1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.375 2006/07/14 14:00:16 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 SC/01 Use a glob alias rather than an array ref in eximstats generated
10 parser. This improves both readability and performance.
12 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
13 Don't display local sender or destination tables in eximstats unless
14 there is data to show.
15 Added average volumes into the eximstats top table text output.
17 SC/03 Collect data on the number of addresses (recipients) as well
18 as the number of messages in eximstats.
20 TF/01 Correct an error in the documentation for the redirect router. Exim
21 does (usually) call initgroups() when daemonizing.
23 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
24 with consistent privilege compared to when running as a daemon.
26 TF/03 Note in the spec that $authenticated_id is not set for local
27 submissions from trusted users.
29 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
30 Thanks to Dean Brooks <dean@iglou.com> for the patch.
32 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
33 by adding some example configuration directives to the default
34 configuration file. A little bit of work is required to uncomment the
35 directives and define how usernames and passwords are checked, but
36 there is now a framework to start from.
38 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
39 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
40 without this. I don't know how relevant this is to other LDAP libraries.
42 PH/02 Add the verb name to the "unknown ACL verb" error.
44 PH/03 Magnus Holmgren's patch for filter_prepend_home.
46 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
48 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
49 directory not expanded when it should be if an expanded home directory
50 was set for the address (which is overridden by the transport).
52 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
55 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
56 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
57 because it is too late at that time, and has no effect.
59 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
60 security issue with \' (bugzilla #107). I could not use the
61 PQescapeStringConn() function, because it needs a PGconn value as one of
64 PH/08 When testing addresses using -bt, indicate those final addresses that
65 are duplicates that would not cause an additional delivery. At least one
66 person was confused, thinking that -bt output corresponded to deliveries.
67 (Suppressing duplicates isn't a good idea as you lose the information
68 about possibly different redirections that led to the duplicates.)
70 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
71 systems where poll() doesn't work, in particular OS X.
73 PH/10 Added more information to debugging output for retry time not reached.
75 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
76 operations in malware.c.
78 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
81 PH/13 If write_rejectlog was set false when logging was sent to syslog with
82 syslog_duplication set false, log lines that would normally be written
83 both the the main log and to the reject log were not written to syslog at
86 PH/14 In the default configuration, change the use of "message" in ACL warn
87 statements to "add_header".
89 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
90 not followed by a command (e.g. "seen endif").
92 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
93 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
96 PH/17 Added an extra condition to the default value of delay_warning_condition
100 { match{$h_precedence:}{(?i)bulk|list|junk} }
101 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} }
104 The Auto-Submitted: header is standardised, whereas I don't think
105 Precedence: ever was.
107 PH/18 Refactored debugging code in route_finduser() to show more information,
108 in particular, the error code if getpwnam() issues one.
114 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
115 other effects) broke the use of negated acl sub-conditions.
117 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
120 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
121 "Deny" causes Exim to reject the incoming connection with a 554 error.
122 Unfortunately, if there is a major crisis, such as a disk failure,
123 tcp-wrappers gives "deny", whereas what one would like would be some
124 kind of temporary error. A kludge has been added to help with this.
125 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
126 554 error is used if errno is still zero or contains ENOENT (which occurs
127 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
130 PH/03 Add -lutil to the default FreeBSD LIBS setting.
132 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
133 errors. Otherwise a message that provokes a temporary error (when other
134 messages do not) can cause a whole host to time out.
136 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
137 addresses were routed directly to files or pipes from a redirect router.
138 File deliveries just didn't batch; pipe deliveries might have suffered
141 PH/06 A failure to get a lock for a hints database would erroneously always say
142 "Failed to get write lock", even when it was really a read lock.
144 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
145 of 0600. This has been changed to use the value of the lockfile_mode
146 option (which defaults to 0600).
148 PH/08 Applied small patch from the Sieve maintainer.
150 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
151 folder from quota calculations, a direct delivery into this folder messed
152 up the contents of the maildirsize file. This was because the regex was
153 used only to exclude .Trash (or whatever) when the size of the mailbox
154 was calculated. There was no check that a delivery was happening into an
155 excluded directory. This bug has been fixed by ignoring all quota
156 processing for deliveries into excluded directories.
158 PH/10 Added the maildirfolder_create_regex option to appendfile.
164 PH/01 The code for finding all the local interface addresses on a FreeBSD
165 system running IPv6 was broken. This may well have applied to all BSD
166 systems, as well as to others that have similar system calls. The broken
167 code found IPv4 interfaces correctly, but gave incorrect values for the
168 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
169 that it would not match correctly against @[] and not recognize the IPv6
172 PH/02 The ipliteral router was not recognizing addresses of the form user@
173 [ipv6:....] because it didn't know about the "ipv6:" prefix.
175 PH/03 Added disable_ipv6.
177 PH/04 Changed $reply_address to use the raw form of the headers instead of the
178 decoded form, because it is most often used to construct To: headers
179 lines in autoreplies, and the decoded form may well be syntactically
180 invalid. However, $reply_address has leading white space removed, and all
181 newlines turned into spaces so that the autoreply transport does not
184 PH/05 If group was specified without a user on a router, and no group or user
185 was specified on a transport, the group from the router was ignored.
187 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
188 for visible compile-time settings that can be used to change these
189 numbers, for those that want even more. Backwards compatibility with old
190 spool files has been maintained. However, going back to a previous Exim
191 release will lost any variables that are in spool files.
193 PH/07 Two small changes when running in the test harness: increase delay when
194 passing a TCP/IP connection to a new process, in case the original
195 process has to generate a bounce, and remove special handling of
196 127.0.0.2 (sic), which is no longer necessary.
198 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
199 be the same on different OS.
201 PH/09 Moved a debug statement in filter processing to avoid a race problem when
204 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
205 whether --show-vars was specified or not
207 JJ/02 exipick: Added support for new ACL variable spool format introduced
210 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
211 syntactically invalid From: or Reply-to: line, and a filter used this to
212 generate an autoreply, and therefore failed to obtain an address for the
213 autoreply, Exim could try to deliver to a non-existent relative file
214 name, causing unrelated and misleading errors. What now happens is that
215 it logs this as a hard delivery error, but does not attempt to create a
218 PH/11 The exinext utility has a -C option for testing purposes, but although
219 the given file was scanned by exinext itself; it wasn't being passed on
220 when Exim was called.
222 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
223 an end-of-file indication when reading a command response.
225 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
226 compiled. In many other places in Exim, IPv6 addresses are always
227 recognized, so I have changed this. It also means that IPv4 domain
228 literals of the form [IPV4:n.n.n.n] are now always recognized.
230 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
231 used if the router is not running as root, for example, when verifying at
232 ACL time, or when using -bh. The debugging output from this situation was
233 non-existent - all you got was a failure to exec. I have made two
236 (a) Failures to set uid/gid, the current directory, or a process leader
237 in a subprocess such as that created by queryprogram now generate
238 suitable debugging ouput when -d is set.
240 (b) The queryprogram router detects when it is not running as root,
241 outputs suitable debugging information if -d is set, and then runs
242 the subprocess without attempting to change uid/gid.
244 PH/15 Minor change to Makefile for building test_host (undocumented testing
247 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
248 additional section of a DNS packet that returns MX or SRV records.
249 Instead, it always explicitly searches for A/AAAA records. This avoids
250 major problems that occur when a DNS server includes only records of one
251 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
252 fixed another bug: if SRV records were looked up and the corresponding
253 address records were *not* found in the additional section, the port
254 values from the SRV records were lost.
256 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
257 using the correct key (the original address) when searching the retry
258 rules in order to find which one to use for generating the retry hint.
260 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
261 adding the default one. Similarly, if it contains a Reply-To: header, the
262 errors_reply_to option, if set, is not used.
264 PH/19 When calculating a retry time, Exim used to measure the "time since
265 failure" by looking at the "first failed" field in the retry record. Now
266 it does not use this if it is later than than the arrival time of the
267 message. Instead it uses the arrival time. This makes for better
268 behaviour in cases where some deliveries succeed, thus re-setting the
269 "first failed" field. An example is a quota failure for a huge message
270 when small messages continue to be delivered. Without this change, the
271 "time since failure" will always be short, possible causing more frequent
272 delivery attempts for the huge message than are intended.
273 [Note: This change was subsequently modified - see PH/04 for 4.62.]
275 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
276 $1, $2, $3) because the numerical variables can be reset during some
277 expansion items (e.g. "match"), thereby losing the authentication data.
279 PH/21 Make -bV show the size of off_t variables so that the test suite can
280 decide whether to run tests for quotas > 2G.
282 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
283 mailbox_size, and mailbox_filecount in the appendfile transport. If a
284 filecount value is greater than 2G or if a quota value is greater than 2G
285 on a system where the size of off_t is not greater than 4, a panic error
288 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
289 never match. The debug and -bh output now contains an explicit error
290 message indicating a malformed IPv4 address or mask.
292 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
293 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
296 PH/25 Do not write to syslog when running in the test harness. The only
297 occasion when this arises is a failure to open the main or panic logs
298 (for which there is an explicit test).
300 PH/26 Added the /no_tell option to "control=freeze".
302 PH/27 If a host name lookup failed very early in a connection, for example, if
303 the IP address matched host_lookup and the reverse lookup yielded a name
304 that did not have a forward lookup, an error message of the form "no IP
305 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
306 could be logged. Now it outputs the IP address instead of "NULL".
308 PH/28 An enabling patch from MH: add new function child_open_exim2() which
309 allows the sender and the authenticated sender to be set when
310 submitting a message from within Exim. Since child_open_exim() is
311 documented for local_scan(), the new function should be too.
313 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
314 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
315 results in an empty string is now treated as unset.
317 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
319 PH/31 Added sender_verify_fail logging option.
321 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
322 needed by RFC 822 but not by RFC 2822 was commented out. I have now
323 tidied the source and removed it altogether.
325 PH/33 When a queue run was abandoned because the load average was too high, a
326 log line was always written; now it is written only if the queue_run log
327 selector is set. In addition, the log line for abandonment now contains
328 information about the queue run such as the pid. This is always present
329 in "start" and "stop" lines but was omitted from the "abandon" line.
331 PH/34 Omit spaces between a header name and the colon in the error message that
332 is given when verify = headers_syntax fails (if there are lots of them,
333 the message gets confusing).
335 PH/35 Change the default for dns_check_names_pattern to allow slashes within
336 names, as there are now some PTR records that contain slashes. This check
337 is only to protect against broken name servers that fall over on strange
338 characters, so the fact that it applies to all lookups doesn't matter.
340 PH/36 Now that the new test suite is complete, we can remove some of the
341 special code in Exim that was needed for the old test suite. For example,
342 sorting DNS records because real resolvers return them in an arbitrary
343 order. The new test suite's fake resolver always returns records in the
346 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
347 bounces) except when queue_only is set, to avoid logging races between
348 the different processes.
350 PH/38 Panic-die if .include specifies a non-absolute path.
352 PH/39 A tweak to the "H" retry rule from its user.
354 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
355 a label. They prevented compilation on older perls.
357 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
358 a warning to be raised on newish perls.
360 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
361 on queue. Changes to match documented behaviour of showing count of
362 messages matching specified criteria.
364 PH/40 Changed the default ident timeout from 30s to 5s.
366 PH/41 Added support for the use of login_cap features, on those BSD systems
367 that have them, for controlling the resources used by pipe deliveries.
369 PH/42 The content-scanning code uses fopen() to create files in which to put
370 message data. Previously it was not paying any attention to the mode of
371 the files. Exim runs with umask(0) because the rest of the code creates
372 files with open(), and sets the required mode explicitly. Thus, these
373 files were ending up world-writeable. This was not a big issue, because,
374 being within the spool directory, they were not world-accessible. I have
375 created a function called modefopen, which takes an additional mode
376 argument. It sets umask(777), creates the file, chmods it to the required
377 mode, then resets the umask. All the relevant calls to fopen() in the
378 content scanning code have been changed to use this function.
380 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
381 to 24 hours. This avoids potential overflow problems when processing G
382 and H retry rules. I suspect nobody ever tinkers with this value.
384 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
386 PH/45 When the plaintext authenticator is running as a client, the server's
387 challenges are checked to ensure they are valid base64 strings. By
388 default, the authentication attempt is cancelled if an invalid string is
389 received. Setting client_ignore_invalid_base64 true ignores these errors.
390 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
391 they are received. Thus, the responses can be made to depend on the
392 challenges. If an invalid string is ignored, an empty string is placed in
395 PH/46 Messages that are created by the autoreply transport now contains a
396 References: header, in accordance with RFCs 2822 and 3834.
398 PH/47 Added authenticated_sender_force to the smtp transport.
400 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
402 PH/49 Installed latest patch from the Sieve maintainer.
404 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
405 was also set, the appendfile transport was unnecessarily scanning a
406 directory of message files (e.g. for maildir delivery) to find the count
407 of files (along with the size), even though it did not need this
408 information. It now does the scan only if it needs to find either the
409 size of the count of files.
411 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
413 PH/52 Two bugs concerned with error handling when the smtp transport is
416 (i) Exim was not creating retry information for temporary errors given
417 for individual recipients after the DATA command when the smtp transport
418 was used in LMTP mode. This meant that they could be retried too
419 frequently, and not timed out correctly.
421 (ii) Exim was setting the flag that allows error details to be returned
422 for LMTP errors on RCPT commands, but not for LMTP errors for individual
423 recipients that were returned after the DATA command.
425 PH/53 This is related to PH/52, but is more general: for any failing address,
426 when detailed error information was permitted to be returned to the
427 sender, but the error was temporary, then after the final timeout, only
428 "retry timeout exceeded" was returned. Now it returns the full error as
429 well as "retry timeout exceeded".
431 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
432 do this, and (what is worse) MTAs that accept it.
434 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
435 will now be deprecated.
437 PH/56 New os.c-cygwin from the Cygwin maintainer.
439 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
440 formats (previously only available in exim formats via -bpr, -bpru,
441 and -bpra. Now also available in native and exiqgrep formats)
443 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
444 with very large, slow to parse queues
446 JJ/08 exipick: added ! as generic prefix to negate any criteria format
448 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
450 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
451 responses to authentication challenges, though it was showing the
452 challenges; (ii) I've removed the CR characters from the debug output for
455 PH/58 Allow for the insertion of a newline as well as a space when a string
456 is turned into more than one encoded-word during RFC 2047 encoding. The
457 Sieve code now uses this.
459 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
460 data_4xx, lost_connection, tls_required.
462 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
463 message was being sent as an SMTP response.
465 PH/61 Add -l and -k options to exicyclog.
467 PH/62 When verifying, if an address was redirected to one new address, so that
468 verification continued, and the new address failed or deferred after
469 having set something in $address_data, the value of $address_data was not
470 passed back to the ACL. This was different to the case when no
471 redirection occurred. The value is now passed back in both cases.
473 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
474 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
475 use login_cap.h, so on its own it isn't the distinguishing feature. The
476 new name refers directly to the setclassresources() function.
478 PH/65 Added configuration files for NetBSD3.
480 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
482 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
483 is preferred over IPv4.
485 PH/68 The bounce_return_message and bounce_return_body options were not being
486 honoured for bounces generated during the reception of non-SMTP messages.
487 In particular, this applied to messages rejected by the ACL. This bug has
488 been fixed. However, if bounce_return_message is true and bounce_return_
489 body is false, the headers that are returned for a non-SMTP message
490 include only those that have been read before the error was detected.
491 (In the case of an ACL rejection, they have all been read.)
493 PH/69 The HTML version of the specification is now built in a directory called
494 spec_html instead of spec.html, because the latter looks like a path with
495 a MIME-type, and this confuses some software.
497 PH/70 Catch two compiler warnings in sieve.c.
499 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
500 function verify_get_ident() calls ip_connect() to connect a socket, but
501 if the "connect()" function timed out, ip_connect() used to close the
502 socket. However, verify_get_ident() also closes the socket later, and in
503 between Exim writes to the log, which may get opened at this point. When
504 the socket was closed in ip_connect(), the log could get the same file
505 descriptor number as the socket. This naturally causes chaos. The fix is
506 not to close the socket in ip_connect(); the socket should be closed by
507 the function that creates it. There was only one place in the code where
508 this was missing, in the iplookup router, which I don't think anybody now
509 uses, but I've fixed it anyway.
511 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
512 well as to direct DNS lookups. Otherwise the handling of names in host
513 lists is inconsistent and therefore confusing.
519 PH/01 Two changes to the default runtime configuration:
521 (1) Move the checks for relay_from_hosts and authenticated clients from
522 after to before the (commented out) DNS black list checks.
524 (2) Add control=submission to the relay_from_hosts and authenticated
525 clients checks, on the grounds that messages accepted by these
526 statements are most likely to be submissions.
528 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
530 (1) Generate an error if the third argument for the ${prvs expansion is
533 (2) Treat a missing third argument of ${prvscheck as if it were an empty
536 (3) Reset the variables that are obtained from the first argument of
537 ${prvscheck and used in the second argument before leaving the code,
538 because their memory is reclaimed, so using them afterwards may do
541 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
542 one (it's much easier than Tom thought :-).
544 (5) Because of (4), we can now allow for the use of $prvscheck_result
545 inside the third argument.
547 PH/03 For some reason, the default setting of PATH when running a command from
548 a pipe transport was just "/usr/bin". I have changed it to
551 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
552 anything to be listed in the output from -bV.
554 PH/05 When a filter generated an autoreply, the entire To: header line was
555 quoted in the delivery log line, like this:
557 => >A.N.Other <ano@some.domain> <original@ddress> ...
559 This has been changed so that it extracts the operative address. There
560 may be more than one such address. If so, they are comma-separated, like
563 => >ano@some.domain,ona@other.domain <original@ddress> ...
565 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
566 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
567 not being looked up in the rDNS to get a host name, Exim was showing the
568 IP address twice in Received: lines, even though the IP addresses were
569 identical. For example:
571 Received: from [1.2.3.4] (helo=[1.2.3.4])
573 However, if the real host name was known, it was omitting the HELO data
574 if it matched the actual IP address. This has been tidied up so that it
575 doesn't show the same IP address twice.
577 PH/07 When both +timestamp and +memory debugging was on, the value given by
578 $tod_xxx expansions could be wrong, because the tod_stamp() function was
579 called by the debug printing, thereby overwriting the timestamp buffer.
580 Debugging no longer uses the tod_stamp() function when +timestamp is set.
582 PH/08 When the original message was included in an autoreply transport, it
583 always said "this is a copy of the message, including all the headers",
584 even if body_only or headers_only was set. It now gives an appropriate
587 PH/09 Applied a patch from the Sieve maintainer which:
589 o fixes some comments
590 o adds the (disabled) notify extension core
591 o adds some debug output for the result of if/elsif tests
592 o points to the current vacation draft in the documentation
593 and documents the missing references header update
597 o fixes a bug in processing the envelope test (when testing
598 multiple envelope elements, the last element determinted the
601 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
602 Electronic Mail") by including:
604 Auto-submitted: auto-generated
606 in the messages that it generates (bounce messages and others, such as
607 warnings). In the case of bounce messages for non-SMTP mesages, there was
608 also a typo: it was using "Auto_submitted" (underscore instead of
609 hyphen). Since every message generated by Exim is necessarily in response
610 to another message, thes have all been changed to:
612 Auto-Submitted: auto-replied
614 in accordance with these statements in the RFC:
616 The auto-replied keyword:
618 - SHOULD be used on messages sent in direct response to another
619 message by an automatic process,
621 - MUST NOT be used on manually-generated messages,
623 - MAY be used on Delivery Status Notifications (DSNs) and Message
624 Disposition Notifications (MDNs),
626 - MUST NOT be used on messages generated by automatic or periodic
627 processes, except for messages which are automatic responses to
630 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
631 to the default Received: header definition.
633 PH/12 Added log selector acl_warn_skipped (default on).
635 PH/13 After a successful wildlsearch lookup, discard the values of numeric
636 variables because (a) they are in the wrong storage pool and (b) even if
637 they were copied, it wouldn't work properly because of the caching.
639 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
640 checking when decoding. Apparently there are clients that generate
641 overlong encoded strings. Why am I not surprised?
643 PH/15 If the first argument of "${if match_address" was not empty, but did not
644 contain an "@" character, Exim crashed. Now it writes a panic log message
645 and treats the condition as false.
647 PH/16 In autoreply, treat an empty string for "once" the same as unset.
649 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
650 extension "envelope-auth". The code is finished and in agreement with
651 other implementations, but there is no documentation so far and in fact,
652 nobody wrote the draft yet. This extension is currently #undef'ed, thus
653 not changing the active code.
655 Print executed "if" and "elsif" statements when debugging is used. This
656 helps a great deal to understand what a filter does.
658 Document more things not specified clearly in RFC3028. I had all this
659 sorted out, when out of a sudden new issues came to my mind. Oops."
661 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
664 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
665 canonical form (as documented). However, after a host name lookup from
666 the IP address, check_host() was doing a simple string comparison with
667 addresses acquired from the DNS when checking that the found name did
668 have the original IP as one of its addresses. Since any found IPv6
669 addresses are likely to be in abbreviated form, the comparison could
670 fail. Luckily, there already exists a function for doing the comparison
671 by converting both addresses to binary, so now that is used instead of
674 PH/20 There was another similar case to PH/19, when a complete host name was
675 given in a host list; looking up its IP address could give an abbreviated
676 form, whereas the current host's name might or might not be abbreviated.
677 The same fix has been applied.
683 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
684 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
687 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
688 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
690 PH/03 Typo: missing ".o" in src/pcre/Makefile.
692 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
693 header line, restrict the check to what is listed in RFCs 2369 and 2929.
694 Also, for "Auto-Submitted", treat anything other than "no" as
695 non-personal, in accordance with RFC 3834. (Previously it treated
696 anything starting "auto-" as non-personal.)
698 TF/01 The control=submission/name=... option had a problem with syntax
699 errors if the name included a slash character. The /name= option
700 now slurps the rest of the string, so it can include any characters
701 but it must come last in the list of options (after /sender_retain
704 PH/05 Some modifications to the interface to the fake nameserver for the new
712 TK/01 Added the "success_on_redirect" address verification option. See
713 NewStuff for rationale and an example.
715 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
717 PH/02 Patch to exigrep to allow it to work on syslog lines.
719 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
720 fread() to skip over the body file's header line, because in Cygwin the
721 header line is locked and is inaccessible.
723 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
724 co-exist for some time) to make it clear that it is the Exim ID that is
725 referenced, not the Message-ID: header line.
727 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
728 string_format() function, because snprintf() does not exist on all
731 PH/06 The use of forbid_filter_existstest now also locks out the use of the
732 ${stat: expansion item.
734 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
735 protocol synchronization error", to keep the pedants happy.
737 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
738 well as for IRIX systems, when gcc is being used. See the host.c source
741 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
743 PH/10 Named domain lists were not working if used in a queue_smtp_domains
746 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
747 transport and to the smtp transport in LMTP mode.
749 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
751 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
752 run a filter in a subprocess. This could lead to confusion in subsequent
753 lookups in the parent process. There should also be a search_tidyup() at
754 the end of the subprocess.
756 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
757 only if the host matched helo_try_verify_hosts, which caused the
758 verification to occur when the EHLO/HELO command was issued. The ACL just
759 tested the remembered result. Now, if a previous verification attempt has
760 not happened, "verify = helo" does it there and then.
762 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
764 TK/03 Fix log output including CR from clamd.
766 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
767 exist provoked a memory error which could cause a segfault.
769 PH/15 Installed PCRE 6.2
771 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
773 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
774 of the problem. Specifically, suggested +O2 rather than +O1 for the
777 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
779 PH/20 If a delivery was routed to a non-standard port by means of an SRV
780 record, the port was not correctly logged when the outgoing_port log
781 selector was set (it logged the transort's default port).
783 PH/21 Added support for host-specific ports to manualroute, queryprogram,
784 fallback_hosts, and "hosts" in the smtp transport.
786 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
787 host errors such as "Connection refused".
789 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
790 authentication with radiusclient 0.4.9:
792 - Error returned from rc_read_config was caught wrongly
793 - Username/password not passed on to radius server due to wrong length.
795 The presumption is that some radiusclient API changes for 4.51/PH/17
796 were not taken care of correctly. The code is still untested by me (my
797 Linux distribution still has 0.3.2 of radiusclient), but it was
798 contributed by a Radius user.
800 PH/24 When doing a callout, the value of $domain wasn't set correctly when
801 expanding the "port" option of the smtp transport.
803 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
804 while reading a MIME header. Thanks to Tom Hughes for a patch.
806 PH/24 Include config.h inside local_scan.h so that configuration settings are
809 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
810 that in an ACL for RCPT (for example), you can examine exactly what was
813 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
814 commands, but it was not correctly comparing the address with the actual
815 client host address. Thus, it would show the EHLO address in Received:
816 header lines when this was not necessary.
818 PH/27 Added the % operator to ${eval:}.
820 PH/28 Exim tries to create and chdir to its spool directory when it starts;
821 it should be ignoring failures (because with -C, for example, it has lost
822 privilege). It wasn't ignoring creation failures other than "already
825 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
828 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
829 because an input line was too long, either on its own, or by virtue of
830 too many continuations, the temporary file was not being removed, and the
831 return code was incorrect.
833 PH/31 Missing "BOOL" in function definition in filtertest.c.
835 PH/32 Applied Sieve patches from the maintainer.
837 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
839 PH/33 Added "verify = not_blind".
841 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
842 Local/Makefile (with some defaults set). These are used in built scripts
843 such as exicyclog, but they have never been used in the exim_install
844 script (though there are many overriding facilities there). I have
845 arranged that the exim_install script now takes note of these two
848 PH/35 Installed configuration files for Dragonfly.
850 PH/36 When a locally submitted message by a trusted user did not contain a
851 From: header, and the sender address was obtained from -f or from an SMTP
852 MAIL command, and the trusted user did not use -F to supply a sender
853 name, $originator_name was incorrectly used when constructing a From:
854 header. Furthermore, $originator_name was used for submission mode
855 messages from external hosts without From: headers in a similar way,
856 which is clearly wrong.
858 PH/37 Added control=suppress_local_fixups.
860 PH/38 When log_selector = +received_sender was set, and the addition of the
861 sender made the log line's construction buffer exactly full, or one byte
862 less than full, an overflow happened when the terminating "\n" was
865 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
866 when the result of a list match is failure because a DNS lookup failed.
868 PH/40 RM_COMMAND is now used in the building process.
870 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
871 the "build-* directories that it finds.
873 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
874 address in a domain literal was a prefix of an interface address.
876 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
877 when verifying a sender address, unless rewrite_headers is false.
879 PH/44 Wrote a long comment about why errors_to addresses are verified as
880 recipients, not senders.
882 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
883 the ratelimit ACL was added.
885 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
887 PH/46 Added extra information about PostgreSQL errors to the error string.
889 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
890 avoiding the need to install special zones in a real server. This is
891 backwards compatible; if it can't find the fake resolver, it drops back.
892 Thus, both old and new test suites can be run.
894 TF/02 Added util/ratelimit.pl
896 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
897 clock is set back in time.
899 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
900 Candler <B.Candler@pobox.com>.
902 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
903 OK for addresses that are the result of redirections.
905 PH/48 A number of further additions for the benefit of the new test suite,
906 including a fake gethostbyname() that interfaces to the fake DNS resolver
909 TF/06 The fix for widen_domains has also been applied to qualify_single and
910 search_parents which are the other dnslookup options that can cause
913 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
916 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
918 TF/07 Exim produced the error message "an SRV record indicated no SMTP
919 service" if it encountered an MX record with an empty target hostname.
920 The message is now "an MX or SRV record indicated no SMTP service".
922 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
923 if the DNS of the sending site is misconfigured. This is quite a
924 common situation. This change restores the behaviour of treating a
925 helo verification defer as a failure.
927 PH/51 If self=fail was set on a router, the bounce message did not include the
928 actual error message.
934 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
936 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
937 command itself ended in error, the underlying message about the transport
938 filter timeout was being overwritten with the pipe command error. Now the
939 underlying error message should be appended to the second error message.
941 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
943 PH/02 Reduce the amount of output that "make" produces by default. Full output
944 can still be requested.
946 PH/03 The warning log line about a condition test deferring for a "warn" verb
947 was being output only once per connection, rather than after each
948 occurrence (because it was using the same function as for successful
949 "warn" verbs). This seems wrong, so I have changed it.
951 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
952 it should not have, which might have caused a crash in the right
953 circumstances, but probably never did.
955 PH/04 Installed a modified version of Tony Finch's patch to make submission
956 mode fix the return path as well as the Sender: header line, and to
957 add a /name= option so that you can make the user's friendly name appear
960 TF/03 Added the control = fakedefer ACL modifier.
962 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
963 Mark Lowes for thorough testing.
965 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
967 TK/03 Merged latest SRS patch from Miles Wilton.
969 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
970 with the definition in sysexits.h (which is #included earlier).
971 Fortunately, Exim does not actually use EX_OK. The code used to try to
972 preserve the sysexits.h value, by assumimg that macro definitions were
973 scanned for macro replacements. I have been disabused of this notion,
974 so now the code just undefines EX_OK before #including unistd.h.
976 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
977 in the smtp transport. When a block could not be written in a single
978 write() function, the timeout was being re-applied to each part-write.
979 This seems wrong - if the receiver was accepting one byte at a time it
980 would take for ever. The timeout is now adjusted when this happens. It
981 doesn't have to be particularly precise.
983 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
984 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
986 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
987 without a domain if the check to <postmaster@domain> fails.
989 SC/01 Eximstats: added -xls and the ability to specify output files
990 (patch written by Frank Heydlauf).
992 SC/02 Eximstats: use FileHandles for outputing results.
994 SC/03 Eximstats: allow any combination of xls, txt, and html output.
996 SC/04 Eximstats: fixed display of large numbers with -nvr option
998 SC/05 Eximstats: fixed merging of reports with empty tables.
1000 SC/06 Eximstats: added the -include_original_destination flag
1002 SC/07 Eximstats: removed tabs and trailing whitespace.
1004 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1006 TK/06 MBOX spool code: Add real "From " MBOX separator line
1007 so the .eml file is really in mbox format (even though
1008 most programs do not really care). Patch from Alex Miller.
1010 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1011 The latter is generated from $received_to and is only set if the
1012 message has one envelope recipient. SA can use these headers,
1013 obviously out-of-the-box. Patch from Alex Miller.
1015 PH/08 The ${def test on a variable was returning false if the variable's
1016 value was "0", contrary to what the specification has always said!
1017 The result should be true unless the variable is empty.
1019 PH/09 The syntax error of a character other than { following "${if
1020 def:variable_name" (after optional whitespace) was not being diagnosed.
1021 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1022 accidental colon was present, for example, could give incorrect results.
1024 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1025 result is used (not including appendfile, where other changes are about
1028 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1029 This involved changing a lot of size variables from int to off_t. It
1030 should work with maildirs and everything.
1032 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1033 spamd dying while we are connected to it.
1035 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1038 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1039 function for BDB 4.3.
1041 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1043 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1044 the sources for PCRE have been re-arranged and more files are now
1047 PH/15 The code I had for printing potentially long long variables in PH/11
1048 above was not the best (it lost precision). The length of off_t variables
1049 is now inspected at build time, and an appropriate printing format (%ld
1050 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1051 to be "long long int" or "long int". This is needed for the internal
1052 formatting function string_vformat().
1054 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1055 the configuration file to be ":syslog", then the script "guesses" where
1056 the logs files are, rather than using the compiled in default. In our
1057 case the guess is not the same as the compiled default, so the script
1058 suddenly stopped working when I started to use syslog. The patch checks
1059 to see if log_file_path is "". If so, it attempts to read it from exim
1060 with no configuration file to get the compiled in version, before it
1061 falls back to the previous guessing code."
1063 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1064 implementing BATV in an Exim configuration. See NewStuff for the gory
1067 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1068 Makefile that are specific to HP-UX.
1070 PH/18 If the "use_postmaster" option was set for a recipient callout together
1071 with the "random" option, the postmaster address was used as the MAIL
1072 FROM address for the random test, but not for the subsequent recipient
1073 test. It is now used for both.
1075 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1076 patch removes a few documentation additions to RFC 3028, because the
1077 latest draft now contains them. It adds the new en;ascii-case comparator
1078 and a new error check for 8bit text in MIME parts. Comparator and
1079 require names are now matched exactly. I enabled the subaddress
1080 extension, but it is not well tested yet (read: it works for me)."
1082 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1083 rework some of the code of TK/09 above to avoid the hardwired use of
1084 "%lld" and "long long". Replaced the call to snprintf() with a call to
1087 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1088 records point to non-existent hosts", "retry timeout exceeded", and
1089 "retry time not reached for any host after a long failure period".
1091 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1092 experimental DomainKeys support:
1094 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1095 (2) On an error such as an illegally used "control", the wrong name for
1096 the control was given.
1098 These problems did NOT occur unless DomainKeys support was compiled.
1100 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1102 PH/24 Added ${if match_ip condition.
1104 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1105 databases so that it will be absolutely obvious if a crash occurs in the
1106 DB library. This is a regular occurrence (often caused by mis-matched
1109 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1110 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1111 picked up on a user's system that detects such things. There doesn't seem
1112 to be a gcc warning option for this - only an attribute that has to be
1113 put on the function's prototype. It seems that in Fedora Core 4 they have
1114 set this on a number of new functions. No doubt there will be more in due
1117 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1118 specify a transport. However, if an address that was verified by such a
1119 router was the subject of a callout, Exim crashed because it tried to
1120 read the rcpt_include_affixes from the non-existent transport. Now it
1121 just assumes that the setting of that option is false. This bug was
1122 introduced by 4.51/PH/31.
1124 PH/28 Changed -d+all to exclude +memory, because that information is very
1125 rarely of interest, but it makes the output a lot bigger. People tend to
1126 do -d+all out of habit.
1128 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1129 code in os-type was giving problems when libc.so lives in lib64, like on
1132 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1133 aren't the modern standard, and it seems that some systems' include files
1134 don't always have them. Exim was already checking for some of the newer
1135 ones like T_AAAA, and defining it itself. I've added checks for all the
1136 record types that Exim uses.
1138 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1139 not automatically generating a new one, as it is supposed to. This
1140 prevented TLS from working. If the file did exist, but contained invalid
1141 data, a new version was generated, as expected. It was only the case of a
1142 non-existent file that was broken.
1144 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1145 with a change in libdomainkeys > 0.64.
1147 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1148 from DNS. If the selector record carries the flag, it now has
1149 precedence over the domain-wide flag.
1151 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1153 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1154 the use of a port name that isn't defined in /etc/services) occurred, the
1155 message was deferred as in a normal delivery, and thus remained on the
1156 spool, instead of being failed because of the mua_wrapper setting. This
1157 is now fixed, and I tidied up some of the mua_wrapper messages at the
1160 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1161 the messages in a hash of arrays rather than using individual hashes.
1162 This is a bit cleaner and results in dramatic memory savings, albeit
1163 at a slight CPU cost.
1165 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1166 as requested by Marc Sherman.
1168 SC/10 Eximstats: added histograms for user specified patterns as requested
1171 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1173 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1174 fopen() in the content-scanning modules that did not already have it.
1180 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1181 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1183 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1185 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1187 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1188 to test Sieve filters that use "vacation".
1190 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1191 that changes the way the GnuTLS parameters are stored in the cache file.
1192 The new format can be generated externally. For backward compatibility,
1193 if the data in the cache doesn't make sense, Exim assumes it has read an
1194 old-format file, and it generates new data and writes a new file. This
1195 means that you can't go back to an older release without removing the
1198 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1199 work if there are any delivery delays because "one_time" forces the
1200 parent to be marked "delivered", so its unseen clone is never tried
1201 again. For this reason, Exim now forbids the simultaneous setting of
1204 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1205 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1206 entry if you want to know the details. The fix, however, neglected to
1207 consider the case when local delivery batching is involved. The test for
1208 "previously delivered" was not happening when checking to see if an
1209 address could be batched with a previous (undelivered) one; under
1210 certain circumstances this could lead to multiple deliveries to the same
1213 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1214 in its include files, and this causes problems building Exim.
1216 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1217 header_syntax) but Exim was just ignoring anything given after a slash.
1218 In particular, this caused confusion with an attempt to use "verify =
1219 reverse_host_lookup/defer_ok". An error is now given when options are
1220 supplied for verify items that do not have them. (Maybe reverse_host_
1221 lookup should have a defer_ok option, but that's a different point.)
1223 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1224 defined by RFC 821) to 2048, because there were problems with some AUTH
1225 commands, and RFC 1869 says the size should be increased for extended
1226 SMTP commands that take arguments.
1228 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1231 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1232 "unknown" error; now it says that the functionality isn't in the binary.
1234 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1235 an address' error message when a string expansion fails (syntax or
1236 whatever). Otherwise the password may appear in the log. Following change
1237 PH/42 below, there is no longer a chance of it appearing in a bounce
1240 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1242 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1243 of its addresses was ever tried. (Bugzilla bug #2.)
1245 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1246 the result incorrectly in the debug output. (It correctly added a newline
1247 to what was transported.)
1249 TF/01 Added $received_time.
1251 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1252 commented out examples of how to interface to a virus scanner and to
1253 SpamAssassin. Also added commented examples of av_scanner and
1254 spamd_address settings.
1256 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1257 and controls are allowed in which ACLs. There were a couple of minor
1258 errors. Some of the entries in the conditions table (which is a table of
1259 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1260 negation of where the condition IS allowed.
1262 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1264 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1265 header file does not have a version number, so I've had to invent a new
1266 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1267 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1268 radiusclient), but it was contributed by a Radius user.
1270 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1271 files or directories, for OpenSSL.
1273 PH/19 When an Exim process that is running as root has to create an Exim log
1274 file, it does so in a subprocess that runs as exim:exim so as to get the
1275 ownership right at creation (otherwise, other Exim processes might see
1276 the file with the wrong ownership). There was no test for failure of this
1277 fork() call, which would lead to the process getting stuck as it waited
1278 for a non-existent subprocess. Forks do occasionally fail when resources
1279 run out. I reviewed all the other calls to fork(); they all seem to check
1282 PH/20 When checking for unexpected SMTP input at connect time (before writing
1283 the banner), Exim was not dealing correctly with a non-positive return
1284 from the read() function. If the client had disconnected by this time,
1285 the result was a log entry for a synchronization error with an empty
1286 string after "input=" when read() returned zero. If read() returned -1
1287 (an event I could not check), uninitialized data bytes were printed.
1288 There were reports of junk text (parts of files, etc) appearing after
1291 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1293 PH/22 Added support for macro redefinition, and (re)definition in between
1294 driver and ACL definitions.
1296 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1297 forgetting to use the resulting value; it was using the unexpanded value.
1299 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1300 hadn't been configured. The fix is from Juergen Kreileder, who
1301 understands it better than I do:
1303 "Here's what I see happening with three configured cyrus_sasl
1304 authenticators configured (plain, login, cram-md5):
1306 On startup auth_cyrus_sasl_init() gets called for each of these.
1307 This means three calls to sasl_listmech() without a specified mech_list.
1308 => SASL tests which mechs of all available mechs actually work
1309 => three warnings about OTP not working
1310 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1312 With the patch, sasl_listmech() also gets called three times. But now
1313 SASL's mech_list option is set to the server_mech specified in the the
1314 authenticator. Or in other words, the answer from sasl_listmech()
1315 gets limited to just the mech you're testing for (which is different
1317 => the return list contains just 'plain' or 'login', 'cram-md5' or
1318 nothing depending on the value of ob->server_mech.
1320 I've just tested the patch: Authentication still works fine,
1321 unavailable mechs specified in the exim configuration are still
1322 caught, and the auth.log warnings about OTP are gone."
1324 PH/25 When debugging is enabled, the contents of the command line are added
1325 to the debugging output, even when log_selector=+arguments is not
1328 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1329 answer is "GNU", and only if the return is "GNU/something" is the answer
1332 PH/27 $acl_verify_message is now set immediately after the failure of a
1333 verification in an ACL, and so is available in subsequent modifiers. In
1334 particular, the message can be preserved by coding like this:
1336 warn !verify = sender
1337 set acl_m0 = $acl_verify_message
1339 Previously, $acl_verify_message was set only while expanding "message"
1340 and "log_message" when a very denied access.
1342 PH/28 Modified OS/os.c-Linux with
1344 -#ifndef OS_LOAD_AVERAGE
1345 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1347 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1348 nomenclature these days.)
1350 PH/29 Installed patch from the Sieve maintainer that adds the options
1351 sieve_useraddress and sieve_subaddress to the redirect router.
1353 PH/30 In these circumstances:
1354 . Two addresses routed to the same list of hosts;
1355 . First host does not offer TLS;
1356 . First host accepts first address;
1357 . First host gives temporary error to second address;
1358 . Second host offers TLS and a TLS session is established;
1359 . Second host accepts second address.
1360 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1361 and peerdn, if requested) that were in fact used only for the second
1364 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1365 attention to any local part prefix or suffix that was matched by the
1366 router that accepted the address. It now behaves in the same way as it
1367 does for delivery: the affixes are removed from the local part unless
1368 rcpt_include_affixes is set on the transport.
1370 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1371 timeout during the DATA phase of an incoming message.
1373 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1374 applied a patch sanctioned by the Sieve maintainer.
1376 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1377 the uid or gid is negative. A case of a negative gid caused this to be
1378 noticed. The fix allows for either to be negative.
1380 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1381 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1384 PH/36 The change PH/12 above was broken. Fixed it.
1386 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1387 the grounds that routing the same address twice would always produce the
1388 same answer. This might have been true once, but it is certainly no
1389 longer true now. Routing a child address may depend on the previous
1390 routing that produced that child. Some complicated redirection strategies
1391 went wrong when messages had multiple recipients, and made Exim's
1392 behaviour dependent on the order in which the addresses were given.
1394 I have moved the duplicate checking until after the routing is complete.
1395 Exim scans the addresses that are assigned to local and remote
1396 transports, and removes any duplicates. This means that more work will be
1397 done, as duplicates will always all be routed, but duplicates are
1398 presumably rare, so I don't expect this is of any significance.
1400 For deliveries to pipes, files, and autoreplies, the duplicate checking
1401 still happens during the routing process, since they are not going to be
1404 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1405 It corrects a timeout issue with spamd. This is Ian's comment: "The
1406 background is that sometimes spamd either never reads data from a
1407 connection it has accepted, or it never writes response data. The exiscan
1408 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1409 blindly assumes that writes won't block so it may never time out."
1411 PH/39 Allow G after quota size as well as K and M.
1413 PH/40 The value set for $authenticated_id in an authenticator may not contain
1414 binary zeroes or newlines because the value is written to log lines and
1415 to spool files. There was no check on this. Now the value is run through
1416 the string_printing() function so that such characters are converted to
1417 printable escape sequences.
1419 PH/41 $message_linecount is a new variable that contains the total number of
1420 lines in the message. Compare $body_linecount, which is the count for the
1423 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1424 bounce and delay warning messages, except in certain special cases, which
1427 (a) An SMTP error message from a remote host;
1428 (b) A message specified in a :fail: redirection;
1429 (c) A message specified in a "fail" command in a system filter;
1430 (d) A message specified in a FAIL return from the queryprogram router;
1431 (e) A message specified by the cannot_route_message router option.
1433 In these cases only, Exim does include the error details in bounce and
1434 warning messages. There are also a few cases where bland messages such
1435 as "unrouteable address" or "local delivery error" are given.
1437 PH/43 $value is now also set for the "else" part of a ${run expansion.
1439 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1440 being worked on, but at least Exim now implements the latest version to
1443 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1444 process to complete was treated as a delivery failure, a timeout while
1445 writing the message to the pipe was logged, but erroneously treated as a
1446 successful delivery. Such timeouts include transport filter timeouts. For
1447 consistency with the overall process timeout, these timeouts are now
1448 treated as errors, giving rise to delivery failures by default. However,
1449 there is now a new Boolean option for the pipe transport called
1450 timeout_defer, which, if set TRUE, converts the failures into defers for
1451 both kinds of timeout. A transport filter timeout is now identified in
1454 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1455 systems where "make" and "gmake" are different, calling "gmake" at top
1456 level broke things. I've arranged for the value of $(MAKE) to be passed
1457 from the Makefile to this script so that it can call the same version of
1461 A note about Exim versions 4.44 and 4.50
1462 ----------------------------------------
1464 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1465 changes of various kinds. As a consequence, a big documentation update was
1466 needed. This delayed the release for rather longer than seemed good, especially
1467 in the light of a couple of (minor) security issues. Therefore, the changes
1468 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1469 release. So 4.44 and 4.50 are in effect two different branches that both start
1472 I have left the 4.50 change log unchanged; it contains all the changes since
1473 4.43. The change log for 4.44 is below; many of its items are identical to
1474 those for 4.50. This seems to be the most sensible way to preserve the
1475 historical information.
1481 1. Minor wording change to the doc/README.SIEVE file.
1483 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1484 computation of the current number of files was incorrect.
1486 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1487 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1490 4. Give more explanation in the error message when the command for a transport
1491 filter fails to execute.
1493 5. There are several places where Exim runs a non-Exim command in a
1494 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1495 was being done only for the command run by the queryprogram router. It is
1496 now done for all such subprocesses. The other cases are: ${run, transport
1497 filters, and the commands run by the lmtp and pipe transports.
1499 6. Added CONFIGURE_GROUP build-time option.
1501 7. Some older OS have a limit of 256 on the maximum number of file
1502 descriptors. Exim was using setrlimit() to set 1000 as a large value
1503 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1504 systems. I've change it so that if it can't get 1000, it tries for 256.
1506 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1507 was an oversight, and furthermore, ever since the addition of extra
1508 controls (e.g. 4.43/32), the checks on when to allow different forms of
1509 "control" were broken. There should now be diagnostics for all cases when a
1510 control that does not make sense is encountered.
1512 9. Added the /retain_sender option to "control=submission".
1514 10. $recipients is now available in the predata ACL (oversight).
1516 11. Tidy the search cache before the fork to do a delivery from a message
1517 received from the command line. Otherwise the child will trigger a lookup
1518 failure and thereby defer the delivery if it tries to use (for example) a
1519 cached ldap connection that the parent has called unbind on.
1521 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1522 of $address_data from the recipient verification was clobbered by the
1523 sender verification.
1525 13. The value of address_data from a sender verification is now available in
1526 $sender_address_data in subsequent conditions in the ACL statement.
1528 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1530 15. Added a new option "connect=<time>" to callout options, to set a different
1533 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1534 was its contents. (It was OK if the option was not defined at all.)
1536 17. A "Completed" log line is now written for messages that are removed from
1537 the spool by the -Mrm option.
1539 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1540 information about exactly what failed.
1542 19. Added -dd to debug only the daemon process.
1544 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1545 handles timeouts, both on the server side and network timeouts. Renamed the
1546 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1548 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1549 It is now set to "smtps".
1551 22. $host_address is now set to the target address during the checking of
1552 ignore_target_hosts.
1554 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1555 being passed; this would have caused $sender_host_name to have been used if
1556 matching the list had actually called for a host name (not very likely,
1557 since this list is usually IP addresses). A host name is now passed as
1560 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1561 code that specifies a non-restarting handler (typically sigaction() in
1562 modern systems) in an attempt to fix a rare and obscure crash bug.
1564 25. Narrowed the window for a race in the daemon that could cause it to ignore
1565 SIGCHLD signals. This is not a major problem, because they are used only to
1566 wake it up if nothing else does.
1568 26. A malformed maildirsize file could cause Exim to calculate negative values
1569 for the mailbox size or file count. Odd effects could occur as a result.
1570 The maildirsize information is now recalculated if the size or filecount
1573 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1574 support for a long time. Removed HAVE_SYS_VFS_H.
1576 28. Installed the latest version of exipick from John Jetmore.
1578 29. In an address list, if the pattern was not a regular expression, an empty
1579 subject address (from a bounce message) matched only if the pattern was an
1580 empty string. Non-empty patterns were not even tested. This was the wrong
1581 because it is perfectly reasonable to use an empty address as part of a
1582 database query. An empty address is now tested by patterns that are
1583 lookups. However, all the other forms of pattern expect the subject to
1584 contain a local part and a domain, and therefore, for them, an empty
1585 address still always fails if the pattern is not itself empty.
1587 30. Exim went into a mad DNS loop when attempting to do a callout where the
1588 host was specified on an smtp transport, and looking it up yielded more
1589 than one IP address.
1591 31. Re-factored the code for checking spool and log partition space into a
1592 function that finds that data and another that does the check. The former
1593 is then used to implement four new variables: $spool_space, $log_space,
1594 $spool_inodes, and $log_inodes.
1596 32. The RFC2047 encoding function was originally intended for short strings
1597 such as real names; it was not keeping to the 75-character limit for
1598 encoded words that the RFC imposes. It now respects the limit, and
1599 generates multiple encoded words if necessary. To be on the safe side, I
1600 have increased the buffer size for the ${rfc2047: expansion operator from
1603 33. It is now permitted to omit both strings after an "if" condition; if the
1604 condition is true, the result is "true". As before, when the second string
1605 is omitted, a false condition yields an empty string. This makes it less
1606 cumbersome to write custom ACL and router conditions.
1608 34. Failure to deliver a bounce message always caused it to be frozen, even if
1609 there was an errors_to setting on the router. The errors_to setting is now
1612 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1613 canonical form (fully expanded) before being placed in
1614 $sender_host_address.
1616 36. The table in the code that translates DNS record types into text (T_A to
1617 "A" for instance) was missing entries for NS and CNAME. It is just possible
1618 that this could have caused confusion if both these types were looked up
1619 for the same domain, because the text type is used as part of Exim's
1620 per-process caching. But the chance of anyone hitting this buglet seems
1623 37. The dnsdb lookup has been extended in a number of ways.
1625 (1) There is a new type, "zns", which walks up the domain tree until it
1626 finds some nameserver records. It should be used with care.
1628 (2) There is a new type, "mxh", which is like "mx" except that it returns
1629 just the host names, not the priorities.
1631 (3) It is now possible to give a list of domains (or IP addresses) to be
1632 looked up. The behaviour when one of the lookups defers can be
1633 controlled by a keyword.
1635 (4) It is now possible to specify the separator character for use when
1636 multiple records are returned.
1638 38. The dnslists ACL condition has been extended: it is now possible to supply
1639 a list of IP addresses and/or domains to be looked up in a particular DNS
1642 39. Added log_selector=+queue_time_overall.
1644 40. When running the queue in the test harness, wait just a tad after forking a
1645 delivery process, to get repeatability of debugging output.
1647 41. Include certificate and key file names in error message when GnuTLS fails
1648 to set them up, because the GnuTLS error message doesn't include the name
1649 of the failing file when there is a problem reading it.
1651 42. Allow both -bf and -bF in the same test run.
1653 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1655 44. The "Exiscan patch" is now merged into the mainline Exim source.
1657 45. Sometimes the final signoff response after QUIT could fail to get
1658 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1659 < 0 before doing a fflush(). This bug looks as though it goes back to the
1660 introduction of TLS in release 3.20, but "sometimes" must have been rare
1661 because the tests only now provoked it.
1663 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1664 (this can affect the format of dates).
1666 47. exim_tidydb, when checking for the continued existence of a message for
1667 which it has found a message-specific retry record, was not finding
1668 messages that were in split spool directories. Consequently, it was
1669 deleting retry records that should have stayed in existence.
1671 48. Steve fixed some bugs in eximstats.
1673 49. The SPA authentication driver was not abandoning authentication and moving
1674 on to the next authenticator when an expansion was forced to fail,
1675 contradicting the general specification for all authenticators. Instead it
1676 was generating a temporary error. It now behaves as specified.
1678 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1679 (the order specifies the preference for clients). The order is now AES256,
1680 AES128, 3DES, ARCFOUR128.
1682 51. Small patch to Sieve code - explicitly set From: when generating an
1685 52. Exim crashed if a remote delivery caused a very long error message to be
1686 recorded - for instance if somebody sent an entire SpamAssassin report back
1687 as a large number of 550 error lines. This bug was coincidentally fixed by
1688 increasing the size of one of Exim's internal buffers (big_buffer) that
1689 happened as part of the Exiscan merge. However, to be on the safe side, I
1690 have made the code more robust (and fixed the comments that describe what
1693 53. Now that there can be additional text after "Completed" in log lines (if
1694 the queue_time_overall log selector is set), a one-byte patch to exigrep
1695 was needed to allow it to recognize "Completed" as not the last thing in
1698 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1699 patch that reportedly fixes this has been added. I am not expert enough to
1700 create a test for it. This is what the patch creator wrote:
1702 "I found a little strange behaviour of ldap code when working with
1703 Windows 2003 AD Domain, where users was placed in more than one
1704 Organization Units. When I tried to give exim partial DN, the exit code
1705 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1706 But simultaneously result of request was absolutely normal ldap result,
1707 so I produce this patch..."
1709 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1710 REFERENCE, so I have modified the code to exclude the patch when that macro
1713 55. Some experimental protocols are using DNS PTR records for new purposes. The
1714 keys for these records are domain names, not reversed IP addresses. The
1715 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1716 leaves it alone. Component reversal etc. now happens only for IP addresses.
1719 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1721 57. Double the size of the debug message buffer (to 2048) so that more of very
1722 long debug lines gets shown.
1724 58. The exicyclog utility now does better if the number of log files to keep
1725 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1727 59. Two changes related to the smtp_active_hostname option:
1729 (1) $smtp_active_hostname is now available as a variable.
1730 (2) The default for smtp_banner uses $smtp_active_hostname instead
1731 of $primary_hostname.
1733 60. The host_aton() function is supposed to be passed a string that is known
1734 to be a valid IP address. However, in the case of IPv6 addresses, it was
1735 not checking this. This is a hostage to fortune. Exim now panics and dies
1736 if the condition is not met. A case was found where this could be provoked
1737 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1738 components; fortuitously, this particular loophole had already been fixed
1739 by change 4.50/55 above.
1741 If there are any other similar loopholes, the new check in host_aton()
1742 itself should stop them being exploited. The report I received stated that
1743 data on the command line could provoke the exploit when Exim was running as
1744 exim, but did not say which command line option was involved. All I could
1745 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1746 running as the user.
1749 61. There was a buffer overflow vulnerability in the SPA authentication code
1750 (which came originally from the Samba project). I have added a test to the
1751 spa_base64_to_bits() function which I hope fixes it.
1754 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1755 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1757 63. The daemon start-up calls getloadavg() while still root for those OS that
1758 need the first call to be done as root, but it missed one case: when
1759 deliver_queue_load_max is set with deliver_drop_privilege. This is
1760 necessary for the benefit of the queue runner, because there is no re-exec
1761 when deliver_drop_privilege is set.
1763 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1764 This has been fixed.
1766 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1767 was in use, was not putting the data itself into the right store pool;
1768 consequently, it could be overwritten for a subsequent message in the same
1769 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1772 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1774 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1775 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1776 the function were treating the return as a boolean value, which happened to
1777 work because 0=false and not-0=true, but is not correct code.
1779 68. The host_aton() function was not handling scoped IPv6 addresses (those
1780 with, for example, "%eth0" on the end) correctly.
1782 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1783 negated items (that is, ~something) in unsigned ints. Some compilers
1784 apparently mutter when there is no cast.
1786 70. If an address verification called from an ACL failed, and did not produce a
1787 user-specific message (i.e. there was only a "system" message), nothing was
1788 put in $acl_verify_message. In this situation, it now puts the system
1791 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1792 session; change 4.31/43 added the unwanted input to the log line - except
1793 that it did not do this in the start of session case. It now does.
1795 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1796 This is wrong and can cause the other end to generate a synchronization
1797 error if it is another Exim or anything else that does the synchronization
1798 check. A QUIT command is no longer sent after a timeout.
1800 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1801 during host lookups.
1803 74. The defer_ok option of callout verification was not working if it was used
1804 when verifying addresses in header lines, that is, for this case:
1806 verify = header_sender/callout=defer_ok
1808 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1809 those file descriptors could be used for SMTP connections. If anything
1810 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1811 could be sent to the SMTP client, causing chaos. The daemon now opens
1812 stdin, stdout, and stderr to /dev/null when it puts itself into the
1815 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1816 log by default. The user can override this with suitable Perl magic.
1818 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1819 the log message when discard triggers, was not working for the DATA ACL or
1820 for the non-SMTP ACL.
1822 78. Error message wording change in sieve.c.
1824 79. If smtp_accept_max_per_host was set, the number of connections could be
1825 restricted to fewer than expected, because the daemon was trying to set up
1826 a new connection before checking whether the processes handling previous
1827 connections had finished. The check for completed processes is now done
1828 earlier. On busy systems, this bug wouldn't be noticed because something
1829 else would have woken the daemon, and it would have reaped the completed
1832 80. If a message was submitted locally by a user whose login name contained one
1833 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1834 It caused a spool format error. I have fixed the spool reading code. A
1835 related problem was that the "from" clause in the Received: line became
1836 illegal because of the space(s). It is now covered by ${quote_local_part.
1838 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1841 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1843 83. Patch from Sieve maintainer to fix unterminated string problem in
1844 "vacation" handling.
1846 84. Some minor changes to the Linux configuration files to help with other
1847 OS variants using glibc.
1849 85. One more patch for Sieve to update vacation handling to latest spec.
1852 ----------------------------------------------------
1853 See the note above about the 4.44 and 4.50 releases.
1854 ----------------------------------------------------
1860 1. Change 4.43/35 introduced a bug that caused file counts to be
1861 incorrectly computed when quota_filecount was set in an appendfile
1864 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1865 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1868 3. Give more explanation in the error message when the command for a transport
1869 filter fails to execute.
1871 4. There are several places where Exim runs a non-Exim command in a
1872 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1873 was being done only for the command run by the queryprogram router. It is
1874 now done for all such subprocesses. The other cases are: ${run, transport
1875 filters, and the commands run by the lmtp and pipe transports.
1877 5. Some older OS have a limit of 256 on the maximum number of file
1878 descriptors. Exim was using setrlimit() to set 1000 as a large value
1879 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1880 systems. I've change it so that if it can't get 1000, it tries for 256.
1882 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1883 was an oversight, and furthermore, ever since the addition of extra
1884 controls (e.g. 4.43/32), the checks on when to allow different forms of
1885 "control" were broken. There should now be diagnostics for all cases when a
1886 control that does not make sense is encountered.
1888 7. $recipients is now available in the predata ACL (oversight).
1890 8. Tidy the search cache before the fork to do a delivery from a message
1891 received from the command line. Otherwise the child will trigger a lookup
1892 failure and thereby defer the delivery if it tries to use (for example) a
1893 cached ldap connection that the parent has called unbind on.
1895 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1896 of $address_data from the recipient verification was clobbered by the
1897 sender verification.
1899 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1900 was its contents. (It was OK if the option was not defined at all.)
1902 11. A "Completed" log line is now written for messages that are removed from
1903 the spool by the -Mrm option.
1905 12. $host_address is now set to the target address during the checking of
1906 ignore_target_hosts.
1908 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1909 being passed; this would have caused $sender_host_name to have been used if
1910 matching the list had actually called for a host name (not very likely,
1911 since this list is usually IP addresses). A host name is now passed as
1914 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1915 code that specifies a non-restarting handler (typically sigaction() in
1916 modern systems) in an attempt to fix a rare and obscure crash bug.
1918 15. Narrowed the window for a race in the daemon that could cause it to ignore
1919 SIGCHLD signals. This is not a major problem, because they are used only to
1920 wake it up if nothing else does.
1922 16. A malformed maildirsize file could cause Exim to calculate negative values
1923 for the mailbox size or file count. Odd effects could occur as a result.
1924 The maildirsize information is now recalculated if the size or filecount
1927 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1928 support for a long time. Removed HAVE_SYS_VFS_H.
1930 18. Updated exipick to current release from John Jetmore.
1932 19. Allow an empty sender to be matched against a lookup in an address list.
1933 Previously the only cases considered were a regular expression, or an
1936 20. Exim went into a mad DNS lookup loop when doing a callout where the
1937 host was specified on the transport, if the DNS lookup yielded more than
1940 21. The RFC2047 encoding function was originally intended for short strings
1941 such as real names; it was not keeping to the 75-character limit for
1942 encoded words that the RFC imposes. It now respects the limit, and
1943 generates multiple encoded words if necessary. To be on the safe side, I
1944 have increased the buffer size for the ${rfc2047: expansion operator from
1947 22. Failure to deliver a bounce message always caused it to be frozen, even if
1948 there was an errors_to setting on the router. The errors_to setting is now
1951 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1952 canonical form (fully expanded) before being placed in
1953 $sender_host_address.
1955 24. Updated eximstats to version 1.33
1957 25. Include certificate and key file names in error message when GnuTLS fails
1958 to set them up, because the GnuTLS error message doesn't include the name
1959 of the failing file when there is a problem reading it.
1961 26. Expand error message when OpenSSL has problems setting up cert/key files.
1964 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1965 (this can affect the format of dates).
1967 28. exim_tidydb, when checking for the continued existence of a message for
1968 which it has found a message-specific retry record, was not finding
1969 messages that were in split spool directories. Consequently, it was
1970 deleting retry records that should have stayed in existence.
1972 29. eximstats updated to version 1.35
1973 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1974 1.35 - bugfix such that pie charts by volume are generated correctly
1976 30. The SPA authentication driver was not abandoning authentication and moving
1977 on to the next authenticator when an expansion was forced to fail,
1978 contradicting the general specification for all authenticators. Instead it
1979 was generating a temporary error. It now behaves as specified.
1981 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1982 (the order specifies the preference for clients). The order is now AES256,
1983 AES128, 3DES, ARCFOUR128.
1985 31. Small patch to Sieve code - explicitly set From: when generating an
1988 32. Exim crashed if a remote delivery caused a very long error message to be
1989 recorded - for instance if somebody sent an entire SpamAssassin report back
1990 as a large number of 550 error lines. This bug was coincidentally fixed by
1991 increasing the size of one of Exim's internal buffers (big_buffer) that
1992 happened as part of the Exiscan merge. However, to be on the safe side, I
1993 have made the code more robust (and fixed the comments that describe what
1996 33. Some experimental protocols are using DNS PTR records for new purposes. The
1997 keys for these records are domain names, not reversed IP addresses. The
1998 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1999 leaves it alone. Component reversal etc. now happens only for IP addresses.
2002 34. The host_aton() function is supposed to be passed a string that is known
2003 to be a valid IP address. However, in the case of IPv6 addresses, it was
2004 not checking this. This is a hostage to fortune. Exim now panics and dies
2005 if the condition is not met. A case was found where this could be provoked
2006 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2007 components; fortuitously, this particular loophole had already been fixed
2008 by change 4.50/55 or 4.44/33 above.
2010 If there are any other similar loopholes, the new check in host_aton()
2011 itself should stop them being exploited. The report I received stated that
2012 data on the command line could provoke the exploit when Exim was running as
2013 exim, but did not say which command line option was involved. All I could
2014 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2015 running as the user.
2018 35. There was a buffer overflow vulnerability in the SPA authentication code
2019 (which came originally from the Samba project). I have added a test to the
2020 spa_base64_to_bits() function which I hope fixes it.
2023 36. The daemon start-up calls getloadavg() while still root for those OS that
2024 need the first call to be done as root, but it missed one case: when
2025 deliver_queue_load_max is set with deliver_drop_privilege. This is
2026 necessary for the benefit of the queue runner, because there is no re-exec
2027 when deliver_drop_privilege is set.
2029 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2030 was in use, was not putting the data itself into the right store pool;
2031 consequently, it could be overwritten for a subsequent message in the same
2032 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2035 38. Sometimes the final signoff response after QUIT could fail to get
2036 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2037 < 0 before doing a fflush(). This bug looks as though it goes back to the
2038 introduction of TLS in release 3.20, but "sometimes" must have been rare
2039 because the tests only now provoked it.
2045 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2046 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2047 Now it returns an integer. A number of calls were still expecting a T/F
2048 return. Fortuitously, in all cases, the tests worked in OK situations,
2049 which is the norm. However, things would have gone wrong on any write
2050 failures on the smtp file descriptor. This function is used when sending
2051 messages over SMTP and also when doing verify callouts.
2053 2. When Exim is called to do synchronous delivery of a locally submitted
2054 message (the -odf or -odi options), it no longer closes stderr before doing
2057 3. Implemented the mua_wrapper option.
2059 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2061 5. Implemented the functions header_remove(), header_testname(),
2062 header_add_at_position(), and receive_remove_recipient(), and exported them
2065 6. If an ACL "warn" statement specified the addition of headers, Exim already
2066 inserted X-ACL-Warn: at the start if there was no header name. However, it
2067 was not making this test for the second and subsequent header lines if
2068 there were newlines in the string. This meant that an invalid header could
2069 be inserted if Exim was badly configured.
2071 7. Allow an ACL "warn" statement to add header lines at the start or after all
2072 the Received: headers, as well as at the end.
2074 8. Added the rcpt_4xx retry error code.
2076 9. Added postmaster_mailfrom=xxx to callout verification option.
2078 10. Added mailfrom=xxxx to the callout verification option, for verify=
2081 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2082 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2084 12. Inserted some casts to stop certain compilers warning when using pointer
2085 differences as field lengths or precisions in printf-type calls (mostly
2086 affecting debugging statements).
2088 13. Added optional readline() support for -be (dynamically loaded).
2090 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2091 same clock tick as a message's arrival, so that its received time was the
2092 same as the "first fail" time on the retry record, and that message
2093 remained on the queue past the ultimate address timeout, every queue runner
2094 would try a delivery (because it was past the ultimate address timeout) but
2095 after another failure, the ultimate address timeout, which should have then
2096 bounced the address, did not kick in. This was a "< instead of <=" error;
2097 in most cases the first failure would have been in the next clock tick
2098 after the received time, and all would be well.
2100 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2101 being recognized when the domain list was tested by the match_domain
2102 condition in an expansion string.
2104 16. Added the ${str2b64: operator.
2106 17. Exim was always calling setrlimit() to set a large limit for the number of
2107 processes, without checking whether the existing limit was already
2108 adequate. (It did check for the limit on file descriptors.) Furthermore,
2109 errors from getrlimit() and setrlimit() were being ignored. Now they are
2110 logged to the main and panic logs, but Exim does carry on, to try to do its
2111 job under whatever limits there are.
2113 18. Imported PCRE 5.0.
2115 19. Trivial typo in log message " temporarily refused connection" (the leading
2118 20. If the log selector return_path_on_delivery was set and an address was
2119 redirected to /dev/null, the delivery process crashed because it assumed
2120 that a return path would always be set for a "successful" delivery. In this
2121 case, the whole delivery is bypassed as an optimization, and therefore no
2124 21. Internal re-arrangement: the function for sending a challenge and reading
2125 a response while authentication was assuming a zero-terminated challenge
2126 string. It's now changed to take a pointer and a length, to allow for
2127 binary data in such strings.
2129 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2131 23. Exim was not respecting finduser_retries when seeking the login of the
2132 uid under which it was called; it was always trying 10 times. (The default
2133 setting of finduser_retries is zero.) Also, it was sleeping after the final
2134 failure, which is pointless.
2136 24. Implemented tls_on_connect_ports.
2138 25. Implemented acl_smtp_predata.
2140 26. If the domain in control=submission is set empty, Exim assumes that the
2141 authenticated id is a complete email address when it generates From: or
2142 Sender: header lines.
2144 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2145 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2146 chown and chgrp in /bin and hostname in /usr/bin.
2148 28. Exim was keeping the "process log" file open after each use, just as it
2149 does for the main log. This opens the possibility of it remaining open for
2150 long periods when the USR1 signal hits a daemon. Occasional processlog
2151 errors were reported, that could have been caused by this. Anyway, it seems
2152 much more sensible not to leave this file open at all, so that is what now
2155 29. The long-running daemon process does not normally write to the log once it
2156 has entered its main loop, and it closes the log before doing so. This is
2157 so that log files can straightforwardly be renamed and moved. However,
2158 there are a couple of unusual error situations where the daemon does write
2159 log entries, and I had neglected to close the log afterwards.
2161 30. The text of an SMTP error response that was received during a remote
2162 delivery was being truncated at 512 bytes. This is too short for some of
2163 the long messages that one sometimes sees. I've increased the limit to
2166 31. It is now possible to make retry rules that apply only when a message has a
2167 specific sender, in particular, an empty sender.
2169 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2170 it possible to be selective about when SMTP synchronization is enforced.
2172 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2174 32. Implemented hosts_connection_nolog.
2176 33. Added an ACL for QUIT.
2178 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2181 35. Added mailbox_size and mailbox_filecount to appendfile.
2183 36. Added control = no_multiline_responses to ACLs.
2185 37. There was a bug in the logic of the code that waits for the clock to tick
2186 in the case where the clock went backwards by a substantial amount such
2187 that the microsecond fraction of "now" was more than the microsecond
2188 fraction of "then" (but the whole seconds number was less).
2190 38. Added support for the libradius Radius client library this is found on
2191 FreeBSD (previously only the radiusclient library was supported).
2197 1. When certain lookups returned multiple values in the form name=value, the
2198 quoting of the values was not always being done properly. Specifically:
2199 (a) If the value started with a double quote, but contained no whitespace,
2201 (b) If the value contained whitespace other than a space character (i.e.
2202 tabs or newlines or carriage returns) it was not quoted.
2203 This fix has been applied to the mysql and pgsql lookups by writing a
2204 separate quoting function and calling it from the lookup code. The fix
2205 should probably also be applied to nisplus, ibase and oracle lookups, but
2206 since I cannot test any of those, I have not disturbed their existing code.
2208 2. A hit in the callout cache for a specific address caused a log line with no
2209 reason for rejecting RCPT. Now it says "Previous (cached) callout
2210 verification failure".
2212 3. There was an off-by-one bug in the queryprogram router. An over-long
2213 return line was truncated at 256 instead of 255 characters, thereby
2214 overflowing its buffer with the terminating zero. As well as fixing this, I
2215 have increased the buffer size to 1024 (and made a note to document this).
2217 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2218 when Exim is waiting for an SMTP response from a remote server, Exim
2219 restarts its select() call on the socket, thereby resetting its timeout.
2220 This is not a problem when such interrupts are rare. Somebody set up a cron
2221 job to run exiwhat every 2 minutes, which is less than the normal select()
2222 timeout (5 or 10 minutes). This meant that the select() timeout never
2223 kicked in because it was always reset. I have fixed this by comparing the
2224 time when an interrupt arrives with the time at the start of the first call
2225 to select(). If more time than the timeout has elapsed, the interrupt is
2226 treated as a timeout.
2228 5. Some internal re-factoring in preparation for the addition of Sieve
2229 extensions (by MH). In particular, the "personal" test is moved to a
2230 separate function, and given an option for scanning Cc: and Bcc: (which is
2231 not set for Exim filters).
2233 6. When Exim created an email address using the login of the caller as the
2234 local part (e.g. when creating a From: or Sender: header line), it was not
2235 quoting the local part when it contained special characters such as @.
2237 7. Installed new OpenBSD configuration files.
2239 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2240 try to make them clearer.
2242 9. Callout options, other than the timeout value, were being ignored when
2243 verifying sender addresses in header lines. For example, when using
2245 verify = header_sender/callout=no_cache
2247 the cache was (incorrectly) being used.
2249 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2250 only the Cygwin environment.
2252 11. When return_path_on_delivery was set as a log selector, if different remote
2253 addresses in the same message used different return paths and parallel
2254 remote delivery occurred, the wrong values would sometimes be logged.
2255 (Whenever a remote delivery process finished, the return path value from
2256 the most recently started remote delivery process was logged.)
2258 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2259 lines when AUTH and/or TLS are in use. This is the "received protocol"
2260 field. Exim used to use "asmtp" for authenticated SMTP, without any
2261 indication (in the protocol name) for TLS use. Now it follows the RFC and
2262 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2263 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2264 names appear in log lines as well as in Received: header lines.
2266 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2267 extensions, and comparison tests, and to fix some bugs.
2269 14. Changes to the "personal" filter test:
2271 (1) The test was buggy in that it was just doing the equivalent of
2272 "contains" tests on header lines. For example, if a user's address was
2273 anne@some.where, the "personal" test would incorrectly be true for
2275 To: susanne@some.where
2277 This test is now done by extracting each address from the header in turn,
2278 and checking the entire address. Other tests that are part of "personal"
2279 are now done using regular expressions (for example, to check local parts
2280 of addresses in From: header lines).
2282 (2) The list of non-personal local parts in From: addresses has been
2283 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2284 taken from the Sieve specification recommendations.
2286 (3) If the message contains any header line starting with "List-" it is
2287 treated as non-personal.
2289 (4) The test for "circular" in the Subject: header line has been removed
2290 because it now seems ill-conceived.
2292 15. Minor typos in src/EDITME comments corrected.
2294 16. Installed latest exipick from John Jetmore.
2296 17. If headers_add on a router specified a text string that was too long for
2297 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2298 of string_sprintf() is now avoided.
2300 18. $message_body_size was not set (it was always zero) when running the DATA
2301 ACL and the local_scan() function.
2303 19. For the "mail" command in an Exim filter, no default was being set for
2304 the once_repeat time, causing a random time value to be used if "once" was
2305 specified. (If the value happened to be <= 0, no repeat happened.) The
2306 default is now 0s, meaning "never repeat". The "vacation" command was OK
2307 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2308 (I found it when inspecting the code).
2310 20. There is now an overall timeout for performing a callout verification. It
2311 defaults to 4 times the callout timeout, which applies to individual SMTP
2312 commands during the callout. The overall timeout applies when there is more
2313 than one host that can be tried. The timeout is checked before trying the
2314 next host. This prevents very long delays if there are a large number of
2315 hosts and all are timing out (e.g. when the network connections are timing
2316 out). The value of the overall timeout can be changed by specifying an
2317 additional sub-option for "callout", called "maxwait". For example:
2319 verify = sender/callout=5s,maxwait=20s
2321 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2322 to the end before writing, but this should make it even safer).
2324 22. Exim was forgetting that it had advertised PIPELINING for the second and
2325 subsequent messages on an SMTP connection. It was also not resetting its
2326 memory on STARTTLS and an internal HELO.
2328 23. When Exim logs an SMTP synchronization error within a session, it now
2329 records whether PIPELINING has been advertised or not.
2331 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2332 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2333 rather than long int.
2335 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2337 26. Added the never_mail option to autoreply.
2343 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2344 crash if the getsockname() call failed; this can happen if a connection is
2345 closed very soon after it is established. The problem was simply in the
2346 order in which certain operations were done, causing Exim to try to write
2347 to the SMTP stream before it had set up the file descriptor. The bug has
2348 been fixed by making things happen in the correct order.
2354 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2355 before the connection was closed, thus losing the rejection response.
2357 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2358 some early Solaris releases, but causes trouble in current releases where
2359 socklen_t is defined.
2361 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2364 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2367 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2368 that releasing the top of it at the end releases what was used for sub-
2369 expansions (unless the block got too big). However, discard this block if
2370 the first thing is a variable or header, so that we can use its block when
2371 it is dynamic (useful for very large $message_headers, for example).
2373 6. Lookups now cache *every* query, not just the most recent. A new, separate
2374 store pool is used for this. It can be recovered when all lookup caches are
2375 flushed. Lookups now release memory at the end of their result strings.
2376 This has involved some general refactoring of the lookup sources.
2378 7. Some code has been added to the store_xxx() functions to reduce the amount
2379 of flapping under certain conditions.
2381 8. log_incoming_interface used to affect only the <= reception log lines. Now
2382 it causes the local interface and port to be added to several more SMTP log
2383 lines, for example "SMTP connection from", and rejection lines.
2385 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2387 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2389 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2390 could be overwritten at the end of the current message (or the start of a
2391 new message if it was set in a HELO ACL). The value is now preserved for
2392 the duration of the SMTP connection.
2394 12. If a transport had a headers_rewrite setting, and a matching header line
2395 contained an unqualified address, that address was qualified, even if it
2396 did not match any rewriting rules. The underlying bug was that the values
2397 of the flags that permit the existence of unqualified sender and recipient
2398 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2399 non-local messages, and by -bnq for local messages) were not being
2400 preserved with the message after it was received.
2402 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2403 "next input=" as part of the text comprising the host identity instead of
2404 the correct text. The code was using the same buffer for two different
2405 strings. However, depending on which order the printing function evaluated
2406 its arguments, the bug did not always show up. Under Linux, for example, my
2407 test suite worked just fine.
2409 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2410 doesn't work with some older versions of Perl. It has been changed to "my",
2411 which in any case is probably the better facility to use.
2413 15. A really picky compiler found some instances of statements for creating
2414 error messages that either had too many or two few arguments for the format
2417 16. The size of the buffer for calls to the DNS resolver has been increased
2418 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2419 for addresses that have a lot of PTR records. This alleviates a problem; it
2420 does not fully solve it.
2422 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2423 buffer now truncates the list and logs the incident, which is the same
2424 action as happens when Exim is looking up a host name and its aliases.
2425 Previously in this situation something unpredictable would happen;
2426 sometimes it was "internal error: store_reset failed".
2428 18. If a server dropped the connection unexpectedly when an Exim client was
2429 using GnuTLS and trying to read a response, the client delivery process
2430 crashed while trying to generate an error log message.
2432 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2433 string, for example:
2435 warn message = H1: something\nH2: something
2437 the text was added as a single header line from Exim's point of view
2438 though it ended up OK in the delivered message. However, searching for the
2439 second and subsequent header lines using $h_h2: did not work. This has been
2440 fixed. Similarly, if a system filter added multiple headers in this way,
2441 the routers could not see them.
2443 20. Expanded the error message when iplsearch is called with an invalid key to
2444 suggest using net-iplsearch in a host list.
2446 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2447 ACLs are no longer actually imposed (and a message to that effect is
2450 22. If a "gecos" field in a passwd entry contained escaped characters, in
2451 particular, if it contained a \" sequence, Exim got it wrong when building
2452 a From: or a Sender: header from that name. A second bug also caused
2453 incorrect handling when an unquoted " was present following a character
2454 that needed quoting.
2456 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2457 was not being matched caselessly.
2459 24. Arranged for all hyphens in the exim.8 source to be escaped with
2462 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2463 properly. Recipient callout cache records were still being keyed to include
2464 the sender, even when use_sender was set false. This led to far more
2465 callouts that were necessary. The sender is no longer included in the key
2466 when use_sender is false.
2468 26. Added "control = submission" modifier to ACLs.
2470 27. Added the ${base62d: operator to decode base 62 numbers.
2472 28. dnsdb lookups can now access SRV records.
2474 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2475 the configuration file.
2477 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2478 (-v) mode. This makes the output for a verbose queue run more intelligible.
2480 31. Added a use_postmaster feature to recipient callouts.
2482 32. Added the $body_zerocount variable, containing the number of binary zero
2483 bytes in the message body.
2485 33. The time of last modification of the "new" subdirectory is now used as the
2486 "mailbox time last read" when there is a quota error for a maildir
2489 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2491 35. Added +ignore_unknown as a special item in host lists.
2493 36. Code for decoding IPv6 addresses in host lists is now included, even if
2494 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2495 address was recognized as an IP address, but was then not correctly decoded
2496 into binary, causing unexpected and incorrect effects when compared with
2503 1. Very minor rewording of debugging text in manualroute to say "list of
2504 hosts" instead of "hostlist".
2506 2. If verify=header_syntax was set, and a header line with an unqualified
2507 address (no domain) and a large number of spaces between the end of the
2508 name and the colon was received, the reception process suffered a buffer
2509 overflow, and (when I tested it) crashed. This was caused by some obsolete
2510 code that should have been removed. The fix is to remove it!
2512 3. When running in the test harness, delay a bit after writing a bounce
2513 message to get a bit more predictability in the log output.
2515 4. Added a call to search_tidyup() just before forking a reception process. In
2516 theory, someone could use a lookup in the expansion of smtp_accept_max_
2517 per_host which, without the tidyup, could leave open a database connection.
2519 5. Added the variables $recipient_data and $sender_data which get set from a
2520 lookup success in an ACL "recipients" or "senders" condition, or a router
2521 "senders" option, similar to $domain_data and $local_part_data.
2523 6. Moved the writing of debug_print from before to after the "senders" test
2526 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2527 problems for message scanning, either using a data ACL, or using
2528 local_scan() because the Received: header was not generated till after they
2529 were called (in order to set the time as the time of reception completion).
2530 I have revised the way this works. The header is now generated after the
2531 body is received, but before the ACL or local_scan() are called. After they
2532 are run, the timestamp in the header is updated.
2538 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2539 before starting a queue runner without re-exec. This happened only when
2540 deliver_drop_privilege was set or when the Exim user was set to root. The
2541 effect of the bug was that timeouts during subsequent deliveries caused
2542 crashes instead of being properly handled. The handler is now left at its
2543 default (and expected) setting.
2545 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2546 message, again when deliver_drop_privilege is set or Exim is run as root.
2547 The bug described in (1) was not present in this case, but the tidying up
2548 of the other signals was missing. I have made the two cases consistent.
2550 3. The ignore_target_hosts setting on a manualroute router was being ignored
2551 for hosts that were looked up using the /MX notation.
2553 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2556 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2557 operated on the sender address. After changing the $sender_address to <>
2558 for the sender address verify, Exim was re-instated it as the original
2559 (before rewriting) address, but remembering that it had rewritten it, so it
2560 wasn't rewriting it again. This bug also had the effect of breaking the
2561 sender address verification caching when the sender address was rewritten.
2563 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2564 This has been changed so that if the ip literal address matches
2565 ignore_target_hosts, the router declines.
2567 7. Added expansion conditions match_domain, match_address, and match_local_
2568 part (NOT match_host).
2570 8. The placeholder for the Received: header didn't have a length field set.
2572 9. Added code to Exim itself and to exim_lock to test for a specific race
2573 condition that could lead to file corruption when using MBX delivery. The
2574 issue is with the lockfile that is created in /tmp. If this file is removed
2575 after a process has opened it but before that process has acquired a lock,
2576 there is the potential for a second process to recreate the file and also
2577 acquire a lock. This could lead to two Exim processes writing to the file
2578 at the same time. The added code performs the same test as UW imapd; it
2579 checks after acquiring the lock that its file descriptor still refers to
2580 the same named file.
2582 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2583 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2586 11. Added the smtp_active_hostname option. If used, this will typically be made
2587 to depend on the incoming interface address. Because $interface_address is
2588 not set up until the daemon has forked a reception process, error responses
2589 that can happen earlier (such as "too many connections") no longer contain
2592 12. If an expansion in a condition on a "warn" statement fails because a lookup
2593 defers, the "warn" statement is abandoned, and the next ACL statement is
2594 processed. Previously this caused the whole ACL to be aborted.
2596 13. Added the iplsearch lookup type.
2598 14. Added ident_timeout as a log selector.
2600 15. Added tls_certificate_verified as a log selector.
2602 16. Added a global option tls_require_ciphers (compare the smtp transport
2603 option of the same name). This controls incoming TLS connections.
2605 17. I finally figured out how to make tls_require_ciphers do a similar thing
2606 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2607 before starting the TLS session.
2609 18. Tabs are now shown as \t in -bP output.
2611 19. If the log selector return_path_on_delivery was set, Exim crashed when
2612 bouncing a message because it had too many Received: header lines.
2614 20. If two routers both had headers_remove settings, and the first one included
2615 a superfluous trailing colon, the final name in the first list and the
2616 first name in the second list were incorrectly joined into one item (with a
2617 colon in the middle).
2623 1. Added -C and -D options to the exinext utility, mainly to make it easier
2624 to include in the automated testing, but these could be helpful when
2625 multiple configurations are in use.
2627 2. The exinext utility was not formatting the output nicely when there was
2628 an alternate port involved in the retry record key, nor when there was a
2629 message id as well (for retries that were specific to a specific message
2630 and a specific host). It was also confused by IPv6 addresses, because of
2631 the additional colons they contain. I have fixed the IPv4 problem, and
2632 patched it up to do a reasonable job for IPv6.
2634 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2635 delivery, the log line now contains "pipelined" if PIPELINING was used.
2637 4. An SMTP transport process used to panic and die if the bind() call to set
2638 an explicit outgoing interface failed. This has been changed; it is now
2639 treated in the same way as a connect() failure.
2641 5. A reference to $sender_host_name in the part of a conditional expansion
2642 that was being skipped was still causing a DNS lookup. This no longer
2645 6. The def: expansion condition was not recognizing references to header lines
2646 that used bh_ and bheader_.
2648 7. Added the _cache feature to named lists.
2650 8. The code for checking quota_filecount in the appendfile transport was
2651 allowing one more file than it should have been.
2653 9. For compatibility with Sendmail, the command line option
2661 and sets the incoming protocol and host name (for trusted callers). The
2662 host name and its colon can be omitted when only the protocol is to be set.
2663 Note the Exim already has two private options, -pd and -ps, that refer to
2664 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2665 "s", but I don't think that's a major issue.
2667 10. A number of refactoring changes to the code, none of which should affect
2670 (a) The number of logging options was getting close to filling up the
2671 32-bit word that was used as a bit map. I have split them into two classes:
2672 those that are passed in the argument to log_write(), and those that are
2673 only ever tested independently outside of that function. These are now in
2674 separate 32-bit words, so there is plenty of room for expansion again.
2675 There is no change in the user interface or the logging behaviour.
2677 (b) When building, for example, log lines, the code previously used a
2678 macro that called string_cat() twice, in order to add two strings. This is
2679 not really sufficiently general. Furthermore, there was one instance where
2680 it was actually wrong because one of the argument was used twice, and in
2681 one call a function was used. (As it happened, calling the function twice
2682 did not affect the overall behaviour.) The macro has been replaced by a
2683 function that can join an arbitrary number of extra strings onto a growing
2686 (c) The code for expansion conditions now uses a table and a binary chop
2687 instead of a serial search (which was left over from when there were very
2688 few conditions). Also, it now recognizes conditions like "pam" even when
2689 the relevant support is not compiled in: a suitably worded error message is
2690 given if an attempt is made to use such a condition.
2692 11. Added ${time_interval:xxxxx}.
2694 12. A bug was causing one of the ddress fields not to be passed back correctly
2695 from remote delivery subprocesses. The field in question was not being
2696 subsequently used, so this caused to problems in practice.
2698 13. Added new log selectors queue_time and deliver_time.
2700 14. Might have fixed a bug in maildirsizefile handling that threw up
2701 "unexpected character" debug warnings, and recalculated the data
2702 unnecessarily. In any case, I expanded the warning message to give more
2705 15. Added the message "Restricted characters in address" to the statements in
2706 the default ACL that block characters like @ and % in local parts.
2708 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2709 Three changes have been made:
2711 (a) There was a serious bug; a negative response to MAIL caused the whole
2712 recipient domain to be cached as invalid, thereby blocking all messages
2713 to all local parts at the same domain, from all senders. This bug has
2714 been fixed. The domain is no longer cached after a negative response to
2715 MAIL if the sender used is not empty.
2717 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2720 (c) A new callout option, "use_sender" has been added for people who want
2721 the modified behaviour.
2727 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2730 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2731 indeed breaks things for older releases.
2733 3. Added additional logging to the case where there is a problem reading data
2734 from a filter that is running in a subprocess using a pipe, in order to
2735 try to track down a specific problem.
2737 4. Testing facility fudge: when running in the test harness and attempting
2738 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2739 getting "No route to host". Convert this to a timeout.
2741 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2744 6. Some OS don't have socklen_t but use size_t instead. This affects the
2745 fifth argument of getsockopt() amongst other things. This is now
2746 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2747 can be set for individual OS. I have set it for SunOS5, OSF1, and
2748 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2749 some earlier ones do not.
2751 7. Change 4.30/15 was not doing the test caselessly.
2753 8. The standard form for an IPv6 address literal was being rejected by address
2754 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2755 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2756 this, as well as the form without the "IPv6" on the front (but only when
2757 address literals are enabled, of course).
2759 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2761 10. Exim crashed if a message with an empty sender address specified by -f
2762 encountered a router with an errors_to setting. This could be provoked only
2763 by a command such as
2767 where an empty string was supplied; "<>" did not hit this bug.
2769 11. Installed PCRE release 4.5.
2771 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2772 remained set. It is now erased.
2774 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2775 times from message ids (which are base 36 rather than the normal 62).
2777 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2778 were being counted as actual protocol errors, and logged if the log
2779 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2780 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2781 and DATA following a set of rejected RCPTs do not count as protocol errors.
2782 In other words, Exim assumes they were pipelined, though this may not
2783 actually be the case. Of course, in all cases the client gets an
2784 appropriate error code.
2786 15. If a lookup fails in an ACL condition, a message about the failure may
2787 be available; it is used if testing the ACL cannot continue, because most
2788 such messages specify what the cause of the deferral is. However, some
2789 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2790 that caused an old message to be retained and used if a later statement
2791 caused a defer, replacing the real cause of the deferral.
2793 16. If an IP address had so many PTR records that the DNS lookup buffer
2794 was not large enough to hold them, Exim could crash while trying to process
2795 the truncated data. It now detects and logs this case.
2797 17. Further to 4.21/58, another change has been made: if (and only if) the
2798 first line of a message (the first header line) ends with CRLF, a bare LF
2799 in a subsequent header line has a space inserted after it, so as not to
2800 terminate the header.
2802 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2803 unnecessarily, used atoi() instead of strtol(), and didn't check the
2804 termination when getting file sizes from file names by regex.
2806 19. Completely re-implemented the support for maildirsize files, in the light
2807 of a number of problems with the previous contributed implementation
2808 (4.30/29). In particular:
2810 . If the quota is zero, the maildirsize file is maintained, but no quota is
2813 . If the maildir directory does not exist, it is created before any attempt
2814 to write a maildirsize file.
2816 . The quota value in the file is just a cache; if the quota is changed in
2817 the transport, the new value overrides.
2819 . A regular expression is available for excluding directories from the
2822 20. The autoreply transport checks the characters in options that define the
2823 message's headers; it allows continued headers, but it was checking with
2824 isspace() after an embedded newline instead of explicitly looking for a
2827 21. If all the "regular" hosts to which an address was routed had passed their
2828 expiry times, and had not reached their retry times, the address was
2829 bounced, even if fallback hosts were defined. Now Exim should go on to try
2832 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2833 equivalent code in the SMTP transport. Some hosts send humungous responses
2834 to HELO/EHLO, more than 1024 it seems.
2836 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2837 gives compiler warnings in some environments. I've now done it "properly",
2840 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2841 (because of problems with the built-in one) was declared to return uschar *
2842 instead of char *, causing compiler failure.
2844 25. Fixed a file descriptor leak when processing alias/forward files.
2846 26. Fixed a minor format string issue in dbfn.c.
2848 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2850 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2851 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2854 29. When a sender address is verified, it is cached, to save repeating the test
2855 when there is more than one recipient in a message. However, when the
2856 verification involves a callout, it is possible for different callout
2857 options to be set for different recipients. It is too complicated to keep
2858 track of this in the cache, so now Exim always runs a verification when a
2859 callout is required, relying on the callout cache for the optimization.
2860 The overhead is duplication of the address routing, but this should not be
2863 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2864 to be verified with callout=postmaster, and the main callout worked but the
2865 postmaster check failed, the verification correctly failed. However, if a
2866 subsequent RCPT command asked for sender verification *without* the
2867 postmaster check, incorrect caching caused this verification also to fail,
2870 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2871 it was not caching the DNS options (qualify_single, search_parents) that
2872 were used when the lookup failed. A subsequent lookup with different
2873 options therefore always gave the same answer, though there were cases
2874 where it should not have. (Example: a "domains = !$mx_any" option on a
2875 dnslookup router: the "domains" option is always processed without any
2876 widening, but the router might have qualify_single set.) Now Exim uses the
2877 cached value only when the same options are set.
2879 32. Added John Jetmore's "exipick" utility to the distribution.
2881 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2882 than a timeout (e.g. a certificate is required, and is not provided), an
2883 Exim server now closes the connection immediately. Previously it waited for
2884 the client to close - but if the client is SSL, it seems that they each
2885 wait for each other, leading to a delay before one of them times out.
2887 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2888 maintained 0.8.x compatibility because I don't think many are using it, and
2889 it is clearly obsolete.
2891 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2894 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2895 client certificate was expired. A simple patch fixes this, though I don't
2896 understand the full logic of why the verify callback is called multiple
2899 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2900 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2901 which causes problems with some clients (such as the Certicom SSL Plus
2902 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2903 disables the coutermeasure allowing Eudora to connect."
2905 38. Exim was not checking that a write() to a log file succeeded. This could
2906 lead to Bad Things if a log got too big, in particular if it hit a file
2907 size limit. Exim now panics and dies if it cannot write to a log file, just
2908 as it does if it cannot open a log file.
2910 39. Modified OS/Makefile-Linux so that it now contains
2912 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2914 The two -D definitions ensure that Exim is compiled with large file
2915 support, which makes it possible to handle log files that are bigger than
2918 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2919 instance) a domain was checked against a named list that involved a lookup,
2920 causing $domain_data to be set, then another domain was checked against the
2921 same list, then the first domain was re-checked, the value of $domain_data
2922 after the final check could be wrong. In particular, if the second check
2923 failed, it could be set empty. This bug probably also applied to
2926 41. The strip_trailing_dot option was not being applied to the address given
2927 with the -f command-line option.
2929 42. The code for reading a message's header from the spool was incrementing
2930 $received_count, but never initializing it. This meant that the value was
2931 incorrect (doubled) while delivering a message in the same process in which
2932 it was received. In the most common configuration of Exim, this never
2933 happens - a fresh exec is done - but it can happen when
2934 deliver_drop_privilege is set.
2936 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2937 it now includes up to 150 characters of the unexpected data in the log
2940 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2941 and building data strings. The size of both of these buffers was 10 000
2942 bytes - far larger than anybody would *ever* want, thought I. Needless to
2943 say, somebody hit the limit. I have increased the maximum line length to
2944 20 000 and the maximum data length of concatenated lines to 100 000. I have
2945 also fixed two bugs, because there was no checking on these buffers. Tsk,
2946 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2947 buffer is too small.
2949 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2950 lsearch lookups. Now it does.
2952 46. When parsing a route_list item in a manualroute router, a fixed-length
2953 buffer was used for the list of hosts. I made this 1024 bytes long,
2954 thinking that nobody would ever have a list of hosts that long. Wrong.
2955 Somebody had a whole pile of complicated expansion conditions, and the
2956 string was silently truncated, leading to an expansion error. It turns out
2957 that it is easier to change to an unlimited length (owing to other changes
2958 that have happened since this code was originally written) than to build
2959 structure for giving a limitation error. The length of the item that
2960 expands into the list of hosts is now unlimited.
2962 47. The lsearch lookup could not handle data where the length of text line was
2963 more than 4095 characters. Such lines were truncated, leading to shortened
2964 data being returned. It should now handle lines of any length.
2966 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2967 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2970 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2971 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2972 no longer happens in comment lines. A list of replacements is now placed
2973 at the head of all of the source files, except those whose only change is
2974 to replace PERL_COMMAND in the very first #! line.
2976 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2977 messages on the queue, with a bottom-up merge sort, using code contributed
2978 by Michael Haardt. This should make operations like -bp somewhat faster on
2979 large queues. It won't affect queue runners, except when queue_run_in_order
2982 51. Installed eximstats 1.31 in the distribution.
2984 52. Added support for SRV lookups to the dnslookup router.
2986 53. If an ACL referred to $message_body or $message_body_end, the value was not
2987 reset for any messages that followed in the same SMTP session.
2989 54. The store-handling optimization for building very long strings was not
2990 differentiating between the different store pools. I don't think this
2991 actually made any difference in practice, but I've tidied it.
2993 55. While running the routers to verify a sender address, $sender_address
2994 was still set to the sender address. This is wrong, because when routing to
2995 send a bounce to the sender, it would be empty. Therefore, I have changed
2996 it so that, while verifying a sender address, $sender_address is set to <>.
2997 (There is no change to what happens when verifying a recipient address.)
2999 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3000 target A or AAAA records (if not already returned) without resetting the
3001 qualify_single or search_parents options of the DNS resolver. These are
3002 inappropriate in this case because the targets of MX and SRV records must
3003 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3004 target that, when qualified, matched something in the local domain. These
3005 two options are now turned off when doing these lookups.
3007 57. It seems that at least some releases of Reiserfs (which does not have the
3008 concept of a fixed number of inodes) returns zero and not -1 for the
3009 number of available inodes. This interacted badly with check_spool_inodes,
3010 which assumed that -1 was the "no such thing" setting. What I have done is
3011 to check that the total number of inodes is greater than zero before doing
3012 the test of how many are available.
3014 58. When a "warn" ACL statement has a log_message modifier, the message is
3015 remembered, and not repeated. This is to avoid a lot of repetition when a
3016 message has many recipients that cause the same warning to be written.
3017 Howewer, Exim was preserving the list of already written lines for an
3018 entire SMTP session, which doesn't seem right. The memory is now reset if a
3019 new message is started.
3021 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3022 debugging output unless log_rewrite was also set.
3024 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3025 of (int)(handle) when we know that handle contains (void *)(-1).
3027 61. The Exim daemon panic-logs an error return when it closes the incoming
3028 connection. However "connection reset by peer" seems to be common, and
3029 isn't really an error worthy of noting specially, so that particular error
3032 62. When Exim is trying to find all the local interfaces, it used to panic and
3033 die if the ioctl to get the interface flags failed. However, it seems that
3034 on at least one OS (Solaris 9) it is possible to have an interface that is
3035 included in the list of interfaces, but for which you get a failure error
3036 for this call. This happens when the interface is not "plumbed" into a
3037 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3038 failure of the "get flags" call assumes that the interface is down.
3040 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3041 makes life easier for people who are doing arithmetic on fields extracted
3042 from dates, where you often get leading zeros that should not be
3043 interpreted as octal.
3045 64. Added qualify_domain to the redirect router, to override the global
3048 65. If a pathologically long header line contained very many addresses (the
3049 report of this problem mentioned 10 000) and each of them was rewritten,
3050 Exim could use up a very large amount of memory. (It kept on making new
3051 copies of the header line as it rewrote, and never released the old ones.)
3052 At the expense of a bit more processing, the header rewriting function has
3053 been changed so that it no longer eats memory in this way.
3055 66. The generation of the Received: header has been moved from the time that a
3056 message starts to be received, to the time that it finishes. The timestamp
3057 in the Received: header should now be very close to that of the <= log
3058 line. There are two side-effects of this change:
3060 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3061 logged header lines no longer include the local Received: line, because
3062 it has not yet been created. The same applies to a copy of the message
3063 that is returned to a non-SMTP sender when a message is rejected.
3065 (b) When a filter file is tested using -bf, no additional Received: header
3066 is added to the test message. After some thought, I decided that this
3069 This change does not affect the value of $received_for. It is still set
3070 after address rewriting, but before local_scan() is called.
3072 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3074 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3075 gave an unhelpful panic error message, and a defer error. I have managed to
3076 change this behaviour so that it now rejects any supplied certificate,
3077 which seems right, as the list of acceptable certificates is empty.
3079 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3080 gave an unhelpful defer error. I have not managed to make this reject any
3081 supplied certificates, but the error message it gives is "no certificate
3082 supplied", which is not helpful.
3084 70. exigrep's output now also includes lines that are not associated with any
3085 message, but which match the given pattern. Implemented by a patch from
3086 Martin Sluka, which also tidied up the Perl a bit.
3088 71. Recipient callout verification, like sender verification, was using <> in
3089 the MAIL FROM command. This isn't really the right thing, since the actual
3090 sender may affect whether the remote host accepts the recipient or not. I
3091 have changed it to use the actual sender in the callout; this means that
3092 the cache record is now keyed on a recipient/sender pair, not just the
3093 recipient address. There doesn't seem to be a real danger of callout loops,
3094 since a callout by the remote host to check the sender would use <>.
3095 [SEE ABOVE: changed after hitting problems.]
3097 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3098 temporary errors. However, in the case of such a code being given after
3099 the end of a data transmission (i.e. after ".") Exim was failing to write
3100 a retry record for the message. (Yes, there was some broken host that was
3101 actually sending 8xx at this point.)
3103 73. An unknown lookup type in a host list could cause Exim to panic-die when
3104 the list was checked. (An example that provoked this was putting <; in the
3105 middle of a list instead of at the start.) If this happened during a DATA
3106 ACL check, a -D file could be left lying around. This kind of configuration
3107 error no longer causes Exim to die; instead it causes a defer errror. The
3108 incident is still logged to the main and panic logs.
3110 74. Buglet left over from Exim 3 conversion. The message "too many messages
3111 in one connection" was written to the rejectlog but not the mainlog, except
3112 when address rewriting (yes!) was being logged.
3114 75. Added write_rejectlog option.
3116 76. When a system filter was run not as root (that is, when system_filter_user
3117 was set), the values of the $n variables were not being returned to the
3118 main process; thus, they were not subsequently available in the $sn
3121 77. Added +return_path_on_delivery log selector.
3123 78. A connection timeout was being treated differently from recipients deferred
3124 when testing hosts_max_try with a message that was older than the host's
3125 retry timeout. (The host should not be counted, thus allowing all hosts to
3126 be tried at least once before bouncing.) This may have been the cause of an
3127 occasionally reported bug whereby a message would remain on the queue
3128 longer than the retry timeout, but would be bounced if a delivery was
3129 forced. I say "may" because I never totally pinned down the problem;
3130 setting up timeout/retry tests is difficult. See also the next item.
3132 79. The ultimate address timeout was not being applied to errors that involved
3133 a combination of host plus message (for example, a timeout on a MAIL
3134 command). When an address resolved to a number of possible hosts, and they
3135 were not all tried for each delivery (e.g. because of hosts_max_try), a
3136 message could remain on the queue longer than the retry timeout.
3138 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3141 81. Fixed an obscure SMTP outgoing bug which required at least the following
3142 conditions: (a) there was another message waiting for the same server;
3143 (b) the server returned 5xx to all RCPT commands in the first message so
3144 that the message was not completed; (c) the server dropped the connection
3145 or gave a negative response to the RSET that Exim sends to abort the
3146 transaction. The observed case was a dropped connection after DATA that had
3147 been sent in pipelining mode. That is, the server had advertised PIPELINING
3148 but was not implementing it correctly. The effect of the bug was incorrect
3149 behaviour, such as trying another host, and this could lead to a crash.
3155 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3156 and daemon.c were passed as pointers to ints; they should have been
3157 pointers to socklen_t variables (which are typically unsigned ints).
3159 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3162 3. Fixed a really odd bug that affected only the testing scheme; patching a
3163 certain fixed string in the binary changed the value of another string that
3164 happened to be identical to the end of the original first string.
3166 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3167 name", it returns that address as the IP address. On some operating
3168 systems (e.g. Solaris), it also passes back the IP address string as the
3169 "host name". However, on others (e.g. Linux), it passes back an empty
3170 string. Exim wasn't checking for this, and was changing the host name to an
3171 empty string, assuming it had been canonicized.
3173 5. Although rare, it is permitted to have more than one PTR record for a given
3174 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3175 all the names associated with an address, because they do in Solaris.
3176 However, it seems that they do not in Linux for data that comes from the
3177 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3178 I found this out when I moved to a new Linux workstation and tried to run
3179 the Exim test suite.
3181 To get round this problem I have changed the code so that it now does its
3182 own call to the DNS to look up PTR records when searching for a host name.
3183 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3184 addresses that are only in /etc/hosts are still found.
3186 This behaviour is, however, controlled by an option called host_lookup_
3187 order, which defaults to "bydns:byaddr". If people want to use the other
3188 order, or indeed, just use one or the other means of lookup, they can
3189 specify it in this variable.
3191 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3192 some operating systems, this comes back from gethostbyaddr() as an empty
3193 string, and this is what Exim used to test for. However, it seems that in
3194 other systems, "." is yielded. Exim now tests for this case too.
3196 7. The values of check_spool_space and check_log_space are now held internally
3197 as a number of kilobytes instead of an absolute number of bytes. If a
3198 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3199 kilobyte. This means that much larger values can be stored.
3201 8. Exim monitor: an attempt to get the action menu when not actually pointing
3202 at a message produces an empty menu entitled "No message selected". This
3203 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3204 no entries in it ("Shell widget menu has zero width and/or height"). So I
3205 have added a single, blank menu entry in this case.
3207 9. Added ${quote_local_part.
3209 10. MIME decoding is now applied to the contents of Subject: header lines when
3212 11. Now that a reference to $sender_host_address automatically causes a reverse
3213 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3214 host lookup before query-style lookups in lists that might use this
3215 variable. This has therefore been abolished, and the "net-" prefix is no
3216 longer necessary for query-style lookups.
3218 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3219 have been a typo for LFLAGS, so it has been changed.
3221 13. The install script calls Exim with "-C /dev/null" in order to find the
3222 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3223 to be output. Howeve, since Exim outputs its version number before the
3224 error, it didn't break the script. It just looked ugly. I fixed this by
3225 always allowing "-C /dev/null" if the caller is root.
3227 14. Ignore overlarge ACL variable number when reading spool file - insurance
3228 against a later release with more variables having written the file.
3230 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3231 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3232 this, as well as the form without the "IPv6" on the front.
3234 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3235 OS/Makefile-Darwin file.
3237 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3238 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3240 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3241 variable. However, some OS use ints for inodes. I've added cast to long int
3242 to get rid of the compiler warning.
3244 19. I had forgotten to lock out "/../" in configuration file names when
3245 ALT_CONFIG_PREFIX was set.
3247 20. Routers used for verification do not need to specify transports. However,
3248 if such a router generated a host list, and callout was configured, Exim
3249 crashed, because it could not find a port number from the (non-existent)
3250 transport. It now assumes port 25 in this circumstance.
3252 21. Added the -t option to exigrep.
3254 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3255 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3256 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3257 them is set, LOOKUP_LSEARCH is forced.
3259 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3261 24. Added sender and host information to the "rejected by local_scan()" log
3262 line; previously there was no indication of these.
3264 25. Added .include_if_exists.
3266 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3267 Linux. It turns out that not all file systems support this. Apparently some
3268 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3269 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3270 not supported on the file descriptor, is now ignored when Exim is trying to
3271 sync a directory. This applies only to Linux.
3273 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3275 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3276 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3279 29. Added support for maildirsize files from supplied patch (modified a bit).
3281 30. The use of :fail: followed by an empty string could lead Exim to respond to
3282 sender verification failures with (e.g.):
3284 550 Verification failed for <xxx>
3285 550 Sender verify failed
3287 where the first response line was missing the '-' that indicates it is not
3288 the final line of the response.
3290 31. The loop for finding the name of the user that called Exim had a hardwired
3291 limit of 10; it now uses the value of finduser_retries, which is used for
3292 all other user lookups.
3294 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3297 33. Exim was neglecting to zero errno before one call of strtol() when
3298 expanding a string and expecting an integer value. On some systems this
3299 resulted in spurious "integer overflow" errors. Also, it was casting the
3300 result into an int without checking.
3302 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3303 did not work. The code looks as if it has *never* worked, though it appears
3304 to have been documented since at least releast 1.62. I have made it work.
3306 35. The "timeout_DNS" error in retry rules, also documented since at least
3307 1.62, also never worked. As it isn't clear exactly what this means, and
3308 clearly it isn't a major issue, I have abolished the feature by treating it
3309 as "timeout", and writing a warning to the main and panic logs.
3311 36. The display of retry rules for -brt wasn't always showing the error code
3314 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3315 timeout_connect_A, timeout_connect_MX.
3317 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3318 to the empty sender.
3320 39. The daemon was not analysing the content of -oX till after it had closed
3321 stderr and disconnected from the controlling terminal. This meant that any
3322 syntax errors were only noted on the panic log, and the return code from
3323 the command was 0. By re-arranging the code a little, I've made the
3324 decoding happen first, so such errors now appear on stderr, and the return
3325 code is 1. However, the actual setting up of the sockets still happens in
3326 the disconnected process, so errors there are still only recorded on the
3329 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3330 connections (as happens on some IP stacks) was logged at start up time as
3331 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3332 it from "IPv6 and IPv4", which means that two separate sockets are being
3335 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3336 says whether AF_INET or AF_INET6 was passed as an argument.
3338 42. Exiwhat output was messed up when time zones were included in log
3341 43. Exiwhat now gives more information about the daemon's listening ports,
3342 and whether -tls-on-connect was used.
3344 44. The "port" option of the smtp transport is now expanded.
3346 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3347 silently ignored. Now an error message is written to the main and panic
3350 46. There's a new ACL modifier called "logwrite" which writes to a log file
3351 as soon as it is encountered.
3353 47. Added $local_user_uid and $local_user_gid at routing time.
3355 48. Exim crashed when trying to verify a sender address that was being
3358 49. Exim was recognizing only a space character after ".include". It now also
3359 recognizes a tab character.
3361 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3362 extracting the relevant information from the specification. The man page no
3363 longer contains scrambled data for the -d option, and I've added a section
3364 at the front about calling Exim under different names.
3366 51. Added "extra_headers" argument to the "mail" command in filter files.
3368 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3371 53. Installed eximstats 1.29.
3373 54. Added transport_filter_timeout as a generic transport option.
3375 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3376 Cc: header lines. This was required by RFC 822, but it not required by RFC
3379 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3380 incoming messages that did not have them. Now it does so only if the
3381 message originates locally, that is, if there is no associated remote host
3382 address. When Resent- header lines are present, this applies to the Resent-
3383 lines rather than the non-Resent- lines.
3385 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3386 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3388 58. Messages for configuration errors now include the name of the main
3389 configuration file - useful now that there may be more than one file in a
3390 list (.included file names were always shown).
3392 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3393 for those rare installations that do not start the daemon as root or run it
3394 setuid root. I've cut out the call to initgroups() if the daemon is not
3397 60. The Exim user and group can now be bound into the binary as text strings
3398 that are looked up at the start of Exim's processing.
3400 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3402 62. Added $mailstore_basename variable.
3404 63. Installed patch to sieve.c from Michael Haardt.
3406 64. When Exim failed to open the panic log after failing to open the main log,
3407 the original message it was trying to log was written to stderr and debug
3408 output, but if they were not available (the usual case in production), it
3409 was lost. Now it is written to syslog before the two lines that record the
3410 failures to open the logs.
3412 65. Users' Exim filters run in subprocesses under the user's uid. It is
3413 possible for a "deliver" command or an alias in a "personal" command to
3414 provoke an address rewrite. If logging of address rewriting is configured,
3415 this fails because the process is not running as root or exim. There may be
3416 a better way of dealing with this, but for the moment (because 4.30 needs
3417 to be released), I have disabled address rewrite logging when running a
3418 filter in a non-root, non-exim process.
3424 1. The buildconfig auxiliary program wasn't quoting the value set for
3425 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3426 not defined. This bug was masked in 4.22 by the effect that was fixed in
3429 2. Some messages that were rejected after a message id was allocated were
3430 shown as "incomplete" by exigrep. It no longer does this for messages that
3431 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3433 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3434 have allow_domain_literals set, the ID did not get logged in the <= line.
3435 Domain literals are now always recognized in Message-ID: header lines.
3437 4. The first argument for a ${extract expansion item is the key name or field
3438 number. Leading and trailing spaces in this item were not being ignored,
3439 causing some misleading effects.
3441 5. When deliver_drop_privilege was set, single queue runner processes started
3442 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3443 same command in the process it spins off) were not dropping privilege.
3445 6. When the daemon running as "exim" started a queue runner, it always
3446 re-executed Exim in the spun-off process. This is a waste of effort when
3447 deliver_drop_privilege is set. The new process now just calls the
3448 queue-runner function directly.
3454 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3457 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3458 ignored. Though the use of -oP was forcing the writing of a pid file, it
3459 was always written to the default place.
3461 3. If the message "no IP address found for host xxxx" is generated during
3462 incoming verification, it is now followed by identification of the incoming
3463 connection (so you can more easily find what provoked it).
3465 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3467 5. Added some features to "harden" Exim a bit more against certain attacks:
3469 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3470 be put in Local/Makefile. This is like the never_users runtime option,
3471 but it cannot be overridden. The default setting is "root".
3473 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3474 prefix string with which any file named in a -C command line option
3477 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3478 is retained for -C and -D only if the caller of Exim is root. Without
3479 it, the exim user may also use -C and -D and retain privilege.
3481 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3482 command line option is disabled.
3484 6. Macro names set by the -D option must start with an upper case letter, just
3485 like macro names defined in the configuration file.
3487 7. Added "dereference=" facility to LDAP.
3489 8. Two instances of the typo "uknown" in the source files are fixed.
3491 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3492 the Configure-Makefile script screwed up while processing it.
3494 10. Incorporated PCRE 4.4.
3496 11. The SMTP synchronization check was not operating right at the start of an
3497 SMTP session. For example, it could not catch a HELO sent before the client
3498 waited for the greeting. There is now a check for outstanding input at the
3499 point when the greeting is written. Because of the duplex, asynchronous
3500 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3501 way, but not yet received, when the check is performed.
3503 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3504 on TCP/IP sockets, because this apparently causes some broken clients to
3507 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3508 unchanged) from the Cygwin maintainer.
3510 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3511 was supported instead of testing for mbx. Effectively a typo.
3513 15. The spa authenticator server code was not checking that the input it
3514 received was valid base64.
3516 16. The debug output line for the "set" modifier in ACLs was not showing the
3517 name of the variable that was being set.
3519 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3521 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3522 lookup on its own. Something else was needed to trigger the lookup. For
3523 example, a match in host_lookup or the need for a host name in a host list.
3524 Now, if $sender_host_name is referenced and the host name has not yet been
3525 looked up, a lookup is performed. If the lookup fails, the variable remains
3526 empty, and $host_lookup_failed is set to "1".
3528 19. Added "eqi" as a case-independent comparison operator.
3530 20. The saslauthd authentication condition could segfault if neither service
3531 nor realm was specified.
3533 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3534 error message that was logged was misleading, and incoming SMTP
3535 connections were dropped. The message is now more accurate, and temporary
3536 errors are given to SMTP connections.
3538 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3539 (except RSET) with a 421 error, until QUIT is received. However, it was
3540 failing to send a response to QUIT.
3542 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3543 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3544 ACL is now run after the helo_try_verify_hosts code.
3546 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3547 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3548 case-independent, so other case variants are also recognized). Apparently
3549 some systems use these upper case variants.
3551 25. If more than two messages were waiting for the same host, and a transport
3552 filter was specified for the transport, Exim sent two messages over the
3553 same TCP/IP connection, and then failed with "socket operation on non-
3554 socket" when it tried to send the third.
3556 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3558 27. The extern definition of crypt16() in expand.c was not being excluded when
3559 the OS had its own crypt16() function.
3561 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3562 as a preferred synonym for return_size_limit, both as an option and as an
3565 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3567 30. Changed the default configuration ACL to relax the local part checking rule
3568 for addresses that are not in any local domains. For these addresses,
3569 slashes and pipe symbols are allowed within local parts, but the sequence
3570 /../ is explicitly forbidden.
3572 31. SPA server authentication was not clearing the challenge buffer before
3575 32. log_message in a "warn" ACL statement was writing to the reject log as
3576 well as to the main log, which contradicts the documentation and doesn't
3577 seem right (because no rejection is happening). So I have stopped it.
3579 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3580 However, I am unable to do any testing of this.
3582 34. Fixed an infelicity in the appendfile transport. When checking directories
3583 for a mailbox, to see if any needed to be created, it was accidentally
3584 using path names with one or more superfluous leading slashes; tracing
3585 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3587 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3588 contents are added to the log line that is written for every discarded
3589 recipient. (Previously a log_message setting was ignored.)
3591 36. The ${quote: operator now quotes the string if it is empty.
3593 37. The install script runs exim in order to find its version number. If for
3594 some reason other than non-existence or emptiness, which it checks, it
3595 could not run './exim', it was installing it with an empty version number,
3596 i.e. as "exim-". This error state is now caught, and the installation is
3599 38. An argument was missing from the function that creates an error message
3600 when Exim fails to connect to the socket for saslauthd authentication.
3601 This could cause Exim to crash, or give a corrupted message.
3603 39. Added isip, isip4, and isip6 to ${if conditions.
3605 40. The ACL variables $acl_xx are now saved with the message, and can be
3606 accessed later in routers, transports, and filters.
3608 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3609 strings in the file are not string-expanded.
3611 42. If a MAIL command specified a SIZE value that was too large to fit into an
3612 int variable, the check against message_size_limit failed. Such values are
3613 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3614 day this will have to be increased, but I don't think I want to be around
3615 when emails are that large.
3622 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3623 iconv() is not standard in FreeBSD.
3625 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3626 IPv6 enabled. The observed symptom was a segmentation fault on return from
3627 the function os_common_find_running_interfaces() in src/os.c.
3629 3. In the check_special_case() function in daemon.c I had used "errno" as an
3630 argument name, which causes warnings on some systems. This was basically a
3631 typo, since it was named "eno" in the comments!
3633 4. The code that waits for the clock to tick (at a resolution of some fraction
3634 of a second) so as to ensure message-id uniqueness was always waiting for
3635 at least one whole tick, when it could have waited for less. [This is
3636 almost certainly not relevant at current processor speeds, where it is
3637 unlikely to ever wait at all. But we try to future-proof.]
3639 5. The function that sleeps for a time interval that includes fractions of a
3640 second contained a race. It did not block SIGALRM between setting the
3641 timer, and suspending (a couple of lines later). If the interval was short
3642 and the sigsuspend() was delayed until after it had expired, the suspension
3643 never ended. On busy systems this could lead to processes getting stuck for
3646 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3647 process, before it forks any delivery processes. The open lookup caching
3648 mechanism meant that the open file or database connection was passed into
3649 the delivery process. The problem was that delivery processes always tidy
3650 up cached lookup data. This could cause a problem for the next delivery
3651 process started by the queue runner, because the external queue runner
3652 process does not know about the closure. So the next delivery process
3653 still has data in the lookup cache. In the case of a file lookup, there was
3654 no problem because closing a file descriptor in a subprocess doesn't affect
3655 the parent. However, if the lookup was caching a connection to a database,
3656 the connection was closed, and the second delivery process was likely to
3657 see errors such as "PGSQL: query failed: server closed the connection
3658 unexpectedly". The problem has been fixed by closing all cached lookups
3659 in a queue runner before running a delivery process.
3661 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3662 seem to have the "const" qualifier which it has on other OS. I've
3665 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3666 *of the same type* (client or server) with the same public name that an
3667 error should be diagnosed.
3669 9. When Exim looked up a host name for an IP address, but failed to find the
3670 original IP address when looking up the host name (a safety check), it
3671 output the message "<ip address> does not match any IP for NULL", which was
3672 confusing, to say the least. The bug was that the host name should have
3673 appeared instead of "NULL".
3675 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3676 user that is built into the binary, and the -C or -D options is used, root
3677 privilege is dropped before the configuration file is read. In addition,
3678 logging is switched to stderr instead of the normal log files. If the
3679 configuration then re-defines the Exim user, the unprivileged environment
3680 is probably not what is expected, so Exim logs a panic warning message (but
3683 However, if deliver_drop_privilege is set, the unprivileged state may well
3684 be exactly what is intended, so the warning has been cut out in that case,
3685 and Exim is allowed to try to write to its normal log files.
3691 1. smtp_return_error_details was not giving details for temporary sender
3692 or receiver verification errors.
3694 2. Diagnose a configuration error if two authenticators have the same public
3697 3. Exim used not to create the message log file for a message until the first
3698 delivery attempt. This could be confusing when incoming messages were held
3699 for policy or load reasons. The message log file is now created at the time
3700 the message is received, and an initial "Received" line is written to it.
3702 4. The automatically generated man page for command line options had a minor
3703 bug that caused no ill effects; however, a more serious problem was that
3704 the procedure for building the man page automatically didn't always
3705 operate. Consequently, release 4.20 contains an out-of-date version. This
3706 shouldn't happen again.
3708 5. When building Exim with embedded Perl support, the script that builds the
3709 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3710 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3712 6. The freeze_tell option was not being used for messages that were frozen on
3713 arrival, either by an ACL or by local_scan().
3715 7. Added the smtp_incomplete_transaction log selector.
3717 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3718 was accepting AUTH without a new EHLO.
3720 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3721 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3724 10. Logging of TCP/IP connections (when configured) now happens in the main
3725 daemon process instead of the child process, so that the TCP/IP connection
3726 count is more accurate (but it can never be perfect).
3728 11. The use of "drop" in a nested ACL was not being handled correctly in the
3729 outer ACL. Now, if condition failure induced by the nested "drop" causes
3730 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3731 or "require"), the connection is dropped.
3733 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3734 that yield "discard" can now be used with an "accept" or a "discard" verb,
3735 but an error is generated for any others (because I can't see a useful way
3736 to define what should happen).
3738 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3739 are now processed in the same way as lines in the Exim configuration file.
3740 In particular, continuation lines are supported.
3742 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3744 15. Added -ti meaning -t -i.
3746 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3747 domains, and warn by logging if others are found.
3749 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3750 returned from GIFCONF when Exim is trying to find the list of interfaces on
3751 a host. The code in os.c has been modified to copy each ifreq to an aligned
3752 structure in all cases.
3754 Also, in some cases, the returned ifreq's were being copied to a 'struct
3755 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3756 means the last couple of bytes of an IPv6 address could be chopped if the
3757 ifreq contained only a normal sockaddr (14 bytes storage).
3759 18. Named domain lists were not supported in the hosts_treat_as_local option.
3760 An entry such as +xxxx was not recognized, and was treated as a literal
3763 19. Ensure that header lines added by a DATA ACL are included in the reject log
3764 if the ACL subsequently rejects the message.
3766 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3767 MD5 (which is deprecated).
3769 21. When testing a filter file using -bf, Exim was writing a message when it
3770 took the sender from a "From " line in the message, but it was not doing so
3771 when it took $return_path from a Return-Path: header line. It now does.
3773 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3774 with a valid header line field name (a series of printing characters
3775 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3777 23. Changed "disc" in the source to "disk" to conform to the documentation and
3778 the book and for uniformity.
3780 24. Ignore Sendmail's -Ooption=value command line item.
3782 25. When execve() failed while trying to run a command in a pipe transport,
3783 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3784 could be confused with a return value of 69 from the command itself. This
3785 has been changed to 127, the value the shell returns if it is asked to run
3786 a non-existent command. The wording for the related log line suggests a
3787 non-existent command as the problem.
3789 26. If received_header_text expands to an empty string, do not add a Received:
3790 header line to the message. (Well, it adds a token one on the spool, but
3791 marks it "old" so that it doesn't get used or transmitted.)
3793 27. Installed eximstats 1.28 (addition of -nt option).
3795 28. There was no check for failure on the call to getsockname() in the daemon
3796 code. This can fail if there is a shortage of resources on the system, with
3797 ENOMEM, for example. A temporary error is now given on failure.
3799 29. Contrary to the C standard, it seems that in some environments, the
3800 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3801 program. Exim now does this explicitly; it affects the formatting of
3802 timestamps using strftime().
3804 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3805 complaints. I've now initialized all the variables, to avoid this.
3807 32. Header lines added by a system filter were not being "seen" during
3808 transport-time rewrites.
3810 33. The info_callback() function passed to OpenSSL is set up with type void
3811 (*)(SSL *, int, int), as described somewhere. However, when calling the
3812 function (actually a macro) that sets it up, the type void(*)() is
3813 expected. I've put in a cast to prevent warnings from picky compilers.
3815 34. If a DNS black list lookup found a CNAME record, but there were no A
3816 records associated with the domain it pointed at, Exim crashed.
3818 35. If a DNS black list lookup returned more than one A record, Exim ignored
3819 all but the first. It now scans all returned addresses if a particular IP
3820 value is being sought. In this situation, the contents of the
3821 $dnslist_value variable are a list of all the addresses, separated by a
3824 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3825 to accept a host name and all its aliases if the forward lookup for any of
3826 them yielded the IP address of the incoming connection. Now it accepts only
3827 those names whose forward lookup yields the correct IP address. Any other
3828 names are discarded. This closes a loophole whereby a rogue DNS
3829 administrator could create reverse DNS records to break through a
3830 wildcarded host restriction in an ACL.
3832 37. If a user filter or a system filter that ran in a subprocess used any of
3833 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3834 the wrong values were passed to the pipe command ($thisaddress had the
3835 value of $0, $0 had the value of $1, etc). This bug was introduced by
3836 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3838 38. Improved the line breaking for long SMTP error messages from ACLs.
3839 Previously, if there was no break point between 40 and 75 characters, Exim
3840 left the rest of the message alone. Two changes have been made: (a) I've
3841 reduced the minimum length to 35 characters; (b) if it can't find a break
3842 point between 35 and 75 characters, it looks ahead and uses the first one
3843 that it finds. This may give the occasional overlong line, but at least the
3844 remaining text gets split now.
3846 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3847 file descriptors might be low, and that setting 1000 would always raise it.
3848 It turns out that in some environments, the limit is already over 1000 and
3849 that lowering it causes trouble. So now Exim takes care not to decrease it.
3851 40. When delivering a message, the value of $return_path is set to $sender_
3852 address at the start of routing (routers may change the value). By an
3853 oversight, this default was not being set up when an address was tested by
3854 -bt or -bv, which affected the outcome if any router or filter referred to
3857 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3858 log only when "message" or "log_message" are set. However, if one of the
3859 conditions was an address verification, or a call to a nested ACL, the
3860 messages generated by the underlying test were being passed through. This
3861 no longer happens. The underlying message is available in $acl_verify_
3862 message for both "message" and "log_message" expansions, so it can be
3863 passed through if needed.
3865 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3866 new expansion $bh_ to give the encoded byte string without charset
3867 translation. Translation happens only if iconv() is available; HAVE_ICONV
3868 indicates this at build time. HEADERS_CHARSET gives the charset to
3869 translate to; headers_charset can change it in the configuration, and
3870 "headers charset" can change it in an individual filter file.
3872 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3873 that creates RFC 2047 encoded "words" labels them as that charset instead
3874 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3875 expansion operator; (ii) when Exim creates a From: line for a local
3876 message; (iii) when a header line is rewritten to include a "phrase" part.
3878 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3879 buggy, causing it to skip the first lines of messages whose message ID
3880 ended in 'D'. This would not have bitten before Exim release 4.14, because
3881 message IDs were unlikely to end in 'D' before then. The effect was to have
3882 incorrect size information for certain domains.
3884 45. #include "config.h" was missing at the start of the crypt16.c module. This
3885 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3888 46. If there was a timeout during a "random" callout check, Exim treated it as
3889 a failure of the random address, and carried on sending RSET and the real
3890 address. If the delay was just some slowness somewhere, the response to the
3891 original RCPT would be taken as a response to RSET and so on, causing
3892 mayhem of various kinds.
3894 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3895 when I implemented it. It didn't allow for the fact that some option values
3896 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3897 the right test for positive values.
3899 48. Domain names in DNS records are case-independent. Exim always looks them up
3900 in lower case. Some resolvers return domain names in exactly the case they
3901 appear in the zone file, that is, they may contain uppercase letters. Not
3902 all resolvers do this - some return always lower case. Exim was treating a
3903 change of case by a resolver as a change of domain, similar to a widening
3904 of a domain abbreviation. This triggered its re-routing code and so it was
3905 trying to route what was effectively the same domain again. This normally
3906 caused routing to fail (because the router wouldn't handle the domain
3907 twice). Now Exim checks for this case specially, and just changes the
3908 casing of the domain that it ultimately uses when it transmits the message
3911 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3914 50. If a filter generated a file delivery with a non-absolute name (possible if
3915 no home directory exists for the router), the forbid_file option was not
3918 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3919 the existing equality matching.
3921 52. Exim was using ints instead of ino_t variables in some places where it was
3922 dealing with inode numbers.
3924 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3925 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3926 called TMPDIR, and if it finds it is different, it changes its value.
3928 54. The smtp_printf() function is now made available to local_scan() so
3929 additional output lines can be written before returning. There is also an
3930 smtp_fflush() function to enable the detection of a dropped connection.
3931 The variables smtp_input and smtp_batched_input are exported to
3934 55. Changed the default runtime configuration: the message "Unknown user"
3935 has been removed from the ACL, and instead placed on the localuser router,
3936 using the cannot_route_message feature. This means that any verification
3937 failures that generate their own messages won't get overridden. Similarly,
3938 the "Unrouteable address" message that was in the ACL for unverifiable
3939 relay addresses has also been removed.
3941 56. Added hosts_avoid_esmtp to the smtp transport.
3943 57. The exicyclog script was not checking for the esoteric option
3944 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3945 will work only if exicyclog is run under the appropriate euid.
3947 58. Following a discussion on the list, the rules by which Exim recognises line
3948 endings on incoming messages have been changed. The -dropcr and drop_cr
3949 options are now no-ops, retained only for backwards compatibility. The
3950 following line terminators are recognized: LF CRLF CR. However, special
3951 processing applies to CR:
3953 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3954 nor a local message in the state where . is a terminator.
3956 (ii) If a bare CR is encountered in a header line, an extra space is added
3957 after the line terminator so as not to end the header. The reasoning
3958 behind this is that bare CRs in header lines are most likely either
3959 to be mistakes, or people trying to play silly games.
3961 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3962 was being incorrectly given as 18 bytes larger than it should have been.
3963 This is a VOB (very old bug).
3965 60. This may never have affected anything current, but just in case it has:
3966 When the local host is found other than at the start of a list of hosts,
3967 the local host, those with the same MX, and any that follow, are discarded.
3968 When the list in question was part of a longer list of hosts, the following
3969 hosts (not currently being processed) were also being discarded. This no
3970 longer happens. I'm not sure if this situation could ever has previously
3973 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3976 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3977 header. This is something that is recommended in a new Internet Draft, and
3978 is something that is documented as being done by Sendmail. There are two
3979 possible values. For messages generated by the autoreply transport, Exim
3982 Auto-Submitted: auto-replied
3984 whereas for all other generated messages (e.g. bounces) it adds
3986 Auto-Submitted: auto-generated
3988 63. The "personal" condition in filters now includes a test for the
3989 Auto-Submitted: header. If it contains the string "auto-" the message it
3990 not considered personal.
3992 64. Added rcpt_include_affixes as a generic transport option.
3994 65. Added queue_only_override (default true).
3996 66. Added the syslog_duplication option.
3998 67. If what should have been the first header line of a message consisted of
3999 a space followed by a colon, Exim was mis-interpreting it as a header line.
4000 It isn't of course - it is syntactically invalid and should therefore be
4001 treated as the start of the message body. The misbehaviour could have
4002 caused a number of strange effects, including loss of data in subsequent
4003 header lines, and spool format errors.
4005 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4006 client host had authenticated. This control can now be exercised by an ACL
4007 for more flexibility.
4009 69. By default, callouts do not happen when testing with -bh. There is now a
4010 variant, -bhc, which does actually run the callout code, including
4011 consulting and updating the callout cache.
4013 70. Added support for saslauthd authentication, courtesy of Alexander
4016 71. If statvfs() failed on the spool or log directories while checking their
4017 size for availability, Exim confusingly gave the error "space shortage".
4018 Furthermore, in debugging mode it crashed with a floating point exception.
4019 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4020 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4021 serious problem, Exim now writes to the main and panic logs when this
4022 happens, with details of the failure. It then refuses to accept the
4023 incoming message, giving the message "spool directory problem" or "log
4024 directory problem" with a 421 code for SMTP messages.
4026 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4027 0, 1, and 2 exist, because some OS complain for execs without them (see
4028 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4029 descriptors. However, the code omitted to check that the open succeeded,
4030 causing mysterious errors if for some reason the permissions on /dev/null
4031 got screwed. Now Exim writes a message to the main and panic logs, and
4032 bombs out if it can't open /dev/null.
4034 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4035 interact so that it is all more flexible. It is supposed to remain
4036 backwards compatible. Also added extra_local_interfaces.
4038 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4039 to bomb out with an assertion failure - to the client this appears as a
4040 connection drop. This problem occurs in the part of the code that was taken
4041 from the Samba project. Fortunately, the assertion is in a very simple
4042 function, so I have fixed this by reproducing the function inline in the
4043 one place where it is called, and arranging for authentication to fail
4044 instead of killing the process with assert().
4046 75. The SPA client code was not working when the server requested OEM rather
4047 than Unicode encoding.
4049 76. Added code to make require_files with a specific uid setting more usable in
4050 the case where statting the file as root fails - usually a non-root-mounted
4051 NFS file system. When this happens and the failure is EACCES, Exim now
4052 forks a subprocess and does the per-uid checking as the relevant uid.
4054 77. Added process_log_path.
4056 78. If log_file_path was not explicitly set, a setting of check_log_space or
4057 check_log_inodes was ignored.
4059 79. If a space check for the spool or log partitions fails, the incident is now
4060 logged. Of course, in the latter case the data may get lost...
4062 80. Added the %p formatting code to string_format() so that it can be used to
4063 print addresses in debug_print(). Adjusted all the address printing in the
4064 debugging in store.c to use %p rather than %d.
4066 81. There was a concern that a line of code in smtp_in.c could overflow a
4067 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4068 initially expressed, the concern was not well-founded, because trailing
4069 spaces are removed early. However, if the trailing spaces were followed by
4070 a NULL, they did not get removed, so the overflow was possible. Two fixes
4073 (a) I re-wrote the offending code in a cleaner fashion.
4074 (b) If an incoming SMTP command contains a NULL character, it is rejected
4077 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4078 runs initgroups(), so that if the Exim user is in any additional groups,
4079 they will be used during message reception.
4085 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob