MAIL FROM command. This isn't really the right thing, since the actual
2930 sender may affect whether the remote host accepts the recipient or not. I
2931 have changed it to use the actual sender in the callout; this means that
2932 the cache record is now keyed on a recipient/sender pair, not just the
2933 recipient address. There doesn't seem to be a real danger of callout loops,
2934 since a callout by the remote host to check the sender would use <>.
2935 [SEE ABOVE: changed after hitting problems.]
2937 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2938 temporary errors. However, in the case of such a code being given after
2939 the end of a data transmission (i.e. after ".") Exim was failing to write
2940 a retry record for the message. (Yes, there was some broken host that was
2941 actually sending 8xx at this point.)
2943 73. An unknown lookup type in a host list could cause Exim to panic-die when
2944 the list was checked. (An example that provoked this was putting <; in the
2945 middle of a list instead of at the start.) If this happened during a DATA
2946 ACL check, a -D file could be left lying around. This kind of configuration
2947 error no longer causes Exim to die; instead it causes a defer errror. The
2948 incident is still logged to the main and panic logs.
2950 74. Buglet left over from Exim 3 conversion. The message "too many messages
2951 in one connection" was written to the rejectlog but not the mainlog, except
2952 when address rewriting (yes!) was being logged.
2954 75. Added write_rejectlog option.
2956 76. When a system filter was run not as root (that is, when system_filter_user
2957 was set), the values of the $n variables were not being returned to the
2958 main process; thus, they were not subsequently available in the $sn
2961 77. Added +return_path_on_delivery log selector.
2963 78. A connection timeout was being treated differently from recipients deferred
2964 when testing hosts_max_try with a message that was older than the host's
2965 retry timeout. (The host should not be counted, thus allowing all hosts to
2966 be tried at least once before bouncing.) This may have been the cause of an
2967 occasionally reported bug whereby a message would remain on the queue
2968 longer than the retry timeout, but would be bounced if a delivery was
2969 forced. I say "may" because I never totally pinned down the problem;
2970 setting up timeout/retry tests is difficult. See also the next item.
2972 79. The ultimate address timeout was not being applied to errors that involved
2973 a combination of host plus message (for example, a timeout on a MAIL
2974 command). When an address resolved to a number of possible hosts, and they
2975 were not all tried for each delivery (e.g. because of hosts_max_try), a
2976 message could remain on the queue longer than the retry timeout.
2978 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2981 81. Fixed an obscure SMTP outgoing bug which required at least the following
2982 conditions: (a) there was another message waiting for the same server;
2983 (b) the server returned 5xx to all RCPT commands in the first message so
2984 that the message was not completed; (c) the server dropped the connection
2985 or gave a negative response to the RSET that Exim sends to abort the
2986 transaction. The observed case was a dropped connection after DATA that had
2987 been sent in pipelining mode. That is, the server had advertised PIPELINING
2988 but was not implementing it correctly. The effect of the bug was incorrect
2989 behaviour, such as trying another host, and this could lead to a crash.
2995 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2996 and daemon.c were passed as pointers to ints; they should have been
2997 pointers to socklen_t variables (which are typically unsigned ints).
2999 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3002 3. Fixed a really odd bug that affected only the testing scheme; patching a
3003 certain fixed string in the binary changed the value of another string that
3004 happened to be identical to the end of the original first string.
3006 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3007 name", it returns that address as the IP address. On some operating
3008 systems (e.g. Solaris), it also passes back the IP address string as the
3009 "host name". However, on others (e.g. Linux), it passes back an empty
3010 string. Exim wasn't checking for this, and was changing the host name to an
3011 empty string, assuming it had been canonicized.
3013 5. Although rare, it is permitted to have more than one PTR record for a given
3014 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3015 all the names associated with an address, because they do in Solaris.
3016 However, it seems that they do not in Linux for data that comes from the
3017 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3018 I found this out when I moved to a new Linux workstation and tried to run
3019 the Exim test suite.
3021 To get round this problem I have changed the code so that it now does its
3022 own call to the DNS to look up PTR records when searching for a host name.
3023 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3024 addresses that are only in /etc/hosts are still found.
3026 This behaviour is, however, controlled by an option called host_lookup_
3027 order, which defaults to "bydns:byaddr". If people want to use the other
3028 order, or indeed, just use one or the other means of lookup, they can
3029 specify it in this variable.
3031 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3032 some operating systems, this comes back from gethostbyaddr() as an empty
3033 string, and this is what Exim used to test for. However, it seems that in
3034 other systems, "." is yielded. Exim now tests for this case too.
3036 7. The values of check_spool_space and check_log_space are now held internally
3037 as a number of kilobytes instead of an absolute number of bytes. If a
3038 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3039 kilobyte. This means that much larger values can be stored.
3041 8. Exim monitor: an attempt to get the action menu when not actually pointing
3042 at a message produces an empty menu entitled "No message selected". This
3043 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3044 no entries in it ("Shell widget menu has zero width and/or height"). So I
3045 have added a single, blank menu entry in this case.
3047 9. Added ${quote_local_part.
3049 10. MIME decoding is now applied to the contents of Subject: header lines when
3052 11. Now that a reference to $sender_host_address automatically causes a reverse
3053 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3054 host lookup before query-style lookups in lists that might use this
3055 variable. This has therefore been abolished, and the "net-" prefix is no
3056 longer necessary for query-style lookups.
3058 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3059 have been a typo for LFLAGS, so it has been changed.
3061 13. The install script calls Exim with "-C /dev/null" in order to find the
3062 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3063 to be output. Howeve, since Exim outputs its version number before the
3064 error, it didn't break the script. It just looked ugly. I fixed this by
3065 always allowing "-C /dev/null" if the caller is root.
3067 14. Ignore overlarge ACL variable number when reading spool file - insurance
3068 against a later release with more variables having written the file.
3070 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3071 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3072 this, as well as the form without the "IPv6" on the front.
3074 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3075 OS/Makefile-Darwin file.
3077 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3078 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3080 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3081 variable. However, some OS use ints for inodes. I've added cast to long int
3082 to get rid of the compiler warning.
3084 19. I had forgotten to lock out "/../" in configuration file names when
3085 ALT_CONFIG_PREFIX was set.
3087 20. Routers used for verification do not need to specify transports. However,
3088 if such a router generated a host list, and callout was configured, Exim
3089 crashed, because it could not find a port number from the (non-existent)
3090 transport. It now assumes port 25 in this circumstance.
3092 21. Added the -t option to exigrep.
3094 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3095 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3096 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3097 them is set, LOOKUP_LSEARCH is forced.
3099 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3101 24. Added sender and host information to the "rejected by local_scan()" log
3102 line; previously there was no indication of these.
3104 25. Added .include_if_exists.
3106 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3107 Linux. It turns out that not all file systems support this. Apparently some
3108 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3109 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3110 not supported on the file descriptor, is now ignored when Exim is trying to
3111 sync a directory. This applies only to Linux.
3113 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3115 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3116 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3119 29. Added support for maildirsize files from supplied patch (modified a bit).
3121 30. The use of :fail: followed by an empty string could lead Exim to respond to
3122 sender verification failures with (e.g.):
3124 550 Verification failed for <xxx>
3125 550 Sender verify failed
3127 where the first response line was missing the '-' that indicates it is not
3128 the final line of the response.
3130 31. The loop for finding the name of the user that called Exim had a hardwired
3131 limit of 10; it now uses the value of finduser_retries, which is used for
3132 all other user lookups.
3134 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3137 33. Exim was neglecting to zero errno before one call of strtol() when
3138 expanding a string and expecting an integer value. On some systems this
3139 resulted in spurious "integer overflow" errors. Also, it was casting the
3140 result into an int without checking.
3142 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3143 did not work. The code looks as if it has *never* worked, though it appears
3144 to have been documented since at least releast 1.62. I have made it work.
3146 35. The "timeout_DNS" error in retry rules, also documented since at least
3147 1.62, also never worked. As it isn't clear exactly what this means, and
3148 clearly it isn't a major issue, I have abolished the feature by treating it
3149 as "timeout", and writing a warning to the main and panic logs.
3151 36. The display of retry rules for -brt wasn't always showing the error code
3154 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3155 timeout_connect_A, timeout_connect_MX.
3157 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3158 to the empty sender.
3160 39. The daemon was not analysing the content of -oX till after it had closed
3161 stderr and disconnected from the controlling terminal. This meant that any
3162 syntax errors were only noted on the panic log, and the return code from
3163 the command was 0. By re-arranging the code a little, I've made the
3164 decoding happen first, so such errors now appear on stderr, and the return
3165 code is 1. However, the actual setting up of the sockets still happens in
3166 the disconnected process, so errors there are still only recorded on the
3169 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3170 connections (as happens on some IP stacks) was logged at start up time as
3171 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3172 it from "IPv6 and IPv4", which means that two separate sockets are being
3175 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3176 says whether AF_INET or AF_INET6 was passed as an argument.
3178 42. Exiwhat output was messed up when time zones were included in log
3181 43. Exiwhat now gives more information about the daemon's listening ports,
3182 and whether -tls-on-connect was used.
3184 44. The "port" option of the smtp transport is now expanded.
3186 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3187 silently ignored. Now an error message is written to the main and panic
3190 46. There's a new ACL modifier called "logwrite" which writes to a log file
3191 as soon as it is encountered.
3193 47. Added $local_user_uid and $local_user_gid at routing time.
3195 48. Exim crashed when trying to verify a sender address that was being
3198 49. Exim was recognizing only a space character after ".include". It now also
3199 recognizes a tab character.
3201 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3202 extracting the relevant information from the specification. The man page no
3203 longer contains scrambled data for the -d option, and I've added a section
3204 at the front about calling Exim under different names.
3206 51. Added "extra_headers" argument to the "mail" command in filter files.
3208 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3211 53. Installed eximstats 1.29.
3213 54. Added transport_filter_timeout as a generic transport option.
3215 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3216 Cc: header lines. This was required by RFC 822, but it not required by RFC
3219 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3220 incoming messages that did not have them. Now it does so only if the
3221 message originates locally, that is, if there is no associated remote host
3222 address. When Resent- header lines are present, this applies to the Resent-
3223 lines rather than the non-Resent- lines.
3225 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3226 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3228 58. Messages for configuration errors now include the name of the main
3229 configuration file - useful now that there may be more than one file in a
3230 list (.included file names were always shown).
3232 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3233 for those rare installations that do not start the daemon as root or run it
3234 setuid root. I've cut out the call to initgroups() if the daemon is not
3237 60. The Exim user and group can now be bound into the binary as text strings
3238 that are looked up at the start of Exim's processing.
3240 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3242 62. Added $mailstore_basename variable.
3244 63. Installed patch to sieve.c from Michael Haardt.
3246 64. When Exim failed to open the panic log after failing to open the main log,
3247 the original message it was trying to log was written to stderr and debug
3248 output, but if they were not available (the usual case in production), it
3249 was lost. Now it is written to syslog before the two lines that record the
3250 failures to open the logs.
3252 65. Users' Exim filters run in subprocesses under the user's uid. It is
3253 possible for a "deliver" command or an alias in a "personal" command to
3254 provoke an address rewrite. If logging of address rewriting is configured,
3255 this fails because the process is not running as root or exim. There may be
3256 a better way of dealing with this, but for the moment (because 4.30 needs
3257 to be released), I have disabled address rewrite logging when running a
3258 filter in a non-root, non-exim process.
3264 1. The buildconfig auxiliary program wasn't quoting the value set for
3265 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3266 not defined. This bug was masked in 4.22 by the effect that was fixed in
3269 2. Some messages that were rejected after a message id was allocated were
3270 shown as "incomplete" by exigrep. It no longer does this for messages that
3271 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3273 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3274 have allow_domain_literals set, the ID did not get logged in the <= line.
3275 Domain literals are now always recognized in Message-ID: header lines.
3277 4. The first argument for a ${extract expansion item is the key name or field
3278 number. Leading and trailing spaces in this item were not being ignored,
3279 causing some misleading effects.
3281 5. When deliver_drop_privilege was set, single queue runner processes started
3282 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3283 same command in the process it spins off) were not dropping privilege.
3285 6. When the daemon running as "exim" started a queue runner, it always
3286 re-executed Exim in the spun-off process. This is a waste of effort when
3287 deliver_drop_privilege is set. The new process now just calls the
3288 queue-runner function directly.
3294 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3297 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3298 ignored. Though the use of -oP was forcing the writing of a pid file, it
3299 was always written to the default place.
3301 3. If the message "no IP address found for host xxxx" is generated during
3302 incoming verification, it is now followed by identification of the incoming
3303 connection (so you can more easily find what provoked it).
3305 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3307 5. Added some features to "harden" Exim a bit more against certain attacks:
3309 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3310 be put in Local/Makefile. This is like the never_users runtime option,
3311 but it cannot be overridden. The default setting is "root".
3313 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3314 prefix string with which any file named in a -C command line option
3317 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3318 is retained for -C and -D only if the caller of Exim is root. Without
3319 it, the exim user may also use -C and -D and retain privilege.
3321 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3322 command line option is disabled.
3324 6. Macro names set by the -D option must start with an upper case letter, just
3325 like macro names defined in the configuration file.
3327 7. Added "dereference=" facility to LDAP.
3329 8. Two instances of the typo "uknown" in the source files are fixed.
3331 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3332 the Configure-Makefile script screwed up while processing it.
3334 10. Incorporated PCRE 4.4.
3336 11. The SMTP synchronization check was not operating right at the start of an
3337 SMTP session. For example, it could not catch a HELO sent before the client
3338 waited for the greeting. There is now a check for outstanding input at the
3339 point when the greeting is written. Because of the duplex, asynchronous
3340 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3341 way, but not yet received, when the check is performed.
3343 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3344 on TCP/IP sockets, because this apparently causes some broken clients to
3347 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3348 unchanged) from the Cygwin maintainer.
3350 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3351 was supported instead of testing for mbx. Effectively a typo.
3353 15. The spa authenticator server code was not checking that the input it
3354 received was valid base64.
3356 16. The debug output line for the "set" modifier in ACLs was not showing the
3357 name of the variable that was being set.
3359 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3361 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3362 lookup on its own. Something else was needed to trigger the lookup. For
3363 example, a match in host_lookup or the need for a host name in a host list.
3364 Now, if $sender_host_name is referenced and the host name has not yet been
3365 looked up, a lookup is performed. If the lookup fails, the variable remains
3366 empty, and $host_lookup_failed is set to "1".
3368 19. Added "eqi" as a case-independent comparison operator.
3370 20. The saslauthd authentication condition could segfault if neither service
3371 nor realm was specified.
3373 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3374 error message that was logged was misleading, and incoming SMTP
3375 connections were dropped. The message is now more accurate, and temporary
3376 errors are given to SMTP connections.
3378 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3379 (except RSET) with a 421 error, until QUIT is received. However, it was
3380 failing to send a response to QUIT.
3382 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3383 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3384 ACL is now run after the helo_try_verify_hosts code.
3386 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3387 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3388 case-independent, so other case variants are also recognized). Apparently
3389 some systems use these upper case variants.
3391 25. If more than two messages were waiting for the same host, and a transport
3392 filter was specified for the transport, Exim sent two messages over the
3393 same TCP/IP connection, and then failed with "socket operation on non-
3394 socket" when it tried to send the third.
3396 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3398 27. The extern definition of crypt16() in expand.c was not being excluded when
3399 the OS had its own crypt16() function.
3401 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3402 as a preferred synonym for return_size_limit, both as an option and as an
3405 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3407 30. Changed the default configuration ACL to relax the local part checking rule
3408 for addresses that are not in any local domains. For these addresses,
3409 slashes and pipe symbols are allowed within local parts, but the sequence
3410 /../ is explicitly forbidden.
3412 31. SPA server authentication was not clearing the challenge buffer before
3415 32. log_message in a "warn" ACL statement was writing to the reject log as
3416 well as to the main log, which contradicts the documentation and doesn't
3417 seem right (because no rejection is happening). So I have stopped it.
3419 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3420 However, I am unable to do any testing of this.
3422 34. Fixed an infelicity in the appendfile transport. When checking directories
3423 for a mailbox, to see if any needed to be created, it was accidentally
3424 using path names with one or more superfluous leading slashes; tracing
3425 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3427 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3428 contents are added to the log line that is written for every discarded
3429 recipient. (Previously a log_message setting was ignored.)
3431 36. The ${quote: operator now quotes the string if it is empty.
3433 37. The install script runs exim in order to find its version number. If for
3434 some reason other than non-existence or emptiness, which it checks, it
3435 could not run './exim', it was installing it with an empty version number,
3436 i.e. as "exim-". This error state is now caught, and the installation is
3439 38. An argument was missing from the function that creates an error message
3440 when Exim fails to connect to the socket for saslauthd authentication.
3441 This could cause Exim to crash, or give a corrupted message.
3443 39. Added isip, isip4, and isip6 to ${if conditions.
3445 40. The ACL variables $acl_xx are now saved with the message, and can be
3446 accessed later in routers, transports, and filters.
3448 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3449 strings in the file are not string-expanded.
3451 42. If a MAIL command specified a SIZE value that was too large to fit into an
3452 int variable, the check against message_size_limit failed. Such values are
3453 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3454 day this will have to be increased, but I don't think I want to be around
3455 when emails are that large.
3462 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3463 iconv() is not standard in FreeBSD.
3465 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3466 IPv6 enabled. The observed symptom was a segmentation fault on return from
3467 the function os_common_find_running_interfaces() in src/os.c.
3469 3. In the check_special_case() function in daemon.c I had used "errno" as an
3470 argument name, which causes warnings on some systems. This was basically a
3471 typo, since it was named "eno" in the comments!
3473 4. The code that waits for the clock to tick (at a resolution of some fraction
3474 of a second) so as to ensure message-id uniqueness was always waiting for
3475 at least one whole tick, when it could have waited for less. [This is
3476 almost certainly not relevant at current processor speeds, where it is
3477 unlikely to ever wait at all. But we try to future-proof.]
3479 5. The function that sleeps for a time interval that includes fractions of a
3480 second contained a race. It did not block SIGALRM between setting the
3481 timer, and suspending (a couple of lines later). If the interval was short
3482 and the sigsuspend() was delayed until after it had expired, the suspension
3483 never ended. On busy systems this could lead to processes getting stuck for
3486 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3487 process, before it forks any delivery processes. The open lookup caching
3488 mechanism meant that the open file or database connection was passed into
3489 the delivery process. The problem was that delivery processes always tidy
3490 up cached lookup data. This could cause a problem for the next delivery
3491 process started by the queue runner, because the external queue runner
3492 process does not know about the closure. So the next delivery process
3493 still has data in the lookup cache. In the case of a file lookup, there was
3494 no problem because closing a file descriptor in a subprocess doesn't affect
3495 the parent. However, if the lookup was caching a connection to a database,
3496 the connection was closed, and the second delivery process was likely to
3497 see errors such as "PGSQL: query failed: server closed the connection
3498 unexpectedly". The problem has been fixed by closing all cached lookups
3499 in a queue runner before running a delivery process.
3501 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3502 seem to have the "const" qualifier which it has on other OS. I've
3505 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3506 *of the same type* (client or server) with the same public name that an
3507 error should be diagnosed.
3509 9. When Exim looked up a host name for an IP address, but failed to find the
3510 original IP address when looking up the host name (a safety check), it
3511 output the message "<ip address> does not match any IP for NULL", which was
3512 confusing, to say the least. The bug was that the host name should have
3513 appeared instead of "NULL".
3515 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3516 user that is built into the binary, and the -C or -D options is used, root
3517 privilege is dropped before the configuration file is read. In addition,
3518 logging is switched to stderr instead of the normal log files. If the
3519 configuration then re-defines the Exim user, the unprivileged environment
3520 is probably not what is expected, so Exim logs a panic warning message (but
3523 However, if deliver_drop_privilege is set, the unprivileged state may well
3524 be exactly what is intended, so the warning has been cut out in that case,
3525 and Exim is allowed to try to write to its normal log files.
3531 1. smtp_return_error_details was not giving details for temporary sender
3532 or receiver verification errors.
3534 2. Diagnose a configuration error if two authenticators have the same public
3537 3. Exim used not to create the message log file for a message until the first
3538 delivery attempt. This could be confusing when incoming messages were held
3539 for policy or load reasons. The message log file is now created at the time
3540 the message is received, and an initial "Received" line is written to it.
3542 4. The automatically generated man page for command line options had a minor
3543 bug that caused no ill effects; however, a more serious problem was that
3544 the procedure for building the man page automatically didn't always
3545 operate. Consequently, release 4.20 contains an out-of-date version. This
3546 shouldn't happen again.
3548 5. When building Exim with embedded Perl support, the script that builds the
3549 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3550 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3552 6. The freeze_tell option was not being used for messages that were frozen on
3553 arrival, either by an ACL or by local_scan().
3555 7. Added the smtp_incomplete_transaction log selector.
3557 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3558 was accepting AUTH without a new EHLO.
3560 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3561 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3564 10. Logging of TCP/IP connections (when configured) now happens in the main
3565 daemon process instead of the child process, so that the TCP/IP connection
3566 count is more accurate (but it can never be perfect).
3568 11. The use of "drop" in a nested ACL was not being handled correctly in the
3569 outer ACL. Now, if condition failure induced by the nested "drop" causes
3570 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3571 or "require"), the connection is dropped.
3573 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3574 that yield "discard" can now be used with an "accept" or a "discard" verb,
3575 but an error is generated for any others (because I can't see a useful way
3576 to define what should happen).
3578 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3579 are now processed in the same way as lines in the Exim configuration file.
3580 In particular, continuation lines are supported.
3582 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3584 15. Added -ti meaning -t -i.
3586 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3587 domains, and warn by logging if others are found.
3589 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3590 returned from GIFCONF when Exim is trying to find the list of interfaces on
3591 a host. The code in os.c has been modified to copy each ifreq to an aligned
3592 structure in all cases.
3594 Also, in some cases, the returned ifreq's were being copied to a 'struct
3595 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3596 means the last couple of bytes of an IPv6 address could be chopped if the
3597 ifreq contained only a normal sockaddr (14 bytes storage).
3599 18. Named domain lists were not supported in the hosts_treat_as_local option.
3600 An entry such as +xxxx was not recognized, and was treated as a literal
3603 19. Ensure that header lines added by a DATA ACL are included in the reject log
3604 if the ACL subsequently rejects the message.
3606 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3607 MD5 (which is deprecated).
3609 21. When testing a filter file using -bf, Exim was writing a message when it
3610 took the sender from a "From " line in the message, but it was not doing so
3611 when it took $return_path from a Return-Path: header line. It now does.
3613 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3614 with a valid header line field name (a series of printing characters
3615 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3617 23. Changed "disc" in the source to "disk" to conform to the documentation and
3618 the book and for uniformity.
3620 24. Ignore Sendmail's -Ooption=value command line item.
3622 25. When execve() failed while trying to run a command in a pipe transport,
3623 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3624 could be confused with a return value of 69 from the command itself. This
3625 has been changed to 127, the value the shell returns if it is asked to run
3626 a non-existent command. The wording for the related log line suggests a
3627 non-existent command as the problem.
3629 26. If received_header_text expands to an empty string, do not add a Received:
3630 header line to the message. (Well, it adds a token one on the spool, but
3631 marks it "old" so that it doesn't get used or transmitted.)
3633 27. Installed eximstats 1.28 (addition of -nt option).
3635 28. There was no check for failure on the call to getsockname() in the daemon
3636 code. This can fail if there is a shortage of resources on the system, with
3637 ENOMEM, for example. A temporary error is now given on failure.
3639 29. Contrary to the C standard, it seems that in some environments, the
3640 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3641 program. Exim now does this explicitly; it affects the formatting of
3642 timestamps using strftime().
3644 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3645 complaints. I've now initialized all the variables, to avoid this.
3647 32. Header lines added by a system filter were not being "seen" during
3648 transport-time rewrites.
3650 33. The info_callback() function passed to OpenSSL is set up with type void
3651 (*)(SSL *, int, int), as described somewhere. However, when calling the
3652 function (actually a macro) that sets it up, the type void(*)() is
3653 expected. I've put in a cast to prevent warnings from picky compilers.
3655 34. If a DNS black list lookup found a CNAME record, but there were no A
3656 records associated with the domain it pointed at, Exim crashed.
3658 35. If a DNS black list lookup returned more than one A record, Exim ignored
3659 all but the first. It now scans all returned addresses if a particular IP
3660 value is being sought. In this situation, the contents of the
3661 $dnslist_value variable are a list of all the addresses, separated by a
3664 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3665 to accept a host name and all its aliases if the forward lookup for any of
3666 them yielded the IP address of the incoming connection. Now it accepts only
3667 those names whose forward lookup yields the correct IP address. Any other
3668 names are discarded. This closes a loophole whereby a rogue DNS
3669 administrator could create reverse DNS records to break through a
3670 wildcarded host restriction in an ACL.
3672 37. If a user filter or a system filter that ran in a subprocess used any of
3673 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3674 the wrong values were passed to the pipe command ($thisaddress had the
3675 value of $0, $0 had the value of $1, etc). This bug was introduced by
3676 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3678 38. Improved the line breaking for long SMTP error messages from ACLs.
3679 Previously, if there was no break point between 40 and 75 characters, Exim
3680 left the rest of the message alone. Two changes have been made: (a) I've
3681 reduced the minimum length to 35 characters; (b) if it can't find a break
3682 point between 35 and 75 characters, it looks ahead and uses the first one
3683 that it finds. This may give the occasional overlong line, but at least the
3684 remaining text gets split now.
3686 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3687 file descriptors might be low, and that setting 1000 would always raise it.
3688 It turns out that in some environments, the limit is already over 1000 and
3689 that lowering it causes trouble. So now Exim takes care not to decrease it.
3691 40. When delivering a message, the value of $return_path is set to $sender_
3692 address at the start of routing (routers may change the value). By an
3693 oversight, this default was not being set up when an address was tested by
3694 -bt or -bv, which affected the outcome if any router or filter referred to
3697 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3698 log only when "message" or "log_message" are set. However, if one of the
3699 conditions was an address verification, or a call to a nested ACL, the
3700 messages generated by the underlying test were being passed through. This
3701 no longer happens. The underlying message is available in $acl_verify_
3702 message for both "message" and "log_message" expansions, so it can be
3703 passed through if needed.
3705 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3706 new expansion $bh_ to give the encoded byte string without charset
3707 translation. Translation happens only if iconv() is available; HAVE_ICONV
3708 indicates this at build time. HEADERS_CHARSET gives the charset to
3709 translate to; headers_charset can change it in the configuration, and
3710 "headers charset" can change it in an individual filter file.
3712 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3713 that creates RFC 2047 encoded "words" labels them as that charset instead
3714 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3715 expansion operator; (ii) when Exim creates a From: line for a local
3716 message; (iii) when a header line is rewritten to include a "phrase" part.
3718 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3719 buggy, causing it to skip the first lines of messages whose message ID
3720 ended in 'D'. This would not have bitten before Exim release 4.14, because
3721 message IDs were unlikely to end in 'D' before then. The effect was to have
3722 incorrect size information for certain domains.
3724 45. #include "config.h" was missing at the start of the crypt16.c module. This
3725 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3728 46. If there was a timeout during a "random" callout check, Exim treated it as
3729 a failure of the random address, and carried on sending RSET and the real
3730 address. If the delay was just some slowness somewhere, the response to the
3731 original RCPT would be taken as a response to RSET and so on, causing
3732 mayhem of various kinds.
3734 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3735 when I implemented it. It didn't allow for the fact that some option values
3736 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3737 the right test for positive values.
3739 48. Domain names in DNS records are case-independent. Exim always looks them up
3740 in lower case. Some resolvers return domain names in exactly the case they
3741 appear in the zone file, that is, they may contain uppercase letters. Not
3742 all resolvers do this - some return always lower case. Exim was treating a
3743 change of case by a resolver as a change of domain, similar to a widening
3744 of a domain abbreviation. This triggered its re-routing code and so it was
3745 trying to route what was effectively the same domain again. This normally
3746 caused routing to fail (because the router wouldn't handle the domain
3747 twice). Now Exim checks for this case specially, and just changes the
3748 casing of the domain that it ultimately uses when it transmits the message
3751 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3754 50. If a filter generated a file delivery with a non-absolute name (possible if
3755 no home directory exists for the router), the forbid_file option was not
3758 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3759 the existing equality matching.
3761 52. Exim was using ints instead of ino_t variables in some places where it was
3762 dealing with inode numbers.
3764 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3765 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3766 called TMPDIR, and if it finds it is different, it changes its value.
3768 54. The smtp_printf() function is now made available to local_scan() so
3769 additional output lines can be written before returning. There is also an
3770 smtp_fflush() function to enable the detection of a dropped connection.
3771 The variables smtp_input and smtp_batched_input are exported to
3774 55. Changed the default runtime configuration: the message "Unknown user"
3775 has been removed from the ACL, and instead placed on the localuser router,
3776 using the cannot_route_message feature. This means that any verification
3777 failures that generate their own messages won't get overridden. Similarly,
3778 the "Unrouteable address" message that was in the ACL for unverifiable
3779 relay addresses has also been removed.
3781 56. Added hosts_avoid_esmtp to the smtp transport.
3783 57. The exicyclog script was not checking for the esoteric option
3784 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3785 will work only if exicyclog is run under the appropriate euid.
3787 58. Following a discussion on the list, the rules by which Exim recognises line
3788 endings on incoming messages have been changed. The -dropcr and drop_cr
3789 options are now no-ops, retained only for backwards compatibility. The
3790 following line terminators are recognized: LF CRLF CR. However, special
3791 processing applies to CR:
3793 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3794 nor a local message in the state where . is a terminator.
3796 (ii) If a bare CR is encountered in a header line, an extra space is added
3797 after the line terminator so as not to end the header. The reasoning
3798 behind this is that bare CRs in header lines are most likely either
3799 to be mistakes, or people trying to play silly games.
3801 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3802 was being incorrectly given as 18 bytes larger than it should have been.
3803 This is a VOB (very old bug).
3805 60. This may never have affected anything current, but just in case it has:
3806 When the local host is found other than at the start of a list of hosts,
3807 the local host, those with the same MX, and any that follow, are discarded.
3808 When the list in question was part of a longer list of hosts, the following
3809 hosts (not currently being processed) were also being discarded. This no
3810 longer happens. I'm not sure if this situation could ever has previously
3813 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3816 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3817 header. This is something that is recommended in a new Internet Draft, and
3818 is something that is documented as being done by Sendmail. There are two
3819 possible values. For messages generated by the autoreply transport, Exim
3822 Auto-Submitted: auto-replied
3824 whereas for all other generated messages (e.g. bounces) it adds
3826 Auto-Submitted: auto-generated
3828 63. The "personal" condition in filters now includes a test for the
3829 Auto-Submitted: header. If it contains the string "auto-" the message it
3830 not considered personal.
3832 64. Added rcpt_include_affixes as a generic transport option.
3834 65. Added queue_only_override (default true).
3836 66. Added the syslog_duplication option.
3838 67. If what should have been the first header line of a message consisted of
3839 a space followed by a colon, Exim was mis-interpreting it as a header line.
3840 It isn't of course - it is syntactically invalid and should therefore be
3841 treated as the start of the message body. The misbehaviour could have
3842 caused a number of strange effects, including loss of data in subsequent
3843 header lines, and spool format errors.
3845 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3846 client host had authenticated. This control can now be exercised by an ACL
3847 for more flexibility.
3849 69. By default, callouts do not happen when testing with -bh. There is now a
3850 variant, -bhc, which does actually run the callout code, including
3851 consulting and updating the callout cache.
3853 70. Added support for saslauthd authentication, courtesy of Alexander
3856 71. If statvfs() failed on the spool or log directories while checking their
3857 size for availability, Exim confusingly gave the error "space shortage".
3858 Furthermore, in debugging mode it crashed with a floating point exception.
3859 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3860 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3861 serious problem, Exim now writes to the main and panic logs when this
3862 happens, with details of the failure. It then refuses to accept the
3863 incoming message, giving the message "spool directory problem" or "log
3864 directory problem" with a 421 code for SMTP messages.
3866 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3867 0, 1, and 2 exist, because some OS complain for execs without them (see
3868 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3869 descriptors. However, the code omitted to check that the open succeeded,
3870 causing mysterious errors if for some reason the permissions on /dev/null
3871 got screwed. Now Exim writes a message to the main and panic logs, and
3872 bombs out if it can't open /dev/null.
3874 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3875 interact so that it is all more flexible. It is supposed to remain
3876 backwards compatible. Also added extra_local_interfaces.
3878 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3879 to bomb out with an assertion failure - to the client this appears as a
3880 connection drop. This problem occurs in the part of the code that was taken
3881 from the Samba project. Fortunately, the assertion is in a very simple
3882 function, so I have fixed this by reproducing the function inline in the
3883 one place where it is called, and arranging for authentication to fail
3884 instead of killing the process with assert().
3886 75. The SPA client code was not working when the server requested OEM rather
3887 than Unicode encoding.
3889 76. Added code to make require_files with a specific uid setting more usable in
3890 the case where statting the file as root fails - usually a non-root-mounted
3891 NFS file system. When this happens and the failure is EACCES, Exim now
3892 forks a subprocess and does the per-uid checking as the relevant uid.
3894 77. Added process_log_path.
3896 78. If log_file_path was not explicitly set, a setting of check_log_space or
3897 check_log_inodes was ignored.
3899 79. If a space check for the spool or log partitions fails, the incident is now
3900 logged. Of course, in the latter case the data may get lost...
3902 80. Added the %p formatting code to string_format() so that it can be used to
3903 print addresses in debug_print(). Adjusted all the address printing in the
3904 debugging in store.c to use %p rather than %d.
3906 81. There was a concern that a line of code in smtp_in.c could overflow a
3907 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3908 initially expressed, the concern was not well-founded, because trailing
3909 spaces are removed early. However, if the trailing spaces were followed by
3910 a NULL, they did not get removed, so the overflow was possible. Two fixes
3913 (a) I re-wrote the offending code in a cleaner fashion.
3914 (b) If an incoming SMTP command contains a NULL character, it is rejected
3917 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3918 runs initgroups(), so that if the Exim user is in any additional groups,
3919 they will be used during message reception.
3925 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob