1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.376 2006/07/14 14:21:27 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.
110 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
111 This is apparently needed in addition to the PH/07 change above to avoid
112 any possible encoding problems.
118 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
119 other effects) broke the use of negated acl sub-conditions.
121 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
124 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
125 "Deny" causes Exim to reject the incoming connection with a 554 error.
126 Unfortunately, if there is a major crisis, such as a disk failure,
127 tcp-wrappers gives "deny", whereas what one would like would be some
128 kind of temporary error. A kludge has been added to help with this.
129 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
130 554 error is used if errno is still zero or contains ENOENT (which occurs
131 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
134 PH/03 Add -lutil to the default FreeBSD LIBS setting.
136 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
137 errors. Otherwise a message that provokes a temporary error (when other
138 messages do not) can cause a whole host to time out.
140 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
141 addresses were routed directly to files or pipes from a redirect router.
142 File deliveries just didn't batch; pipe deliveries might have suffered
145 PH/06 A failure to get a lock for a hints database would erroneously always say
146 "Failed to get write lock", even when it was really a read lock.
148 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
149 of 0600. This has been changed to use the value of the lockfile_mode
150 option (which defaults to 0600).
152 PH/08 Applied small patch from the Sieve maintainer.
154 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
155 folder from quota calculations, a direct delivery into this folder messed
156 up the contents of the maildirsize file. This was because the regex was
157 used only to exclude .Trash (or whatever) when the size of the mailbox
158 was calculated. There was no check that a delivery was happening into an
159 excluded directory. This bug has been fixed by ignoring all quota
160 processing for deliveries into excluded directories.
162 PH/10 Added the maildirfolder_create_regex option to appendfile.
168 PH/01 The code for finding all the local interface addresses on a FreeBSD
169 system running IPv6 was broken. This may well have applied to all BSD
170 systems, as well as to others that have similar system calls. The broken
171 code found IPv4 interfaces correctly, but gave incorrect values for the
172 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
173 that it would not match correctly against @[] and not recognize the IPv6
176 PH/02 The ipliteral router was not recognizing addresses of the form user@
177 [ipv6:....] because it didn't know about the "ipv6:" prefix.
179 PH/03 Added disable_ipv6.
181 PH/04 Changed $reply_address to use the raw form of the headers instead of the
182 decoded form, because it is most often used to construct To: headers
183 lines in autoreplies, and the decoded form may well be syntactically
184 invalid. However, $reply_address has leading white space removed, and all
185 newlines turned into spaces so that the autoreply transport does not
188 PH/05 If group was specified without a user on a router, and no group or user
189 was specified on a transport, the group from the router was ignored.
191 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
192 for visible compile-time settings that can be used to change these
193 numbers, for those that want even more. Backwards compatibility with old
194 spool files has been maintained. However, going back to a previous Exim
195 release will lost any variables that are in spool files.
197 PH/07 Two small changes when running in the test harness: increase delay when
198 passing a TCP/IP connection to a new process, in case the original
199 process has to generate a bounce, and remove special handling of
200 127.0.0.2 (sic), which is no longer necessary.
202 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
203 be the same on different OS.
205 PH/09 Moved a debug statement in filter processing to avoid a race problem when
208 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
209 whether --show-vars was specified or not
211 JJ/02 exipick: Added support for new ACL variable spool format introduced
214 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
215 syntactically invalid From: or Reply-to: line, and a filter used this to
216 generate an autoreply, and therefore failed to obtain an address for the
217 autoreply, Exim could try to deliver to a non-existent relative file
218 name, causing unrelated and misleading errors. What now happens is that
219 it logs this as a hard delivery error, but does not attempt to create a
222 PH/11 The exinext utility has a -C option for testing purposes, but although
223 the given file was scanned by exinext itself; it wasn't being passed on
224 when Exim was called.
226 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
227 an end-of-file indication when reading a command response.
229 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
230 compiled. In many other places in Exim, IPv6 addresses are always
231 recognized, so I have changed this. It also means that IPv4 domain
232 literals of the form [IPV4:n.n.n.n] are now always recognized.
234 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
235 used if the router is not running as root, for example, when verifying at
236 ACL time, or when using -bh. The debugging output from this situation was
237 non-existent - all you got was a failure to exec. I have made two
240 (a) Failures to set uid/gid, the current directory, or a process leader
241 in a subprocess such as that created by queryprogram now generate
242 suitable debugging ouput when -d is set.
244 (b) The queryprogram router detects when it is not running as root,
245 outputs suitable debugging information if -d is set, and then runs
246 the subprocess without attempting to change uid/gid.
248 PH/15 Minor change to Makefile for building test_host (undocumented testing
251 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
252 additional section of a DNS packet that returns MX or SRV records.
253 Instead, it always explicitly searches for A/AAAA records. This avoids
254 major problems that occur when a DNS server includes only records of one
255 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
256 fixed another bug: if SRV records were looked up and the corresponding
257 address records were *not* found in the additional section, the port
258 values from the SRV records were lost.
260 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
261 using the correct key (the original address) when searching the retry
262 rules in order to find which one to use for generating the retry hint.
264 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
265 adding the default one. Similarly, if it contains a Reply-To: header, the
266 errors_reply_to option, if set, is not used.
268 PH/19 When calculating a retry time, Exim used to measure the "time since
269 failure" by looking at the "first failed" field in the retry record. Now
270 it does not use this if it is later than than the arrival time of the
271 message. Instead it uses the arrival time. This makes for better
272 behaviour in cases where some deliveries succeed, thus re-setting the
273 "first failed" field. An example is a quota failure for a huge message
274 when small messages continue to be delivered. Without this change, the
275 "time since failure" will always be short, possible causing more frequent
276 delivery attempts for the huge message than are intended.
277 [Note: This change was subsequently modified - see PH/04 for 4.62.]
279 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
280 $1, $2, $3) because the numerical variables can be reset during some
281 expansion items (e.g. "match"), thereby losing the authentication data.
283 PH/21 Make -bV show the size of off_t variables so that the test suite can
284 decide whether to run tests for quotas > 2G.
286 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
287 mailbox_size, and mailbox_filecount in the appendfile transport. If a
288 filecount value is greater than 2G or if a quota value is greater than 2G
289 on a system where the size of off_t is not greater than 4, a panic error
292 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
293 never match. The debug and -bh output now contains an explicit error
294 message indicating a malformed IPv4 address or mask.
296 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
297 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
300 PH/25 Do not write to syslog when running in the test harness. The only
301 occasion when this arises is a failure to open the main or panic logs
302 (for which there is an explicit test).
304 PH/26 Added the /no_tell option to "control=freeze".
306 PH/27 If a host name lookup failed very early in a connection, for example, if
307 the IP address matched host_lookup and the reverse lookup yielded a name
308 that did not have a forward lookup, an error message of the form "no IP
309 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
310 could be logged. Now it outputs the IP address instead of "NULL".
312 PH/28 An enabling patch from MH: add new function child_open_exim2() which
313 allows the sender and the authenticated sender to be set when
314 submitting a message from within Exim. Since child_open_exim() is
315 documented for local_scan(), the new function should be too.
317 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
318 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
319 results in an empty string is now treated as unset.
321 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
323 PH/31 Added sender_verify_fail logging option.
325 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
326 needed by RFC 822 but not by RFC 2822 was commented out. I have now
327 tidied the source and removed it altogether.
329 PH/33 When a queue run was abandoned because the load average was too high, a
330 log line was always written; now it is written only if the queue_run log
331 selector is set. In addition, the log line for abandonment now contains
332 information about the queue run such as the pid. This is always present
333 in "start" and "stop" lines but was omitted from the "abandon" line.
335 PH/34 Omit spaces between a header name and the colon in the error message that
336 is given when verify = headers_syntax fails (if there are lots of them,
337 the message gets confusing).
339 PH/35 Change the default for dns_check_names_pattern to allow slashes within
340 names, as there are now some PTR records that contain slashes. This check
341 is only to protect against broken name servers that fall over on strange
342 characters, so the fact that it applies to all lookups doesn't matter.
344 PH/36 Now that the new test suite is complete, we can remove some of the
345 special code in Exim that was needed for the old test suite. For example,
346 sorting DNS records because real resolvers return them in an arbitrary
347 order. The new test suite's fake resolver always returns records in the
350 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
351 bounces) except when queue_only is set, to avoid logging races between
352 the different processes.
354 PH/38 Panic-die if .include specifies a non-absolute path.
356 PH/39 A tweak to the "H" retry rule from its user.
358 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
359 a label. They prevented compilation on older perls.
361 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
362 a warning to be raised on newish perls.
364 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
365 on queue. Changes to match documented behaviour of showing count of
366 messages matching specified criteria.
368 PH/40 Changed the default ident timeout from 30s to 5s.
370 PH/41 Added support for the use of login_cap features, on those BSD systems
371 that have them, for controlling the resources used by pipe deliveries.
373 PH/42 The content-scanning code uses fopen() to create files in which to put
374 message data. Previously it was not paying any attention to the mode of
375 the files. Exim runs with umask(0) because the rest of the code creates
376 files with open(), and sets the required mode explicitly. Thus, these
377 files were ending up world-writeable. This was not a big issue, because,
378 being within the spool directory, they were not world-accessible. I have
379 created a function called modefopen, which takes an additional mode
380 argument. It sets umask(777), creates the file, chmods it to the required
381 mode, then resets the umask. All the relevant calls to fopen() in the
382 content scanning code have been changed to use this function.
384 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
385 to 24 hours. This avoids potential overflow problems when processing G
386 and H retry rules. I suspect nobody ever tinkers with this value.
388 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
390 PH/45 When the plaintext authenticator is running as a client, the server's
391 challenges are checked to ensure they are valid base64 strings. By
392 default, the authentication attempt is cancelled if an invalid string is
393 received. Setting client_ignore_invalid_base64 true ignores these errors.
394 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
395 they are received. Thus, the responses can be made to depend on the
396 challenges. If an invalid string is ignored, an empty string is placed in
399 PH/46 Messages that are created by the autoreply transport now contains a
400 References: header, in accordance with RFCs 2822 and 3834.
402 PH/47 Added authenticated_sender_force to the smtp transport.
404 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
406 PH/49 Installed latest patch from the Sieve maintainer.
408 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
409 was also set, the appendfile transport was unnecessarily scanning a
410 directory of message files (e.g. for maildir delivery) to find the count
411 of files (along with the size), even though it did not need this
412 information. It now does the scan only if it needs to find either the
413 size of the count of files.
415 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
417 PH/52 Two bugs concerned with error handling when the smtp transport is
420 (i) Exim was not creating retry information for temporary errors given
421 for individual recipients after the DATA command when the smtp transport
422 was used in LMTP mode. This meant that they could be retried too
423 frequently, and not timed out correctly.
425 (ii) Exim was setting the flag that allows error details to be returned
426 for LMTP errors on RCPT commands, but not for LMTP errors for individual
427 recipients that were returned after the DATA command.
429 PH/53 This is related to PH/52, but is more general: for any failing address,
430 when detailed error information was permitted to be returned to the
431 sender, but the error was temporary, then after the final timeout, only
432 "retry timeout exceeded" was returned. Now it returns the full error as
433 well as "retry timeout exceeded".
435 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
436 do this, and (what is worse) MTAs that accept it.
438 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
439 will now be deprecated.
441 PH/56 New os.c-cygwin from the Cygwin maintainer.
443 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
444 formats (previously only available in exim formats via -bpr, -bpru,
445 and -bpra. Now also available in native and exiqgrep formats)
447 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
448 with very large, slow to parse queues
450 JJ/08 exipick: added ! as generic prefix to negate any criteria format
452 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
454 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
455 responses to authentication challenges, though it was showing the
456 challenges; (ii) I've removed the CR characters from the debug output for
459 PH/58 Allow for the insertion of a newline as well as a space when a string
460 is turned into more than one encoded-word during RFC 2047 encoding. The
461 Sieve code now uses this.
463 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
464 data_4xx, lost_connection, tls_required.
466 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
467 message was being sent as an SMTP response.
469 PH/61 Add -l and -k options to exicyclog.
471 PH/62 When verifying, if an address was redirected to one new address, so that
472 verification continued, and the new address failed or deferred after
473 having set something in $address_data, the value of $address_data was not
474 passed back to the ACL. This was different to the case when no
475 redirection occurred. The value is now passed back in both cases.
477 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
478 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
479 use login_cap.h, so on its own it isn't the distinguishing feature. The
480 new name refers directly to the setclassresources() function.
482 PH/65 Added configuration files for NetBSD3.
484 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
486 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
487 is preferred over IPv4.
489 PH/68 The bounce_return_message and bounce_return_body options were not being
490 honoured for bounces generated during the reception of non-SMTP messages.
491 In particular, this applied to messages rejected by the ACL. This bug has
492 been fixed. However, if bounce_return_message is true and bounce_return_
493 body is false, the headers that are returned for a non-SMTP message
494 include only those that have been read before the error was detected.
495 (In the case of an ACL rejection, they have all been read.)
497 PH/69 The HTML version of the specification is now built in a directory called
498 spec_html instead of spec.html, because the latter looks like a path with
499 a MIME-type, and this confuses some software.
501 PH/70 Catch two compiler warnings in sieve.c.
503 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
504 function verify_get_ident() calls ip_connect() to connect a socket, but
505 if the "connect()" function timed out, ip_connect() used to close the
506 socket. However, verify_get_ident() also closes the socket later, and in
507 between Exim writes to the log, which may get opened at this point. When
508 the socket was closed in ip_connect(), the log could get the same file
509 descriptor number as the socket. This naturally causes chaos. The fix is
510 not to close the socket in ip_connect(); the socket should be closed by
511 the function that creates it. There was only one place in the code where
512 this was missing, in the iplookup router, which I don't think anybody now
513 uses, but I've fixed it anyway.
515 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
516 well as to direct DNS lookups. Otherwise the handling of names in host
517 lists is inconsistent and therefore confusing.
523 PH/01 Two changes to the default runtime configuration:
525 (1) Move the checks for relay_from_hosts and authenticated clients from
526 after to before the (commented out) DNS black list checks.
528 (2) Add control=submission to the relay_from_hosts and authenticated
529 clients checks, on the grounds that messages accepted by these
530 statements are most likely to be submissions.
532 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
534 (1) Generate an error if the third argument for the ${prvs expansion is
537 (2) Treat a missing third argument of ${prvscheck as if it were an empty
540 (3) Reset the variables that are obtained from the first argument of
541 ${prvscheck and used in the second argument before leaving the code,
542 because their memory is reclaimed, so using them afterwards may do
545 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
546 one (it's much easier than Tom thought :-).
548 (5) Because of (4), we can now allow for the use of $prvscheck_result
549 inside the third argument.
551 PH/03 For some reason, the default setting of PATH when running a command from
552 a pipe transport was just "/usr/bin". I have changed it to
555 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
556 anything to be listed in the output from -bV.
558 PH/05 When a filter generated an autoreply, the entire To: header line was
559 quoted in the delivery log line, like this:
561 => >A.N.Other <ano@some.domain> <original@ddress> ...
563 This has been changed so that it extracts the operative address. There
564 may be more than one such address. If so, they are comma-separated, like
567 => >ano@some.domain,ona@other.domain <original@ddress> ...
569 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
570 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
571 not being looked up in the rDNS to get a host name, Exim was showing the
572 IP address twice in Received: lines, even though the IP addresses were
573 identical. For example:
575 Received: from [1.2.3.4] (helo=[1.2.3.4])
577 However, if the real host name was known, it was omitting the HELO data
578 if it matched the actual IP address. This has been tidied up so that it
579 doesn't show the same IP address twice.
581 PH/07 When both +timestamp and +memory debugging was on, the value given by
582 $tod_xxx expansions could be wrong, because the tod_stamp() function was
583 called by the debug printing, thereby overwriting the timestamp buffer.
584 Debugging no longer uses the tod_stamp() function when +timestamp is set.
586 PH/08 When the original message was included in an autoreply transport, it
587 always said "this is a copy of the message, including all the headers",
588 even if body_only or headers_only was set. It now gives an appropriate
591 PH/09 Applied a patch from the Sieve maintainer which:
593 o fixes some comments
594 o adds the (disabled) notify extension core
595 o adds some debug output for the result of if/elsif tests
596 o points to the current vacation draft in the documentation
597 and documents the missing references header update
601 o fixes a bug in processing the envelope test (when testing
602 multiple envelope elements, the last element determinted the
605 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
606 Electronic Mail") by including:
608 Auto-submitted: auto-generated
610 in the messages that it generates (bounce messages and others, such as
611 warnings). In the case of bounce messages for non-SMTP mesages, there was
612 also a typo: it was using "Auto_submitted" (underscore instead of
613 hyphen). Since every message generated by Exim is necessarily in response
614 to another message, thes have all been changed to:
616 Auto-Submitted: auto-replied
618 in accordance with these statements in the RFC:
620 The auto-replied keyword:
622 - SHOULD be used on messages sent in direct response to another
623 message by an automatic process,
625 - MUST NOT be used on manually-generated messages,
627 - MAY be used on Delivery Status Notifications (DSNs) and Message
628 Disposition Notifications (MDNs),
630 - MUST NOT be used on messages generated by automatic or periodic
631 processes, except for messages which are automatic responses to
634 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
635 to the default Received: header definition.
637 PH/12 Added log selector acl_warn_skipped (default on).
639 PH/13 After a successful wildlsearch lookup, discard the values of numeric
640 variables because (a) they are in the wrong storage pool and (b) even if
641 they were copied, it wouldn't work properly because of the caching.
643 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
644 checking when decoding. Apparently there are clients that generate
645 overlong encoded strings. Why am I not surprised?
647 PH/15 If the first argument of "${if match_address" was not empty, but did not
648 contain an "@" character, Exim crashed. Now it writes a panic log message
649 and treats the condition as false.
651 PH/16 In autoreply, treat an empty string for "once" the same as unset.
653 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
654 extension "envelope-auth". The code is finished and in agreement with
655 other implementations, but there is no documentation so far and in fact,
656 nobody wrote the draft yet. This extension is currently #undef'ed, thus
657 not changing the active code.
659 Print executed "if" and "elsif" statements when debugging is used. This
660 helps a great deal to understand what a filter does.
662 Document more things not specified clearly in RFC3028. I had all this
663 sorted out, when out of a sudden new issues came to my mind. Oops."
665 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
668 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
669 canonical form (as documented). However, after a host name lookup from
670 the IP address, check_host() was doing a simple string comparison with
671 addresses acquired from the DNS when checking that the found name did
672 have the original IP as one of its addresses. Since any found IPv6
673 addresses are likely to be in abbreviated form, the comparison could
674 fail. Luckily, there already exists a function for doing the comparison
675 by converting both addresses to binary, so now that is used instead of
678 PH/20 There was another similar case to PH/19, when a complete host name was
679 given in a host list; looking up its IP address could give an abbreviated
680 form, whereas the current host's name might or might not be abbreviated.
681 The same fix has been applied.
687 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
688 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
691 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
692 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
694 PH/03 Typo: missing ".o" in src/pcre/Makefile.
696 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
697 header line, restrict the check to what is listed in RFCs 2369 and 2929.
698 Also, for "Auto-Submitted", treat anything other than "no" as
699 non-personal, in accordance with RFC 3834. (Previously it treated
700 anything starting "auto-" as non-personal.)
702 TF/01 The control=submission/name=... option had a problem with syntax
703 errors if the name included a slash character. The /name= option
704 now slurps the rest of the string, so it can include any characters
705 but it must come last in the list of options (after /sender_retain
708 PH/05 Some modifications to the interface to the fake nameserver for the new
716 TK/01 Added the "success_on_redirect" address verification option. See
717 NewStuff for rationale and an example.
719 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
721 PH/02 Patch to exigrep to allow it to work on syslog lines.
723 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
724 fread() to skip over the body file's header line, because in Cygwin the
725 header line is locked and is inaccessible.
727 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
728 co-exist for some time) to make it clear that it is the Exim ID that is
729 referenced, not the Message-ID: header line.
731 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
732 string_format() function, because snprintf() does not exist on all
735 PH/06 The use of forbid_filter_existstest now also locks out the use of the
736 ${stat: expansion item.
738 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
739 protocol synchronization error", to keep the pedants happy.
741 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
742 well as for IRIX systems, when gcc is being used. See the host.c source
745 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
747 PH/10 Named domain lists were not working if used in a queue_smtp_domains
750 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
751 transport and to the smtp transport in LMTP mode.
753 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
755 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
756 run a filter in a subprocess. This could lead to confusion in subsequent
757 lookups in the parent process. There should also be a search_tidyup() at
758 the end of the subprocess.
760 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
761 only if the host matched helo_try_verify_hosts, which caused the
762 verification to occur when the EHLO/HELO command was issued. The ACL just
763 tested the remembered result. Now, if a previous verification attempt has
764 not happened, "verify = helo" does it there and then.
766 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
768 TK/03 Fix log output including CR from clamd.
770 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
771 exist provoked a memory error which could cause a segfault.
773 PH/15 Installed PCRE 6.2
775 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
777 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
778 of the problem. Specifically, suggested +O2 rather than +O1 for the
781 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
783 PH/20 If a delivery was routed to a non-standard port by means of an SRV
784 record, the port was not correctly logged when the outgoing_port log
785 selector was set (it logged the transort's default port).
787 PH/21 Added support for host-specific ports to manualroute, queryprogram,
788 fallback_hosts, and "hosts" in the smtp transport.
790 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
791 host errors such as "Connection refused".
793 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
794 authentication with radiusclient 0.4.9:
796 - Error returned from rc_read_config was caught wrongly
797 - Username/password not passed on to radius server due to wrong length.
799 The presumption is that some radiusclient API changes for 4.51/PH/17
800 were not taken care of correctly. The code is still untested by me (my
801 Linux distribution still has 0.3.2 of radiusclient), but it was
802 contributed by a Radius user.
804 PH/24 When doing a callout, the value of $domain wasn't set correctly when
805 expanding the "port" option of the smtp transport.
807 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
808 while reading a MIME header. Thanks to Tom Hughes for a patch.
810 PH/24 Include config.h inside local_scan.h so that configuration settings are
813 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
814 that in an ACL for RCPT (for example), you can examine exactly what was
817 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
818 commands, but it was not correctly comparing the address with the actual
819 client host address. Thus, it would show the EHLO address in Received:
820 header lines when this was not necessary.
822 PH/27 Added the % operator to ${eval:}.
824 PH/28 Exim tries to create and chdir to its spool directory when it starts;
825 it should be ignoring failures (because with -C, for example, it has lost
826 privilege). It wasn't ignoring creation failures other than "already
829 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
832 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
833 because an input line was too long, either on its own, or by virtue of
834 too many continuations, the temporary file was not being removed, and the
835 return code was incorrect.
837 PH/31 Missing "BOOL" in function definition in filtertest.c.
839 PH/32 Applied Sieve patches from the maintainer.
841 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
843 PH/33 Added "verify = not_blind".
845 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
846 Local/Makefile (with some defaults set). These are used in built scripts
847 such as exicyclog, but they have never been used in the exim_install
848 script (though there are many overriding facilities there). I have
849 arranged that the exim_install script now takes note of these two
852 PH/35 Installed configuration files for Dragonfly.
854 PH/36 When a locally submitted message by a trusted user did not contain a
855 From: header, and the sender address was obtained from -f or from an SMTP
856 MAIL command, and the trusted user did not use -F to supply a sender
857 name, $originator_name was incorrectly used when constructing a From:
858 header. Furthermore, $originator_name was used for submission mode
859 messages from external hosts without From: headers in a similar way,
860 which is clearly wrong.
862 PH/37 Added control=suppress_local_fixups.
864 PH/38 When log_selector = +received_sender was set, and the addition of the
865 sender made the log line's construction buffer exactly full, or one byte
866 less than full, an overflow happened when the terminating "\n" was
869 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
870 when the result of a list match is failure because a DNS lookup failed.
872 PH/40 RM_COMMAND is now used in the building process.
874 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
875 the "build-* directories that it finds.
877 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
878 address in a domain literal was a prefix of an interface address.
880 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
881 when verifying a sender address, unless rewrite_headers is false.
883 PH/44 Wrote a long comment about why errors_to addresses are verified as
884 recipients, not senders.
886 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
887 the ratelimit ACL was added.
889 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
891 PH/46 Added extra information about PostgreSQL errors to the error string.
893 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
894 avoiding the need to install special zones in a real server. This is
895 backwards compatible; if it can't find the fake resolver, it drops back.
896 Thus, both old and new test suites can be run.
898 TF/02 Added util/ratelimit.pl
900 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
901 clock is set back in time.
903 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
904 Candler <B.Candler@pobox.com>.
906 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
907 OK for addresses that are the result of redirections.
909 PH/48 A number of further additions for the benefit of the new test suite,
910 including a fake gethostbyname() that interfaces to the fake DNS resolver
913 TF/06 The fix for widen_domains has also been applied to qualify_single and
914 search_parents which are the other dnslookup options that can cause
917 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
920 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
922 TF/07 Exim produced the error message "an SRV record indicated no SMTP
923 service" if it encountered an MX record with an empty target hostname.
924 The message is now "an MX or SRV record indicated no SMTP service".
926 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
927 if the DNS of the sending site is misconfigured. This is quite a
928 common situation. This change restores the behaviour of treating a
929 helo verification defer as a failure.
931 PH/51 If self=fail was set on a router, the bounce message did not include the
932 actual error message.
938 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
940 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
941 command itself ended in error, the underlying message about the transport
942 filter timeout was being overwritten with the pipe command error. Now the
943 underlying error message should be appended to the second error message.
945 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
947 PH/02 Reduce the amount of output that "make" produces by default. Full output
948 can still be requested.
950 PH/03 The warning log line about a condition test deferring for a "warn" verb
951 was being output only once per connection, rather than after each
952 occurrence (because it was using the same function as for successful
953 "warn" verbs). This seems wrong, so I have changed it.
955 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
956 it should not have, which might have caused a crash in the right
957 circumstances, but probably never did.
959 PH/04 Installed a modified version of Tony Finch's patch to make submission
960 mode fix the return path as well as the Sender: header line, and to
961 add a /name= option so that you can make the user's friendly name appear
964 TF/03 Added the control = fakedefer ACL modifier.
966 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
967 Mark Lowes for thorough testing.
969 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
971 TK/03 Merged latest SRS patch from Miles Wilton.
973 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
974 with the definition in sysexits.h (which is #included earlier).
975 Fortunately, Exim does not actually use EX_OK. The code used to try to
976 preserve the sysexits.h value, by assumimg that macro definitions were
977 scanned for macro replacements. I have been disabused of this notion,
978 so now the code just undefines EX_OK before #including unistd.h.
980 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
981 in the smtp transport. When a block could not be written in a single
982 write() function, the timeout was being re-applied to each part-write.
983 This seems wrong - if the receiver was accepting one byte at a time it
984 would take for ever. The timeout is now adjusted when this happens. It
985 doesn't have to be particularly precise.
987 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
988 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
990 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
991 without a domain if the check to <postmaster@domain> fails.
993 SC/01 Eximstats: added -xls and the ability to specify output files
994 (patch written by Frank Heydlauf).
996 SC/02 Eximstats: use FileHandles for outputing results.
998 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1000 SC/04 Eximstats: fixed display of large numbers with -nvr option
1002 SC/05 Eximstats: fixed merging of reports with empty tables.
1004 SC/06 Eximstats: added the -include_original_destination flag
1006 SC/07 Eximstats: removed tabs and trailing whitespace.
1008 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1010 TK/06 MBOX spool code: Add real "From " MBOX separator line
1011 so the .eml file is really in mbox format (even though
1012 most programs do not really care). Patch from Alex Miller.
1014 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1015 The latter is generated from $received_to and is only set if the
1016 message has one envelope recipient. SA can use these headers,
1017 obviously out-of-the-box. Patch from Alex Miller.
1019 PH/08 The ${def test on a variable was returning false if the variable's
1020 value was "0", contrary to what the specification has always said!
1021 The result should be true unless the variable is empty.
1023 PH/09 The syntax error of a character other than { following "${if
1024 def:variable_name" (after optional whitespace) was not being diagnosed.
1025 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1026 accidental colon was present, for example, could give incorrect results.
1028 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1029 result is used (not including appendfile, where other changes are about
1032 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1033 This involved changing a lot of size variables from int to off_t. It
1034 should work with maildirs and everything.
1036 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1037 spamd dying while we are connected to it.
1039 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1042 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1043 function for BDB 4.3.
1045 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1047 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1048 the sources for PCRE have been re-arranged and more files are now
1051 PH/15 The code I had for printing potentially long long variables in PH/11
1052 above was not the best (it lost precision). The length of off_t variables
1053 is now inspected at build time, and an appropriate printing format (%ld
1054 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1055 to be "long long int" or "long int". This is needed for the internal
1056 formatting function string_vformat().
1058 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1059 the configuration file to be ":syslog", then the script "guesses" where
1060 the logs files are, rather than using the compiled in default. In our
1061 case the guess is not the same as the compiled default, so the script
1062 suddenly stopped working when I started to use syslog. The patch checks
1063 to see if log_file_path is "". If so, it attempts to read it from exim
1064 with no configuration file to get the compiled in version, before it
1065 falls back to the previous guessing code."
1067 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1068 implementing BATV in an Exim configuration. See NewStuff for the gory
1071 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1072 Makefile that are specific to HP-UX.
1074 PH/18 If the "use_postmaster" option was set for a recipient callout together
1075 with the "random" option, the postmaster address was used as the MAIL
1076 FROM address for the random test, but not for the subsequent recipient
1077 test. It is now used for both.
1079 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1080 patch removes a few documentation additions to RFC 3028, because the
1081 latest draft now contains them. It adds the new en;ascii-case comparator
1082 and a new error check for 8bit text in MIME parts. Comparator and
1083 require names are now matched exactly. I enabled the subaddress
1084 extension, but it is not well tested yet (read: it works for me)."
1086 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1087 rework some of the code of TK/09 above to avoid the hardwired use of
1088 "%lld" and "long long". Replaced the call to snprintf() with a call to
1091 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1092 records point to non-existent hosts", "retry timeout exceeded", and
1093 "retry time not reached for any host after a long failure period".
1095 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1096 experimental DomainKeys support:
1098 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1099 (2) On an error such as an illegally used "control", the wrong name for
1100 the control was given.
1102 These problems did NOT occur unless DomainKeys support was compiled.
1104 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1106 PH/24 Added ${if match_ip condition.
1108 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1109 databases so that it will be absolutely obvious if a crash occurs in the
1110 DB library. This is a regular occurrence (often caused by mis-matched
1113 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1114 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1115 picked up on a user's system that detects such things. There doesn't seem
1116 to be a gcc warning option for this - only an attribute that has to be
1117 put on the function's prototype. It seems that in Fedora Core 4 they have
1118 set this on a number of new functions. No doubt there will be more in due
1121 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1122 specify a transport. However, if an address that was verified by such a
1123 router was the subject of a callout, Exim crashed because it tried to
1124 read the rcpt_include_affixes from the non-existent transport. Now it
1125 just assumes that the setting of that option is false. This bug was
1126 introduced by 4.51/PH/31.
1128 PH/28 Changed -d+all to exclude +memory, because that information is very
1129 rarely of interest, but it makes the output a lot bigger. People tend to
1130 do -d+all out of habit.
1132 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1133 code in os-type was giving problems when libc.so lives in lib64, like on
1136 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1137 aren't the modern standard, and it seems that some systems' include files
1138 don't always have them. Exim was already checking for some of the newer
1139 ones like T_AAAA, and defining it itself. I've added checks for all the
1140 record types that Exim uses.
1142 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1143 not automatically generating a new one, as it is supposed to. This
1144 prevented TLS from working. If the file did exist, but contained invalid
1145 data, a new version was generated, as expected. It was only the case of a
1146 non-existent file that was broken.
1148 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1149 with a change in libdomainkeys > 0.64.
1151 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1152 from DNS. If the selector record carries the flag, it now has
1153 precedence over the domain-wide flag.
1155 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1157 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1158 the use of a port name that isn't defined in /etc/services) occurred, the
1159 message was deferred as in a normal delivery, and thus remained on the
1160 spool, instead of being failed because of the mua_wrapper setting. This
1161 is now fixed, and I tidied up some of the mua_wrapper messages at the
1164 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1165 the messages in a hash of arrays rather than using individual hashes.
1166 This is a bit cleaner and results in dramatic memory savings, albeit
1167 at a slight CPU cost.
1169 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1170 as requested by Marc Sherman.
1172 SC/10 Eximstats: added histograms for user specified patterns as requested
1175 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1177 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1178 fopen() in the content-scanning modules that did not already have it.
1184 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1185 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1187 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1189 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1191 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1192 to test Sieve filters that use "vacation".
1194 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1195 that changes the way the GnuTLS parameters are stored in the cache file.
1196 The new format can be generated externally. For backward compatibility,
1197 if the data in the cache doesn't make sense, Exim assumes it has read an
1198 old-format file, and it generates new data and writes a new file. This
1199 means that you can't go back to an older release without removing the
1202 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1203 work if there are any delivery delays because "one_time" forces the
1204 parent to be marked "delivered", so its unseen clone is never tried
1205 again. For this reason, Exim now forbids the simultaneous setting of
1208 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1209 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1210 entry if you want to know the details. The fix, however, neglected to
1211 consider the case when local delivery batching is involved. The test for
1212 "previously delivered" was not happening when checking to see if an
1213 address could be batched with a previous (undelivered) one; under
1214 certain circumstances this could lead to multiple deliveries to the same
1217 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1218 in its include files, and this causes problems building Exim.
1220 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1221 header_syntax) but Exim was just ignoring anything given after a slash.
1222 In particular, this caused confusion with an attempt to use "verify =
1223 reverse_host_lookup/defer_ok". An error is now given when options are
1224 supplied for verify items that do not have them. (Maybe reverse_host_
1225 lookup should have a defer_ok option, but that's a different point.)
1227 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1228 defined by RFC 821) to 2048, because there were problems with some AUTH
1229 commands, and RFC 1869 says the size should be increased for extended
1230 SMTP commands that take arguments.
1232 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1235 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1236 "unknown" error; now it says that the functionality isn't in the binary.
1238 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1239 an address' error message when a string expansion fails (syntax or
1240 whatever). Otherwise the password may appear in the log. Following change
1241 PH/42 below, there is no longer a chance of it appearing in a bounce
1244 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1246 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1247 of its addresses was ever tried. (Bugzilla bug #2.)
1249 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1250 the result incorrectly in the debug output. (It correctly added a newline
1251 to what was transported.)
1253 TF/01 Added $received_time.
1255 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1256 commented out examples of how to interface to a virus scanner and to
1257 SpamAssassin. Also added commented examples of av_scanner and
1258 spamd_address settings.
1260 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1261 and controls are allowed in which ACLs. There were a couple of minor
1262 errors. Some of the entries in the conditions table (which is a table of
1263 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1264 negation of where the condition IS allowed.
1266 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1268 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1269 header file does not have a version number, so I've had to invent a new
1270 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1271 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1272 radiusclient), but it was contributed by a Radius user.
1274 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1275 files or directories, for OpenSSL.
1277 PH/19 When an Exim process that is running as root has to create an Exim log
1278 file, it does so in a subprocess that runs as exim:exim so as to get the
1279 ownership right at creation (otherwise, other Exim processes might see
1280 the file with the wrong ownership). There was no test for failure of this
1281 fork() call, which would lead to the process getting stuck as it waited
1282 for a non-existent subprocess. Forks do occasionally fail when resources
1283 run out. I reviewed all the other calls to fork(); they all seem to check
1286 PH/20 When checking for unexpected SMTP input at connect time (before writing
1287 the banner), Exim was not dealing correctly with a non-positive return
1288 from the read() function. If the client had disconnected by this time,
1289 the result was a log entry for a synchronization error with an empty
1290 string after "input=" when read() returned zero. If read() returned -1
1291 (an event I could not check), uninitialized data bytes were printed.
1292 There were reports of junk text (parts of files, etc) appearing after
1295 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1297 PH/22 Added support for macro redefinition, and (re)definition in between
1298 driver and ACL definitions.
1300 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1301 forgetting to use the resulting value; it was using the unexpanded value.
1303 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1304 hadn't been configured. The fix is from Juergen Kreileder, who
1305 understands it better than I do:
1307 "Here's what I see happening with three configured cyrus_sasl
1308 authenticators configured (plain, login, cram-md5):
1310 On startup auth_cyrus_sasl_init() gets called for each of these.
1311 This means three calls to sasl_listmech() without a specified mech_list.
1312 => SASL tests which mechs of all available mechs actually work
1313 => three warnings about OTP not working
1314 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1316 With the patch, sasl_listmech() also gets called three times. But now
1317 SASL's mech_list option is set to the server_mech specified in the the
1318 authenticator. Or in other words, the answer from sasl_listmech()
1319 gets limited to just the mech you're testing for (which is different
1321 => the return list contains just 'plain' or 'login', 'cram-md5' or
1322 nothing depending on the value of ob->server_mech.
1324 I've just tested the patch: Authentication still works fine,
1325 unavailable mechs specified in the exim configuration are still
1326 caught, and the auth.log warnings about OTP are gone."
1328 PH/25 When debugging is enabled, the contents of the command line are added
1329 to the debugging output, even when log_selector=+arguments is not
1332 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1333 answer is "GNU", and only if the return is "GNU/something" is the answer
1336 PH/27 $acl_verify_message is now set immediately after the failure of a
1337 verification in an ACL, and so is available in subsequent modifiers. In
1338 particular, the message can be preserved by coding like this:
1340 warn !verify = sender
1341 set acl_m0 = $acl_verify_message
1343 Previously, $acl_verify_message was set only while expanding "message"
1344 and "log_message" when a very denied access.
1346 PH/28 Modified OS/os.c-Linux with
1348 -#ifndef OS_LOAD_AVERAGE
1349 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1351 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1352 nomenclature these days.)
1354 PH/29 Installed patch from the Sieve maintainer that adds the options
1355 sieve_useraddress and sieve_subaddress to the redirect router.
1357 PH/30 In these circumstances:
1358 . Two addresses routed to the same list of hosts;
1359 . First host does not offer TLS;
1360 . First host accepts first address;
1361 . First host gives temporary error to second address;
1362 . Second host offers TLS and a TLS session is established;
1363 . Second host accepts second address.
1364 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1365 and peerdn, if requested) that were in fact used only for the second
1368 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1369 attention to any local part prefix or suffix that was matched by the
1370 router that accepted the address. It now behaves in the same way as it
1371 does for delivery: the affixes are removed from the local part unless
1372 rcpt_include_affixes is set on the transport.
1374 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1375 timeout during the DATA phase of an incoming message.
1377 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1378 applied a patch sanctioned by the Sieve maintainer.
1380 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1381 the uid or gid is negative. A case of a negative gid caused this to be
1382 noticed. The fix allows for either to be negative.
1384 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1385 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1388 PH/36 The change PH/12 above was broken. Fixed it.
1390 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1391 the grounds that routing the same address twice would always produce the
1392 same answer. This might have been true once, but it is certainly no
1393 longer true now. Routing a child address may depend on the previous
1394 routing that produced that child. Some complicated redirection strategies
1395 went wrong when messages had multiple recipients, and made Exim's
1396 behaviour dependent on the order in which the addresses were given.
1398 I have moved the duplicate checking until after the routing is complete.
1399 Exim scans the addresses that are assigned to local and remote
1400 transports, and removes any duplicates. This means that more work will be
1401 done, as duplicates will always all be routed, but duplicates are
1402 presumably rare, so I don't expect this is of any significance.
1404 For deliveries to pipes, files, and autoreplies, the duplicate checking
1405 still happens during the routing process, since they are not going to be
1408 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1409 It corrects a timeout issue with spamd. This is Ian's comment: "The
1410 background is that sometimes spamd either never reads data from a
1411 connection it has accepted, or it never writes response data. The exiscan
1412 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1413 blindly assumes that writes won't block so it may never time out."
1415 PH/39 Allow G after quota size as well as K and M.
1417 PH/40 The value set for $authenticated_id in an authenticator may not contain
1418 binary zeroes or newlines because the value is written to log lines and
1419 to spool files. There was no check on this. Now the value is run through
1420 the string_printing() function so that such characters are converted to
1421 printable escape sequences.
1423 PH/41 $message_linecount is a new variable that contains the total number of
1424 lines in the message. Compare $body_linecount, which is the count for the
1427 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1428 bounce and delay warning messages, except in certain special cases, which
1431 (a) An SMTP error message from a remote host;
1432 (b) A message specified in a :fail: redirection;
1433 (c) A message specified in a "fail" command in a system filter;
1434 (d) A message specified in a FAIL return from the queryprogram router;
1435 (e) A message specified by the cannot_route_message router option.
1437 In these cases only, Exim does include the error details in bounce and
1438 warning messages. There are also a few cases where bland messages such
1439 as "unrouteable address" or "local delivery error" are given.
1441 PH/43 $value is now also set for the "else" part of a ${run expansion.
1443 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1444 being worked on, but at least Exim now implements the latest version to
1447 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1448 process to complete was treated as a delivery failure, a timeout while
1449 writing the message to the pipe was logged, but erroneously treated as a
1450 successful delivery. Such timeouts include transport filter timeouts. For
1451 consistency with the overall process timeout, these timeouts are now
1452 treated as errors, giving rise to delivery failures by default. However,
1453 there is now a new Boolean option for the pipe transport called
1454 timeout_defer, which, if set TRUE, converts the failures into defers for
1455 both kinds of timeout. A transport filter timeout is now identified in
1458 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1459 systems where "make" and "gmake" are different, calling "gmake" at top
1460 level broke things. I've arranged for the value of $(MAKE) to be passed
1461 from the Makefile to this script so that it can call the same version of
1465 A note about Exim versions 4.44 and 4.50
1466 ----------------------------------------
1468 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1469 changes of various kinds. As a consequence, a big documentation update was
1470 needed. This delayed the release for rather longer than seemed good, especially
1471 in the light of a couple of (minor) security issues. Therefore, the changes
1472 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1473 release. So 4.44 and 4.50 are in effect two different branches that both start
1476 I have left the 4.50 change log unchanged; it contains all the changes since
1477 4.43. The change log for 4.44 is below; many of its items are identical to
1478 those for 4.50. This seems to be the most sensible way to preserve the
1479 historical information.
1485 1. Minor wording change to the doc/README.SIEVE file.
1487 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1488 computation of the current number of files was incorrect.
1490 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1491 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1494 4. Give more explanation in the error message when the command for a transport
1495 filter fails to execute.
1497 5. There are several places where Exim runs a non-Exim command in a
1498 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1499 was being done only for the command run by the queryprogram router. It is
1500 now done for all such subprocesses. The other cases are: ${run, transport
1501 filters, and the commands run by the lmtp and pipe transports.
1503 6. Added CONFIGURE_GROUP build-time option.
1505 7. Some older OS have a limit of 256 on the maximum number of file
1506 descriptors. Exim was using setrlimit() to set 1000 as a large value
1507 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1508 systems. I've change it so that if it can't get 1000, it tries for 256.
1510 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1511 was an oversight, and furthermore, ever since the addition of extra
1512 controls (e.g. 4.43/32), the checks on when to allow different forms of
1513 "control" were broken. There should now be diagnostics for all cases when a
1514 control that does not make sense is encountered.
1516 9. Added the /retain_sender option to "control=submission".
1518 10. $recipients is now available in the predata ACL (oversight).
1520 11. Tidy the search cache before the fork to do a delivery from a message
1521 received from the command line. Otherwise the child will trigger a lookup
1522 failure and thereby defer the delivery if it tries to use (for example) a
1523 cached ldap connection that the parent has called unbind on.
1525 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1526 of $address_data from the recipient verification was clobbered by the
1527 sender verification.
1529 13. The value of address_data from a sender verification is now available in
1530 $sender_address_data in subsequent conditions in the ACL statement.
1532 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1534 15. Added a new option "connect=<time>" to callout options, to set a different
1537 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1538 was its contents. (It was OK if the option was not defined at all.)
1540 17. A "Completed" log line is now written for messages that are removed from
1541 the spool by the -Mrm option.
1543 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1544 information about exactly what failed.
1546 19. Added -dd to debug only the daemon process.
1548 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1549 handles timeouts, both on the server side and network timeouts. Renamed the
1550 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1552 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1553 It is now set to "smtps".
1555 22. $host_address is now set to the target address during the checking of
1556 ignore_target_hosts.
1558 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1559 being passed; this would have caused $sender_host_name to have been used if
1560 matching the list had actually called for a host name (not very likely,
1561 since this list is usually IP addresses). A host name is now passed as
1564 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1565 code that specifies a non-restarting handler (typically sigaction() in
1566 modern systems) in an attempt to fix a rare and obscure crash bug.
1568 25. Narrowed the window for a race in the daemon that could cause it to ignore
1569 SIGCHLD signals. This is not a major problem, because they are used only to
1570 wake it up if nothing else does.
1572 26. A malformed maildirsize file could cause Exim to calculate negative values
1573 for the mailbox size or file count. Odd effects could occur as a result.
1574 The maildirsize information is now recalculated if the size or filecount
1577 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1578 support for a long time. Removed HAVE_SYS_VFS_H.
1580 28. Installed the latest version of exipick from John Jetmore.
1582 29. In an address list, if the pattern was not a regular expression, an empty
1583 subject address (from a bounce message) matched only if the pattern was an
1584 empty string. Non-empty patterns were not even tested. This was the wrong
1585 because it is perfectly reasonable to use an empty address as part of a
1586 database query. An empty address is now tested by patterns that are
1587 lookups. However, all the other forms of pattern expect the subject to
1588 contain a local part and a domain, and therefore, for them, an empty
1589 address still always fails if the pattern is not itself empty.
1591 30. Exim went into a mad DNS loop when attempting to do a callout where the
1592 host was specified on an smtp transport, and looking it up yielded more
1593 than one IP address.
1595 31. Re-factored the code for checking spool and log partition space into a
1596 function that finds that data and another that does the check. The former
1597 is then used to implement four new variables: $spool_space, $log_space,
1598 $spool_inodes, and $log_inodes.
1600 32. The RFC2047 encoding function was originally intended for short strings
1601 such as real names; it was not keeping to the 75-character limit for
1602 encoded words that the RFC imposes. It now respects the limit, and
1603 generates multiple encoded words if necessary. To be on the safe side, I
1604 have increased the buffer size for the ${rfc2047: expansion operator from
1607 33. It is now permitted to omit both strings after an "if" condition; if the
1608 condition is true, the result is "true". As before, when the second string
1609 is omitted, a false condition yields an empty string. This makes it less
1610 cumbersome to write custom ACL and router conditions.
1612 34. Failure to deliver a bounce message always caused it to be frozen, even if
1613 there was an errors_to setting on the router. The errors_to setting is now
1616 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1617 canonical form (fully expanded) before being placed in
1618 $sender_host_address.
1620 36. The table in the code that translates DNS record types into text (T_A to
1621 "A" for instance) was missing entries for NS and CNAME. It is just possible
1622 that this could have caused confusion if both these types were looked up
1623 for the same domain, because the text type is used as part of Exim's
1624 per-process caching. But the chance of anyone hitting this buglet seems
1627 37. The dnsdb lookup has been extended in a number of ways.
1629 (1) There is a new type, "zns", which walks up the domain tree until it
1630 finds some nameserver records. It should be used with care.
1632 (2) There is a new type, "mxh", which is like "mx" except that it returns
1633 just the host names, not the priorities.
1635 (3) It is now possible to give a list of domains (or IP addresses) to be
1636 looked up. The behaviour when one of the lookups defers can be
1637 controlled by a keyword.
1639 (4) It is now possible to specify the separator character for use when
1640 multiple records are returned.
1642 38. The dnslists ACL condition has been extended: it is now possible to supply
1643 a list of IP addresses and/or domains to be looked up in a particular DNS
1646 39. Added log_selector=+queue_time_overall.
1648 40. When running the queue in the test harness, wait just a tad after forking a
1649 delivery process, to get repeatability of debugging output.
1651 41. Include certificate and key file names in error message when GnuTLS fails
1652 to set them up, because the GnuTLS error message doesn't include the name
1653 of the failing file when there is a problem reading it.
1655 42. Allow both -bf and -bF in the same test run.
1657 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1659 44. The "Exiscan patch" is now merged into the mainline Exim source.
1661 45. Sometimes the final signoff response after QUIT could fail to get
1662 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1663 < 0 before doing a fflush(). This bug looks as though it goes back to the
1664 introduction of TLS in release 3.20, but "sometimes" must have been rare
1665 because the tests only now provoked it.
1667 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1668 (this can affect the format of dates).
1670 47. exim_tidydb, when checking for the continued existence of a message for
1671 which it has found a message-specific retry record, was not finding
1672 messages that were in split spool directories. Consequently, it was
1673 deleting retry records that should have stayed in existence.
1675 48. Steve fixed some bugs in eximstats.
1677 49. The SPA authentication driver was not abandoning authentication and moving
1678 on to the next authenticator when an expansion was forced to fail,
1679 contradicting the general specification for all authenticators. Instead it
1680 was generating a temporary error. It now behaves as specified.
1682 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1683 (the order specifies the preference for clients). The order is now AES256,
1684 AES128, 3DES, ARCFOUR128.
1686 51. Small patch to Sieve code - explicitly set From: when generating an
1689 52. Exim crashed if a remote delivery caused a very long error message to be
1690 recorded - for instance if somebody sent an entire SpamAssassin report back
1691 as a large number of 550 error lines. This bug was coincidentally fixed by
1692 increasing the size of one of Exim's internal buffers (big_buffer) that
1693 happened as part of the Exiscan merge. However, to be on the safe side, I
1694 have made the code more robust (and fixed the comments that describe what
1697 53. Now that there can be additional text after "Completed" in log lines (if
1698 the queue_time_overall log selector is set), a one-byte patch to exigrep
1699 was needed to allow it to recognize "Completed" as not the last thing in
1702 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1703 patch that reportedly fixes this has been added. I am not expert enough to
1704 create a test for it. This is what the patch creator wrote:
1706 "I found a little strange behaviour of ldap code when working with
1707 Windows 2003 AD Domain, where users was placed in more than one
1708 Organization Units. When I tried to give exim partial DN, the exit code
1709 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1710 But simultaneously result of request was absolutely normal ldap result,
1711 so I produce this patch..."
1713 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1714 REFERENCE, so I have modified the code to exclude the patch when that macro
1717 55. Some experimental protocols are using DNS PTR records for new purposes. The
1718 keys for these records are domain names, not reversed IP addresses. The
1719 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1720 leaves it alone. Component reversal etc. now happens only for IP addresses.
1723 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1725 57. Double the size of the debug message buffer (to 2048) so that more of very
1726 long debug lines gets shown.
1728 58. The exicyclog utility now does better if the number of log files to keep
1729 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1731 59. Two changes related to the smtp_active_hostname option:
1733 (1) $smtp_active_hostname is now available as a variable.
1734 (2) The default for smtp_banner uses $smtp_active_hostname instead
1735 of $primary_hostname.
1737 60. The host_aton() function is supposed to be passed a string that is known
1738 to be a valid IP address. However, in the case of IPv6 addresses, it was
1739 not checking this. This is a hostage to fortune. Exim now panics and dies
1740 if the condition is not met. A case was found where this could be provoked
1741 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1742 components; fortuitously, this particular loophole had already been fixed
1743 by change 4.50/55 above.
1745 If there are any other similar loopholes, the new check in host_aton()
1746 itself should stop them being exploited. The report I received stated that
1747 data on the command line could provoke the exploit when Exim was running as
1748 exim, but did not say which command line option was involved. All I could
1749 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1750 running as the user.
1753 61. There was a buffer overflow vulnerability in the SPA authentication code
1754 (which came originally from the Samba project). I have added a test to the
1755 spa_base64_to_bits() function which I hope fixes it.
1758 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1759 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1761 63. The daemon start-up calls getloadavg() while still root for those OS that
1762 need the first call to be done as root, but it missed one case: when
1763 deliver_queue_load_max is set with deliver_drop_privilege. This is
1764 necessary for the benefit of the queue runner, because there is no re-exec
1765 when deliver_drop_privilege is set.
1767 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1768 This has been fixed.
1770 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1771 was in use, was not putting the data itself into the right store pool;
1772 consequently, it could be overwritten for a subsequent message in the same
1773 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1776 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1778 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1779 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1780 the function were treating the return as a boolean value, which happened to
1781 work because 0=false and not-0=true, but is not correct code.
1783 68. The host_aton() function was not handling scoped IPv6 addresses (those
1784 with, for example, "%eth0" on the end) correctly.
1786 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1787 negated items (that is, ~something) in unsigned ints. Some compilers
1788 apparently mutter when there is no cast.
1790 70. If an address verification called from an ACL failed, and did not produce a
1791 user-specific message (i.e. there was only a "system" message), nothing was
1792 put in $acl_verify_message. In this situation, it now puts the system
1795 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1796 session; change 4.31/43 added the unwanted input to the log line - except
1797 that it did not do this in the start of session case. It now does.
1799 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1800 This is wrong and can cause the other end to generate a synchronization
1801 error if it is another Exim or anything else that does the synchronization
1802 check. A QUIT command is no longer sent after a timeout.
1804 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1805 during host lookups.
1807 74. The defer_ok option of callout verification was not working if it was used
1808 when verifying addresses in header lines, that is, for this case:
1810 verify = header_sender/callout=defer_ok
1812 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1813 those file descriptors could be used for SMTP connections. If anything
1814 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1815 could be sent to the SMTP client, causing chaos. The daemon now opens
1816 stdin, stdout, and stderr to /dev/null when it puts itself into the
1819 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1820 log by default. The user can override this with suitable Perl magic.
1822 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1823 the log message when discard triggers, was not working for the DATA ACL or
1824 for the non-SMTP ACL.
1826 78. Error message wording change in sieve.c.
1828 79. If smtp_accept_max_per_host was set, the number of connections could be
1829 restricted to fewer than expected, because the daemon was trying to set up
1830 a new connection before checking whether the processes handling previous
1831 connections had finished. The check for completed processes is now done
1832 earlier. On busy systems, this bug wouldn't be noticed because something
1833 else would have woken the daemon, and it would have reaped the completed
1836 80. If a message was submitted locally by a user whose login name contained one
1837 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1838 It caused a spool format error. I have fixed the spool reading code. A
1839 related problem was that the "from" clause in the Received: line became
1840 illegal because of the space(s). It is now covered by ${quote_local_part.
1842 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1845 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1847 83. Patch from Sieve maintainer to fix unterminated string problem in
1848 "vacation" handling.
1850 84. Some minor changes to the Linux configuration files to help with other
1851 OS variants using glibc.
1853 85. One more patch for Sieve to update vacation handling to latest spec.
1856 ----------------------------------------------------
1857 See the note above about the 4.44 and 4.50 releases.
1858 ----------------------------------------------------
1864 1. Change 4.43/35 introduced a bug that caused file counts to be
1865 incorrectly computed when quota_filecount was set in an appendfile
1868 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1869 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1872 3. Give more explanation in the error message when the command for a transport
1873 filter fails to execute.
1875 4. There are several places where Exim runs a non-Exim command in a
1876 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1877 was being done only for the command run by the queryprogram router. It is
1878 now done for all such subprocesses. The other cases are: ${run, transport
1879 filters, and the commands run by the lmtp and pipe transports.
1881 5. Some older OS have a limit of 256 on the maximum number of file
1882 descriptors. Exim was using setrlimit() to set 1000 as a large value
1883 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1884 systems. I've change it so that if it can't get 1000, it tries for 256.
1886 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1887 was an oversight, and furthermore, ever since the addition of extra
1888 controls (e.g. 4.43/32), the checks on when to allow different forms of
1889 "control" were broken. There should now be diagnostics for all cases when a
1890 control that does not make sense is encountered.
1892 7. $recipients is now available in the predata ACL (oversight).
1894 8. Tidy the search cache before the fork to do a delivery from a message
1895 received from the command line. Otherwise the child will trigger a lookup
1896 failure and thereby defer the delivery if it tries to use (for example) a
1897 cached ldap connection that the parent has called unbind on.
1899 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1900 of $address_data from the recipient verification was clobbered by the
1901 sender verification.
1903 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1904 was its contents. (It was OK if the option was not defined at all.)
1906 11. A "Completed" log line is now written for messages that are removed from
1907 the spool by the -Mrm option.
1909 12. $host_address is now set to the target address during the checking of
1910 ignore_target_hosts.
1912 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1913 being passed; this would have caused $sender_host_name to have been used if
1914 matching the list had actually called for a host name (not very likely,
1915 since this list is usually IP addresses). A host name is now passed as
1918 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1919 code that specifies a non-restarting handler (typically sigaction() in
1920 modern systems) in an attempt to fix a rare and obscure crash bug.
1922 15. Narrowed the window for a race in the daemon that could cause it to ignore
1923 SIGCHLD signals. This is not a major problem, because they are used only to
1924 wake it up if nothing else does.
1926 16. A malformed maildirsize file could cause Exim to calculate negative values
1927 for the mailbox size or file count. Odd effects could occur as a result.
1928 The maildirsize information is now recalculated if the size or filecount
1931 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1932 support for a long time. Removed HAVE_SYS_VFS_H.
1934 18. Updated exipick to current release from John Jetmore.
1936 19. Allow an empty sender to be matched against a lookup in an address list.
1937 Previously the only cases considered were a regular expression, or an
1940 20. Exim went into a mad DNS lookup loop when doing a callout where the
1941 host was specified on the transport, if the DNS lookup yielded more than
1944 21. The RFC2047 encoding function was originally intended for short strings
1945 such as real names; it was not keeping to the 75-character limit for
1946 encoded words that the RFC imposes. It now respects the limit, and
1947 generates multiple encoded words if necessary. To be on the safe side, I
1948 have increased the buffer size for the ${rfc2047: expansion operator from
1951 22. Failure to deliver a bounce message always caused it to be frozen, even if
1952 there was an errors_to setting on the router. The errors_to setting is now
1955 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1956 canonical form (fully expanded) before being placed in
1957 $sender_host_address.
1959 24. Updated eximstats to version 1.33
1961 25. Include certificate and key file names in error message when GnuTLS fails
1962 to set them up, because the GnuTLS error message doesn't include the name
1963 of the failing file when there is a problem reading it.
1965 26. Expand error message when OpenSSL has problems setting up cert/key files.
1968 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1969 (this can affect the format of dates).
1971 28. exim_tidydb, when checking for the continued existence of a message for
1972 which it has found a message-specific retry record, was not finding
1973 messages that were in split spool directories. Consequently, it was
1974 deleting retry records that should have stayed in existence.
1976 29. eximstats updated to version 1.35
1977 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1978 1.35 - bugfix such that pie charts by volume are generated correctly
1980 30. The SPA authentication driver was not abandoning authentication and moving
1981 on to the next authenticator when an expansion was forced to fail,
1982 contradicting the general specification for all authenticators. Instead it
1983 was generating a temporary error. It now behaves as specified.
1985 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1986 (the order specifies the preference for clients). The order is now AES256,
1987 AES128, 3DES, ARCFOUR128.
1989 31. Small patch to Sieve code - explicitly set From: when generating an
1992 32. Exim crashed if a remote delivery caused a very long error message to be
1993 recorded - for instance if somebody sent an entire SpamAssassin report back
1994 as a large number of 550 error lines. This bug was coincidentally fixed by
1995 increasing the size of one of Exim's internal buffers (big_buffer) that
1996 happened as part of the Exiscan merge. However, to be on the safe side, I
1997 have made the code more robust (and fixed the comments that describe what
2000 33. Some experimental protocols are using DNS PTR records for new purposes. The
2001 keys for these records are domain names, not reversed IP addresses. The
2002 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2003 leaves it alone. Component reversal etc. now happens only for IP addresses.
2006 34. The host_aton() function is supposed to be passed a string that is known
2007 to be a valid IP address. However, in the case of IPv6 addresses, it was
2008 not checking this. This is a hostage to fortune. Exim now panics and dies
2009 if the condition is not met. A case was found where this could be provoked
2010 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2011 components; fortuitously, this particular loophole had already been fixed
2012 by change 4.50/55 or 4.44/33 above.
2014 If there are any other similar loopholes, the new check in host_aton()
2015 itself should stop them being exploited. The report I received stated that
2016 data on the command line could provoke the exploit when Exim was running as
2017 exim, but did not say which command line option was involved. All I could
2018 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2019 running as the user.
2022 35. There was a buffer overflow vulnerability in the SPA authentication code
2023 (which came originally from the Samba project). I have added a test to the
2024 spa_base64_to_bits() function which I hope fixes it.
2027 36. The daemon start-up calls getloadavg() while still root for those OS that
2028 need the first call to be done as root, but it missed one case: when
2029 deliver_queue_load_max is set with deliver_drop_privilege. This is
2030 necessary for the benefit of the queue runner, because there is no re-exec
2031 when deliver_drop_privilege is set.
2033 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2034 was in use, was not putting the data itself into the right store pool;
2035 consequently, it could be overwritten for a subsequent message in the same
2036 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2039 38. Sometimes the final signoff response after QUIT could fail to get
2040 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2041 < 0 before doing a fflush(). This bug looks as though it goes back to the
2042 introduction of TLS in release 3.20, but "sometimes" must have been rare
2043 because the tests only now provoked it.
2049 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2050 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2051 Now it returns an integer. A number of calls were still expecting a T/F
2052 return. Fortuitously, in all cases, the tests worked in OK situations,
2053 which is the norm. However, things would have gone wrong on any write
2054 failures on the smtp file descriptor. This function is used when sending
2055 messages over SMTP and also when doing verify callouts.
2057 2. When Exim is called to do synchronous delivery of a locally submitted
2058 message (the -odf or -odi options), it no longer closes stderr before doing
2061 3. Implemented the mua_wrapper option.
2063 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2065 5. Implemented the functions header_remove(), header_testname(),
2066 header_add_at_position(), and receive_remove_recipient(), and exported them
2069 6. If an ACL "warn" statement specified the addition of headers, Exim already
2070 inserted X-ACL-Warn: at the start if there was no header name. However, it
2071 was not making this test for the second and subsequent header lines if
2072 there were newlines in the string. This meant that an invalid header could
2073 be inserted if Exim was badly configured.
2075 7. Allow an ACL "warn" statement to add header lines at the start or after all
2076 the Received: headers, as well as at the end.
2078 8. Added the rcpt_4xx retry error code.
2080 9. Added postmaster_mailfrom=xxx to callout verification option.
2082 10. Added mailfrom=xxxx to the callout verification option, for verify=
2085 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2086 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2088 12. Inserted some casts to stop certain compilers warning when using pointer
2089 differences as field lengths or precisions in printf-type calls (mostly
2090 affecting debugging statements).
2092 13. Added optional readline() support for -be (dynamically loaded).
2094 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2095 same clock tick as a message's arrival, so that its received time was the
2096 same as the "first fail" time on the retry record, and that message
2097 remained on the queue past the ultimate address timeout, every queue runner
2098 would try a delivery (because it was past the ultimate address timeout) but
2099 after another failure, the ultimate address timeout, which should have then
2100 bounced the address, did not kick in. This was a "< instead of <=" error;
2101 in most cases the first failure would have been in the next clock tick
2102 after the received time, and all would be well.
2104 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2105 being recognized when the domain list was tested by the match_domain
2106 condition in an expansion string.
2108 16. Added the ${str2b64: operator.
2110 17. Exim was always calling setrlimit() to set a large limit for the number of
2111 processes, without checking whether the existing limit was already
2112 adequate. (It did check for the limit on file descriptors.) Furthermore,
2113 errors from getrlimit() and setrlimit() were being ignored. Now they are
2114 logged to the main and panic logs, but Exim does carry on, to try to do its
2115 job under whatever limits there are.
2117 18. Imported PCRE 5.0.
2119 19. Trivial typo in log message " temporarily refused connection" (the leading
2122 20. If the log selector return_path_on_delivery was set and an address was
2123 redirected to /dev/null, the delivery process crashed because it assumed
2124 that a return path would always be set for a "successful" delivery. In this
2125 case, the whole delivery is bypassed as an optimization, and therefore no
2128 21. Internal re-arrangement: the function for sending a challenge and reading
2129 a response while authentication was assuming a zero-terminated challenge
2130 string. It's now changed to take a pointer and a length, to allow for
2131 binary data in such strings.
2133 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2135 23. Exim was not respecting finduser_retries when seeking the login of the
2136 uid under which it was called; it was always trying 10 times. (The default
2137 setting of finduser_retries is zero.) Also, it was sleeping after the final
2138 failure, which is pointless.
2140 24. Implemented tls_on_connect_ports.
2142 25. Implemented acl_smtp_predata.
2144 26. If the domain in control=submission is set empty, Exim assumes that the
2145 authenticated id is a complete email address when it generates From: or
2146 Sender: header lines.
2148 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2149 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2150 chown and chgrp in /bin and hostname in /usr/bin.
2152 28. Exim was keeping the "process log" file open after each use, just as it
2153 does for the main log. This opens the possibility of it remaining open for
2154 long periods when the USR1 signal hits a daemon. Occasional processlog
2155 errors were reported, that could have been caused by this. Anyway, it seems
2156 much more sensible not to leave this file open at all, so that is what now
2159 29. The long-running daemon process does not normally write to the log once it
2160 has entered its main loop, and it closes the log before doing so. This is
2161 so that log files can straightforwardly be renamed and moved. However,
2162 there are a couple of unusual error situations where the daemon does write
2163 log entries, and I had neglected to close the log afterwards.
2165 30. The text of an SMTP error response that was received during a remote
2166 delivery was being truncated at 512 bytes. This is too short for some of
2167 the long messages that one sometimes sees. I've increased the limit to
2170 31. It is now possible to make retry rules that apply only when a message has a
2171 specific sender, in particular, an empty sender.
2173 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2174 it possible to be selective about when SMTP synchronization is enforced.
2176 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2178 32. Implemented hosts_connection_nolog.
2180 33. Added an ACL for QUIT.
2182 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2185 35. Added mailbox_size and mailbox_filecount to appendfile.
2187 36. Added control = no_multiline_responses to ACLs.
2189 37. There was a bug in the logic of the code that waits for the clock to tick
2190 in the case where the clock went backwards by a substantial amount such
2191 that the microsecond fraction of "now" was more than the microsecond
2192 fraction of "then" (but the whole seconds number was less).
2194 38. Added support for the libradius Radius client library this is found on
2195 FreeBSD (previously only the radiusclient library was supported).
2201 1. When certain lookups returned multiple values in the form name=value, the
2202 quoting of the values was not always being done properly. Specifically:
2203 (a) If the value started with a double quote, but contained no whitespace,
2205 (b) If the value contained whitespace other than a space character (i.e.
2206 tabs or newlines or carriage returns) it was not quoted.
2207 This fix has been applied to the mysql and pgsql lookups by writing a
2208 separate quoting function and calling it from the lookup code. The fix
2209 should probably also be applied to nisplus, ibase and oracle lookups, but
2210 since I cannot test any of those, I have not disturbed their existing code.
2212 2. A hit in the callout cache for a specific address caused a log line with no
2213 reason for rejecting RCPT. Now it says "Previous (cached) callout
2214 verification failure".
2216 3. There was an off-by-one bug in the queryprogram router. An over-long
2217 return line was truncated at 256 instead of 255 characters, thereby
2218 overflowing its buffer with the terminating zero. As well as fixing this, I
2219 have increased the buffer size to 1024 (and made a note to document this).
2221 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2222 when Exim is waiting for an SMTP response from a remote server, Exim
2223 restarts its select() call on the socket, thereby resetting its timeout.
2224 This is not a problem when such interrupts are rare. Somebody set up a cron
2225 job to run exiwhat every 2 minutes, which is less than the normal select()
2226 timeout (5 or 10 minutes). This meant that the select() timeout never
2227 kicked in because it was always reset. I have fixed this by comparing the
2228 time when an interrupt arrives with the time at the start of the first call
2229 to select(). If more time than the timeout has elapsed, the interrupt is
2230 treated as a timeout.
2232 5. Some internal re-factoring in preparation for the addition of Sieve
2233 extensions (by MH). In particular, the "personal" test is moved to a
2234 separate function, and given an option for scanning Cc: and Bcc: (which is
2235 not set for Exim filters).
2237 6. When Exim created an email address using the login of the caller as the
2238 local part (e.g. when creating a From: or Sender: header line), it was not
2239 quoting the local part when it contained special characters such as @.
2241 7. Installed new OpenBSD configuration files.
2243 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2244 try to make them clearer.
2246 9. Callout options, other than the timeout value, were being ignored when
2247 verifying sender addresses in header lines. For example, when using
2249 verify = header_sender/callout=no_cache
2251 the cache was (incorrectly) being used.
2253 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2254 only the Cygwin environment.
2256 11. When return_path_on_delivery was set as a log selector, if different remote
2257 addresses in the same message used different return paths and parallel
2258 remote delivery occurred, the wrong values would sometimes be logged.
2259 (Whenever a remote delivery process finished, the return path value from
2260 the most recently started remote delivery process was logged.)
2262 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2263 lines when AUTH and/or TLS are in use. This is the "received protocol"
2264 field. Exim used to use "asmtp" for authenticated SMTP, without any
2265 indication (in the protocol name) for TLS use. Now it follows the RFC and
2266 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2267 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2268 names appear in log lines as well as in Received: header lines.
2270 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2271 extensions, and comparison tests, and to fix some bugs.
2273 14. Changes to the "personal" filter test:
2275 (1) The test was buggy in that it was just doing the equivalent of
2276 "contains" tests on header lines. For example, if a user's address was
2277 anne@some.where, the "personal" test would incorrectly be true for
2279 To: susanne@some.where
2281 This test is now done by extracting each address from the header in turn,
2282 and checking the entire address. Other tests that are part of "personal"
2283 are now done using regular expressions (for example, to check local parts
2284 of addresses in From: header lines).
2286 (2) The list of non-personal local parts in From: addresses has been
2287 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2288 taken from the Sieve specification recommendations.
2290 (3) If the message contains any header line starting with "List-" it is
2291 treated as non-personal.
2293 (4) The test for "circular" in the Subject: header line has been removed
2294 because it now seems ill-conceived.
2296 15. Minor typos in src/EDITME comments corrected.
2298 16. Installed latest exipick from John Jetmore.
2300 17. If headers_add on a router specified a text string that was too long for
2301 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2302 of string_sprintf() is now avoided.
2304 18. $message_body_size was not set (it was always zero) when running the DATA
2305 ACL and the local_scan() function.
2307 19. For the "mail" command in an Exim filter, no default was being set for
2308 the once_repeat time, causing a random time value to be used if "once" was
2309 specified. (If the value happened to be <= 0, no repeat happened.) The
2310 default is now 0s, meaning "never repeat". The "vacation" command was OK
2311 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2312 (I found it when inspecting the code).
2314 20. There is now an overall timeout for performing a callout verification. It
2315 defaults to 4 times the callout timeout, which applies to individual SMTP
2316 commands during the callout. The overall timeout applies when there is more
2317 than one host that can be tried. The timeout is checked before trying the
2318 next host. This prevents very long delays if there are a large number of
2319 hosts and all are timing out (e.g. when the network connections are timing
2320 out). The value of the overall timeout can be changed by specifying an
2321 additional sub-option for "callout", called "maxwait". For example:
2323 verify = sender/callout=5s,maxwait=20s
2325 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2326 to the end before writing, but this should make it even safer).
2328 22. Exim was forgetting that it had advertised PIPELINING for the second and
2329 subsequent messages on an SMTP connection. It was also not resetting its
2330 memory on STARTTLS and an internal HELO.
2332 23. When Exim logs an SMTP synchronization error within a session, it now
2333 records whether PIPELINING has been advertised or not.
2335 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2336 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2337 rather than long int.
2339 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2341 26. Added the never_mail option to autoreply.
2347 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2348 crash if the getsockname() call failed; this can happen if a connection is
2349 closed very soon after it is established. The problem was simply in the
2350 order in which certain operations were done, causing Exim to try to write
2351 to the SMTP stream before it had set up the file descriptor. The bug has
2352 been fixed by making things happen in the correct order.
2358 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2359 before the connection was closed, thus losing the rejection response.
2361 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2362 some early Solaris releases, but causes trouble in current releases where
2363 socklen_t is defined.
2365 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2368 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2371 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2372 that releasing the top of it at the end releases what was used for sub-
2373 expansions (unless the block got too big). However, discard this block if
2374 the first thing is a variable or header, so that we can use its block when
2375 it is dynamic (useful for very large $message_headers, for example).
2377 6. Lookups now cache *every* query, not just the most recent. A new, separate
2378 store pool is used for this. It can be recovered when all lookup caches are
2379 flushed. Lookups now release memory at the end of their result strings.
2380 This has involved some general refactoring of the lookup sources.
2382 7. Some code has been added to the store_xxx() functions to reduce the amount
2383 of flapping under certain conditions.
2385 8. log_incoming_interface used to affect only the <= reception log lines. Now
2386 it causes the local interface and port to be added to several more SMTP log
2387 lines, for example "SMTP connection from", and rejection lines.
2389 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2391 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2393 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2394 could be overwritten at the end of the current message (or the start of a
2395 new message if it was set in a HELO ACL). The value is now preserved for
2396 the duration of the SMTP connection.
2398 12. If a transport had a headers_rewrite setting, and a matching header line
2399 contained an unqualified address, that address was qualified, even if it
2400 did not match any rewriting rules. The underlying bug was that the values
2401 of the flags that permit the existence of unqualified sender and recipient
2402 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2403 non-local messages, and by -bnq for local messages) were not being
2404 preserved with the message after it was received.
2406 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2407 "next input=" as part of the text comprising the host identity instead of
2408 the correct text. The code was using the same buffer for two different
2409 strings. However, depending on which order the printing function evaluated
2410 its arguments, the bug did not always show up. Under Linux, for example, my
2411 test suite worked just fine.
2413 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2414 doesn't work with some older versions of Perl. It has been changed to "my",
2415 which in any case is probably the better facility to use.
2417 15. A really picky compiler found some instances of statements for creating
2418 error messages that either had too many or two few arguments for the format
2421 16. The size of the buffer for calls to the DNS resolver has been increased
2422 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2423 for addresses that have a lot of PTR records. This alleviates a problem; it
2424 does not fully solve it.
2426 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2427 buffer now truncates the list and logs the incident, which is the same
2428 action as happens when Exim is looking up a host name and its aliases.
2429 Previously in this situation something unpredictable would happen;
2430 sometimes it was "internal error: store_reset failed".
2432 18. If a server dropped the connection unexpectedly when an Exim client was
2433 using GnuTLS and trying to read a response, the client delivery process
2434 crashed while trying to generate an error log message.
2436 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2437 string, for example:
2439 warn message = H1: something\nH2: something
2441 the text was added as a single header line from Exim's point of view
2442 though it ended up OK in the delivered message. However, searching for the
2443 second and subsequent header lines using $h_h2: did not work. This has been
2444 fixed. Similarly, if a system filter added multiple headers in this way,
2445 the routers could not see them.
2447 20. Expanded the error message when iplsearch is called with an invalid key to
2448 suggest using net-iplsearch in a host list.
2450 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2451 ACLs are no longer actually imposed (and a message to that effect is
2454 22. If a "gecos" field in a passwd entry contained escaped characters, in
2455 particular, if it contained a \" sequence, Exim got it wrong when building
2456 a From: or a Sender: header from that name. A second bug also caused
2457 incorrect handling when an unquoted " was present following a character
2458 that needed quoting.
2460 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2461 was not being matched caselessly.
2463 24. Arranged for all hyphens in the exim.8 source to be escaped with
2466 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2467 properly. Recipient callout cache records were still being keyed to include
2468 the sender, even when use_sender was set false. This led to far more
2469 callouts that were necessary. The sender is no longer included in the key
2470 when use_sender is false.
2472 26. Added "control = submission" modifier to ACLs.
2474 27. Added the ${base62d: operator to decode base 62 numbers.
2476 28. dnsdb lookups can now access SRV records.
2478 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2479 the configuration file.
2481 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2482 (-v) mode. This makes the output for a verbose queue run more intelligible.
2484 31. Added a use_postmaster feature to recipient callouts.
2486 32. Added the $body_zerocount variable, containing the number of binary zero
2487 bytes in the message body.
2489 33. The time of last modification of the "new" subdirectory is now used as the
2490 "mailbox time last read" when there is a quota error for a maildir
2493 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2495 35. Added +ignore_unknown as a special item in host lists.
2497 36. Code for decoding IPv6 addresses in host lists is now included, even if
2498 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2499 address was recognized as an IP address, but was then not correctly decoded
2500 into binary, causing unexpected and incorrect effects when compared with
2507 1. Very minor rewording of debugging text in manualroute to say "list of
2508 hosts" instead of "hostlist".
2510 2. If verify=header_syntax was set, and a header line with an unqualified
2511 address (no domain) and a large number of spaces between the end of the
2512 name and the colon was received, the reception process suffered a buffer
2513 overflow, and (when I tested it) crashed. This was caused by some obsolete
2514 code that should have been removed. The fix is to remove it!
2516 3. When running in the test harness, delay a bit after writing a bounce
2517 message to get a bit more predictability in the log output.
2519 4. Added a call to search_tidyup() just before forking a reception process. In
2520 theory, someone could use a lookup in the expansion of smtp_accept_max_
2521 per_host which, without the tidyup, could leave open a database connection.
2523 5. Added the variables $recipient_data and $sender_data which get set from a
2524 lookup success in an ACL "recipients" or "senders" condition, or a router
2525 "senders" option, similar to $domain_data and $local_part_data.
2527 6. Moved the writing of debug_print from before to after the "senders" test
2530 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2531 problems for message scanning, either using a data ACL, or using
2532 local_scan() because the Received: header was not generated till after they
2533 were called (in order to set the time as the time of reception completion).
2534 I have revised the way this works. The header is now generated after the
2535 body is received, but before the ACL or local_scan() are called. After they
2536 are run, the timestamp in the header is updated.
2542 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2543 before starting a queue runner without re-exec. This happened only when
2544 deliver_drop_privilege was set or when the Exim user was set to root. The
2545 effect of the bug was that timeouts during subsequent deliveries caused
2546 crashes instead of being properly handled. The handler is now left at its
2547 default (and expected) setting.
2549 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2550 message, again when deliver_drop_privilege is set or Exim is run as root.
2551 The bug described in (1) was not present in this case, but the tidying up
2552 of the other signals was missing. I have made the two cases consistent.
2554 3. The ignore_target_hosts setting on a manualroute router was being ignored
2555 for hosts that were looked up using the /MX notation.
2557 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2560 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2561 operated on the sender address. After changing the $sender_address to <>
2562 for the sender address verify, Exim was re-instated it as the original
2563 (before rewriting) address, but remembering that it had rewritten it, so it
2564 wasn't rewriting it again. This bug also had the effect of breaking the
2565 sender address verification caching when the sender address was rewritten.
2567 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2568 This has been changed so that if the ip literal address matches
2569 ignore_target_hosts, the router declines.
2571 7. Added expansion conditions match_domain, match_address, and match_local_
2572 part (NOT match_host).
2574 8. The placeholder for the Received: header didn't have a length field set.
2576 9. Added code to Exim itself and to exim_lock to test for a specific race
2577 condition that could lead to file corruption when using MBX delivery. The
2578 issue is with the lockfile that is created in /tmp. If this file is removed
2579 after a process has opened it but before that process has acquired a lock,
2580 there is the potential for a second process to recreate the file and also
2581 acquire a lock. This could lead to two Exim processes writing to the file
2582 at the same time. The added code performs the same test as UW imapd; it
2583 checks after acquiring the lock that its file descriptor still refers to
2584 the same named file.
2586 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2587 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2590 11. Added the smtp_active_hostname option. If used, this will typically be made
2591 to depend on the incoming interface address. Because $interface_address is
2592 not set up until the daemon has forked a reception process, error responses
2593 that can happen earlier (such as "too many connections") no longer contain
2596 12. If an expansion in a condition on a "warn" statement fails because a lookup
2597 defers, the "warn" statement is abandoned, and the next ACL statement is
2598 processed. Previously this caused the whole ACL to be aborted.
2600 13. Added the iplsearch lookup type.
2602 14. Added ident_timeout as a log selector.
2604 15. Added tls_certificate_verified as a log selector.
2606 16. Added a global option tls_require_ciphers (compare the smtp transport
2607 option of the same name). This controls incoming TLS connections.
2609 17. I finally figured out how to make tls_require_ciphers do a similar thing
2610 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2611 before starting the TLS session.
2613 18. Tabs are now shown as \t in -bP output.
2615 19. If the log selector return_path_on_delivery was set, Exim crashed when
2616 bouncing a message because it had too many Received: header lines.
2618 20. If two routers both had headers_remove settings, and the first one included
2619 a superfluous trailing colon, the final name in the first list and the
2620 first name in the second list were incorrectly joined into one item (with a
2621 colon in the middle).
2627 1. Added -C and -D options to the exinext utility, mainly to make it easier
2628 to include in the automated testing, but these could be helpful when
2629 multiple configurations are in use.
2631 2. The exinext utility was not formatting the output nicely when there was
2632 an alternate port involved in the retry record key, nor when there was a
2633 message id as well (for retries that were specific to a specific message
2634 and a specific host). It was also confused by IPv6 addresses, because of
2635 the additional colons they contain. I have fixed the IPv4 problem, and
2636 patched it up to do a reasonable job for IPv6.
2638 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2639 delivery, the log line now contains "pipelined" if PIPELINING was used.
2641 4. An SMTP transport process used to panic and die if the bind() call to set
2642 an explicit outgoing interface failed. This has been changed; it is now
2643 treated in the same way as a connect() failure.
2645 5. A reference to $sender_host_name in the part of a conditional expansion
2646 that was being skipped was still causing a DNS lookup. This no longer
2649 6. The def: expansion condition was not recognizing references to header lines
2650 that used bh_ and bheader_.
2652 7. Added the _cache feature to named lists.
2654 8. The code for checking quota_filecount in the appendfile transport was
2655 allowing one more file than it should have been.
2657 9. For compatibility with Sendmail, the command line option
2665 and sets the incoming protocol and host name (for trusted callers). The
2666 host name and its colon can be omitted when only the protocol is to be set.
2667 Note the Exim already has two private options, -pd and -ps, that refer to
2668 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2669 "s", but I don't think that's a major issue.
2671 10. A number of refactoring changes to the code, none of which should affect
2674 (a) The number of logging options was getting close to filling up the
2675 32-bit word that was used as a bit map. I have split them into two classes:
2676 those that are passed in the argument to log_write(), and those that are
2677 only ever tested independently outside of that function. These are now in
2678 separate 32-bit words, so there is plenty of room for expansion again.
2679 There is no change in the user interface or the logging behaviour.
2681 (b) When building, for example, log lines, the code previously used a
2682 macro that called string_cat() twice, in order to add two strings. This is
2683 not really sufficiently general. Furthermore, there was one instance where
2684 it was actually wrong because one of the argument was used twice, and in
2685 one call a function was used. (As it happened, calling the function twice
2686 did not affect the overall behaviour.) The macro has been replaced by a
2687 function that can join an arbitrary number of extra strings onto a growing
2690 (c) The code for expansion conditions now uses a table and a binary chop
2691 instead of a serial search (which was left over from when there were very
2692 few conditions). Also, it now recognizes conditions like "pam" even when
2693 the relevant support is not compiled in: a suitably worded error message is
2694 given if an attempt is made to use such a condition.
2696 11. Added ${time_interval:xxxxx}.
2698 12. A bug was causing one of the ddress fields not to be passed back correctly
2699 from remote delivery subprocesses. The field in question was not being
2700 subsequently used, so this caused to problems in practice.
2702 13. Added new log selectors queue_time and deliver_time.
2704 14. Might have fixed a bug in maildirsizefile handling that threw up
2705 "unexpected character" debug warnings, and recalculated the data
2706 unnecessarily. In any case, I expanded the warning message to give more
2709 15. Added the message "Restricted characters in address" to the statements in
2710 the default ACL that block characters like @ and % in local parts.
2712 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2713 Three changes have been made:
2715 (a) There was a serious bug; a negative response to MAIL caused the whole
2716 recipient domain to be cached as invalid, thereby blocking all messages
2717 to all local parts at the same domain, from all senders. This bug has
2718 been fixed. The domain is no longer cached after a negative response to
2719 MAIL if the sender used is not empty.
2721 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2724 (c) A new callout option, "use_sender" has been added for people who want
2725 the modified behaviour.
2731 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2734 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2735 indeed breaks things for older releases.
2737 3. Added additional logging to the case where there is a problem reading data
2738 from a filter that is running in a subprocess using a pipe, in order to
2739 try to track down a specific problem.
2741 4. Testing facility fudge: when running in the test harness and attempting
2742 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2743 getting "No route to host". Convert this to a timeout.
2745 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2748 6. Some OS don't have socklen_t but use size_t instead. This affects the
2749 fifth argument of getsockopt() amongst other things. This is now
2750 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2751 can be set for individual OS. I have set it for SunOS5, OSF1, and
2752 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2753 some earlier ones do not.
2755 7. Change 4.30/15 was not doing the test caselessly.
2757 8. The standard form for an IPv6 address literal was being rejected by address
2758 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2759 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2760 this, as well as the form without the "IPv6" on the front (but only when
2761 address literals are enabled, of course).
2763 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2765 10. Exim crashed if a message with an empty sender address specified by -f
2766 encountered a router with an errors_to setting. This could be provoked only
2767 by a command such as
2771 where an empty string was supplied; "<>" did not hit this bug.
2773 11. Installed PCRE release 4.5.
2775 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2776 remained set. It is now erased.
2778 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2779 times from message ids (which are base 36 rather than the normal 62).
2781 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2782 were being counted as actual protocol errors, and logged if the log
2783 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2784 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2785 and DATA following a set of rejected RCPTs do not count as protocol errors.
2786 In other words, Exim assumes they were pipelined, though this may not
2787 actually be the case. Of course, in all cases the client gets an
2788 appropriate error code.
2790 15. If a lookup fails in an ACL condition, a message about the failure may
2791 be available; it is used if testing the ACL cannot continue, because most
2792 such messages specify what the cause of the deferral is. However, some
2793 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2794 that caused an old message to be retained and used if a later statement
2795 caused a defer, replacing the real cause of the deferral.
2797 16. If an IP address had so many PTR records that the DNS lookup buffer
2798 was not large enough to hold them, Exim could crash while trying to process
2799 the truncated data. It now detects and logs this case.
2801 17. Further to 4.21/58, another change has been made: if (and only if) the
2802 first line of a message (the first header line) ends with CRLF, a bare LF
2803 in a subsequent header line has a space inserted after it, so as not to
2804 terminate the header.
2806 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2807 unnecessarily, used atoi() instead of strtol(), and didn't check the
2808 termination when getting file sizes from file names by regex.
2810 19. Completely re-implemented the support for maildirsize files, in the light
2811 of a number of problems with the previous contributed implementation
2812 (4.30/29). In particular:
2814 . If the quota is zero, the maildirsize file is maintained, but no quota is
2817 . If the maildir directory does not exist, it is created before any attempt
2818 to write a maildirsize file.
2820 . The quota value in the file is just a cache; if the quota is changed in
2821 the transport, the new value overrides.
2823 . A regular expression is available for excluding directories from the
2826 20. The autoreply transport checks the characters in options that define the
2827 message's headers; it allows continued headers, but it was checking with
2828 isspace() after an embedded newline instead of explicitly looking for a
2831 21. If all the "regular" hosts to which an address was routed had passed their
2832 expiry times, and had not reached their retry times, the address was
2833 bounced, even if fallback hosts were defined. Now Exim should go on to try
2836 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2837 equivalent code in the SMTP transport. Some hosts send humungous responses
2838 to HELO/EHLO, more than 1024 it seems.
2840 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2841 gives compiler warnings in some environments. I've now done it "properly",
2844 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2845 (because of problems with the built-in one) was declared to return uschar *
2846 instead of char *, causing compiler failure.
2848 25. Fixed a file descriptor leak when processing alias/forward files.
2850 26. Fixed a minor format string issue in dbfn.c.
2852 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2854 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2855 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2858 29. When a sender address is verified, it is cached, to save repeating the test
2859 when there is more than one recipient in a message. However, when the
2860 verification involves a callout, it is possible for different callout
2861 options to be set for different recipients. It is too complicated to keep
2862 track of this in the cache, so now Exim always runs a verification when a
2863 callout is required, relying on the callout cache for the optimization.
2864 The overhead is duplication of the address routing, but this should not be
2867 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2868 to be verified with callout=postmaster, and the main callout worked but the
2869 postmaster check failed, the verification correctly failed. However, if a
2870 subsequent RCPT command asked for sender verification *without* the
2871 postmaster check, incorrect caching caused this verification also to fail,
2874 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2875 it was not caching the DNS options (qualify_single, search_parents) that
2876 were used when the lookup failed. A subsequent lookup with different
2877 options therefore always gave the same answer, though there were cases
2878 where it should not have. (Example: a "domains = !$mx_any" option on a
2879 dnslookup router: the "domains" option is always processed without any
2880 widening, but the router might have qualify_single set.) Now Exim uses the
2881 cached value only when the same options are set.
2883 32. Added John Jetmore's "exipick" utility to the distribution.
2885 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2886 than a timeout (e.g. a certificate is required, and is not provided), an
2887 Exim server now closes the connection immediately. Previously it waited for
2888 the client to close - but if the client is SSL, it seems that they each
2889 wait for each other, leading to a delay before one of them times out.
2891 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2892 maintained 0.8.x compatibility because I don't think many are using it, and
2893 it is clearly obsolete.
2895 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2898 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2899 client certificate was expired. A simple patch fixes this, though I don't
2900 understand the full logic of why the verify callback is called multiple
2903 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2904 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2905 which causes problems with some clients (such as the Certicom SSL Plus
2906 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2907 disables the coutermeasure allowing Eudora to connect."
2909 38. Exim was not checking that a write() to a log file succeeded. This could
2910 lead to Bad Things if a log got too big, in particular if it hit a file
2911 size limit. Exim now panics and dies if it cannot write to a log file, just
2912 as it does if it cannot open a log file.
2914 39. Modified OS/Makefile-Linux so that it now contains
2916 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2918 The two -D definitions ensure that Exim is compiled with large file
2919 support, which makes it possible to handle log files that are bigger than
2922 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2923 instance) a domain was checked against a named list that involved a lookup,
2924 causing $domain_data to be set, then another domain was checked against the
2925 same list, then the first domain was re-checked, the value of $domain_data
2926 after the final check could be wrong. In particular, if the second check
2927 failed, it could be set empty. This bug probably also applied to
2930 41. The strip_trailing_dot option was not being applied to the address given
2931 with the -f command-line option.
2933 42. The code for reading a message's header from the spool was incrementing
2934 $received_count, but never initializing it. This meant that the value was
2935 incorrect (doubled) while delivering a message in the same process in which
2936 it was received. In the most common configuration of Exim, this never
2937 happens - a fresh exec is done - but it can happen when
2938 deliver_drop_privilege is set.
2940 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2941 it now includes up to 150 characters of the unexpected data in the log
2944 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2945 and building data strings. The size of both of these buffers was 10 000
2946 bytes - far larger than anybody would *ever* want, thought I. Needless to
2947 say, somebody hit the limit. I have increased the maximum line length to
2948 20 000 and the maximum data length of concatenated lines to 100 000. I have
2949 also fixed two bugs, because there was no checking on these buffers. Tsk,
2950 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2951 buffer is too small.
2953 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2954 lsearch lookups. Now it does.
2956 46. When parsing a route_list item in a manualroute router, a fixed-length
2957 buffer was used for the list of hosts. I made this 1024 bytes long,
2958 thinking that nobody would ever have a list of hosts that long. Wrong.
2959 Somebody had a whole pile of complicated expansion conditions, and the
2960 string was silently truncated, leading to an expansion error. It turns out
2961 that it is easier to change to an unlimited length (owing to other changes
2962 that have happened since this code was originally written) than to build
2963 structure for giving a limitation error. The length of the item that
2964 expands into the list of hosts is now unlimited.
2966 47. The lsearch lookup could not handle data where the length of text line was
2967 more than 4095 characters. Such lines were truncated, leading to shortened
2968 data being returned. It should now handle lines of any length.
2970 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2971 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2974 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2975 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2976 no longer happens in comment lines. A list of replacements is now placed
2977 at the head of all of the source files, except those whose only change is
2978 to replace PERL_COMMAND in the very first #! line.
2980 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2981 messages on the queue, with a bottom-up merge sort, using code contributed
2982 by Michael Haardt. This should make operations like -bp somewhat faster on
2983 large queues. It won't affect queue runners, except when queue_run_in_order
2986 51. Installed eximstats 1.31 in the distribution.
2988 52. Added support for SRV lookups to the dnslookup router.
2990 53. If an ACL referred to $message_body or $message_body_end, the value was not
2991 reset for any messages that followed in the same SMTP session.
2993 54. The store-handling optimization for building very long strings was not
2994 differentiating between the different store pools. I don't think this
2995 actually made any difference in practice, but I've tidied it.
2997 55. While running the routers to verify a sender address, $sender_address
2998 was still set to the sender address. This is wrong, because when routing to
2999 send a bounce to the sender, it would be empty. Therefore, I have changed
3000 it so that, while verifying a sender address, $sender_address is set to <>.
3001 (There is no change to what happens when verifying a recipient address.)
3003 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3004 target A or AAAA records (if not already returned) without resetting the
3005 qualify_single or search_parents options of the DNS resolver. These are
3006 inappropriate in this case because the targets of MX and SRV records must
3007 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3008 target that, when qualified, matched something in the local domain. These
3009 two options are now turned off when doing these lookups.
3011 57. It seems that at least some releases of Reiserfs (which does not have the
3012 concept of a fixed number of inodes) returns zero and not -1 for the
3013 number of available inodes. This interacted badly with check_spool_inodes,
3014 which assumed that -1 was the "no such thing" setting. What I have done is
3015 to check that the total number of inodes is greater than zero before doing
3016 the test of how many are available.
3018 58. When a "warn" ACL statement has a log_message modifier, the message is
3019 remembered, and not repeated. This is to avoid a lot of repetition when a
3020 message has many recipients that cause the same warning to be written.
3021 Howewer, Exim was preserving the list of already written lines for an
3022 entire SMTP session, which doesn't seem right. The memory is now reset if a
3023 new message is started.
3025 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3026 debugging output unless log_rewrite was also set.
3028 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3029 of (int)(handle) when we know that handle contains (void *)(-1).
3031 61. The Exim daemon panic-logs an error return when it closes the incoming
3032 connection. However "connection reset by peer" seems to be common, and
3033 isn't really an error worthy of noting specially, so that particular error
3036 62. When Exim is trying to find all the local interfaces, it used to panic and
3037 die if the ioctl to get the interface flags failed. However, it seems that
3038 on at least one OS (Solaris 9) it is possible to have an interface that is
3039 included in the list of interfaces, but for which you get a failure error
3040 for this call. This happens when the interface is not "plumbed" into a
3041 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3042 failure of the "get flags" call assumes that the interface is down.
3044 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3045 makes life easier for people who are doing arithmetic on fields extracted
3046 from dates, where you often get leading zeros that should not be
3047 interpreted as octal.
3049 64. Added qualify_domain to the redirect router, to override the global
3052 65. If a pathologically long header line contained very many addresses (the
3053 report of this problem mentioned 10 000) and each of them was rewritten,
3054 Exim could use up a very large amount of memory. (It kept on making new
3055 copies of the header line as it rewrote, and never released the old ones.)
3056 At the expense of a bit more processing, the header rewriting function has
3057 been changed so that it no longer eats memory in this way.
3059 66. The generation of the Received: header has been moved from the time that a
3060 message starts to be received, to the time that it finishes. The timestamp
3061 in the Received: header should now be very close to that of the <= log
3062 line. There are two side-effects of this change:
3064 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3065 logged header lines no longer include the local Received: line, because
3066 it has not yet been created. The same applies to a copy of the message
3067 that is returned to a non-SMTP sender when a message is rejected.
3069 (b) When a filter file is tested using -bf, no additional Received: header
3070 is added to the test message. After some thought, I decided that this
3073 This change does not affect the value of $received_for. It is still set
3074 after address rewriting, but before local_scan() is called.
3076 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3078 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3079 gave an unhelpful panic error message, and a defer error. I have managed to
3080 change this behaviour so that it now rejects any supplied certificate,
3081 which seems right, as the list of acceptable certificates is empty.
3083 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3084 gave an unhelpful defer error. I have not managed to make this reject any
3085 supplied certificates, but the error message it gives is "no certificate
3086 supplied", which is not helpful.
3088 70. exigrep's output now also includes lines that are not associated with any
3089 message, but which match the given pattern. Implemented by a patch from
3090 Martin Sluka, which also tidied up the Perl a bit.
3092 71. Recipient callout verification, like sender verification, was using <> in
3093 the MAIL FROM command. This isn't really the right thing, since the actual
3094 sender may affect whether the remote host accepts the recipient or not. I
3095 have changed it to use the actual sender in the callout; this means that
3096 the cache record is now keyed on a recipient/sender pair, not just the
3097 recipient address. There doesn't seem to be a real danger of callout loops,
3098 since a callout by the remote host to check the sender would use <>.
3099 [SEE ABOVE: changed after hitting problems.]
3101 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3102 temporary errors. However, in the case of such a code being given after
3103 the end of a data transmission (i.e. after ".") Exim was failing to write
3104 a retry record for the message. (Yes, there was some broken host that was
3105 actually sending 8xx at this point.)
3107 73. An unknown lookup type in a host list could cause Exim to panic-die when
3108 the list was checked. (An example that provoked this was putting <; in the
3109 middle of a list instead of at the start.) If this happened during a DATA
3110 ACL check, a -D file could be left lying around. This kind of configuration
3111 error no longer causes Exim to die; instead it causes a defer errror. The
3112 incident is still logged to the main and panic logs.
3114 74. Buglet left over from Exim 3 conversion. The message "too many messages
3115 in one connection" was written to the rejectlog but not the mainlog, except
3116 when address rewriting (yes!) was being logged.
3118 75. Added write_rejectlog option.
3120 76. When a system filter was run not as root (that is, when system_filter_user
3121 was set), the values of the $n variables were not being returned to the
3122 main process; thus, they were not subsequently available in the $sn
3125 77. Added +return_path_on_delivery log selector.
3127 78. A connection timeout was being treated differently from recipients deferred
3128 when testing hosts_max_try with a message that was older than the host's
3129 retry timeout. (The host should not be counted, thus allowing all hosts to
3130 be tried at least once before bouncing.) This may have been the cause of an
3131 occasionally reported bug whereby a message would remain on the queue
3132 longer than the retry timeout, but would be bounced if a delivery was
3133 forced. I say "may" because I never totally pinned down the problem;
3134 setting up timeout/retry tests is difficult. See also the next item.
3136 79. The ultimate address timeout was not being applied to errors that involved
3137 a combination of host plus message (for example, a timeout on a MAIL
3138 command). When an address resolved to a number of possible hosts, and they
3139 were not all tried for each delivery (e.g. because of hosts_max_try), a
3140 message could remain on the queue longer than the retry timeout.
3142 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3145 81. Fixed an obscure SMTP outgoing bug which required at least the following
3146 conditions: (a) there was another message waiting for the same server;
3147 (b) the server returned 5xx to all RCPT commands in the first message so
3148 that the message was not completed; (c) the server dropped the connection
3149 or gave a negative response to the RSET that Exim sends to abort the
3150 transaction. The observed case was a dropped connection after DATA that had
3151 been sent in pipelining mode. That is, the server had advertised PIPELINING
3152 but was not implementing it correctly. The effect of the bug was incorrect
3153 behaviour, such as trying another host, and this could lead to a crash.
3159 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3160 and daemon.c were passed as pointers to ints; they should have been
3161 pointers to socklen_t variables (which are typically unsigned ints).
3163 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3166 3. Fixed a really odd bug that affected only the testing scheme; patching a
3167 certain fixed string in the binary changed the value of another string that
3168 happened to be identical to the end of the original first string.
3170 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3171 name", it returns that address as the IP address. On some operating
3172 systems (e.g. Solaris), it also passes back the IP address string as the
3173 "host name". However, on others (e.g. Linux), it passes back an empty
3174 string. Exim wasn't checking for this, and was changing the host name to an
3175 empty string, assuming it had been canonicized.
3177 5. Although rare, it is permitted to have more than one PTR record for a given
3178 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3179 all the names associated with an address, because they do in Solaris.
3180 However, it seems that they do not in Linux for data that comes from the
3181 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3182 I found this out when I moved to a new Linux workstation and tried to run
3183 the Exim test suite.
3185 To get round this problem I have changed the code so that it now does its
3186 own call to the DNS to look up PTR records when searching for a host name.
3187 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3188 addresses that are only in /etc/hosts are still found.
3190 This behaviour is, however, controlled by an option called host_lookup_
3191 order, which defaults to "bydns:byaddr". If people want to use the other
3192 order, or indeed, just use one or the other means of lookup, they can
3193 specify it in this variable.
3195 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3196 some operating systems, this comes back from gethostbyaddr() as an empty
3197 string, and this is what Exim used to test for. However, it seems that in
3198 other systems, "." is yielded. Exim now tests for this case too.
3200 7. The values of check_spool_space and check_log_space are now held internally
3201 as a number of kilobytes instead of an absolute number of bytes. If a
3202 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3203 kilobyte. This means that much larger values can be stored.
3205 8. Exim monitor: an attempt to get the action menu when not actually pointing
3206 at a message produces an empty menu entitled "No message selected". This
3207 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3208 no entries in it ("Shell widget menu has zero width and/or height"). So I
3209 have added a single, blank menu entry in this case.
3211 9. Added ${quote_local_part.
3213 10. MIME decoding is now applied to the contents of Subject: header lines when
3216 11. Now that a reference to $sender_host_address automatically causes a reverse
3217 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3218 host lookup before query-style lookups in lists that might use this
3219 variable. This has therefore been abolished, and the "net-" prefix is no
3220 longer necessary for query-style lookups.
3222 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3223 have been a typo for LFLAGS, so it has been changed.
3225 13. The install script calls Exim with "-C /dev/null" in order to find the
3226 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3227 to be output. Howeve, since Exim outputs its version number before the
3228 error, it didn't break the script. It just looked ugly. I fixed this by
3229 always allowing "-C /dev/null" if the caller is root.
3231 14. Ignore overlarge ACL variable number when reading spool file - insurance
3232 against a later release with more variables having written the file.
3234 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3235 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3236 this, as well as the form without the "IPv6" on the front.
3238 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3239 OS/Makefile-Darwin file.
3241 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3242 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3244 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3245 variable. However, some OS use ints for inodes. I've added cast to long int
3246 to get rid of the compiler warning.
3248 19. I had forgotten to lock out "/../" in configuration file names when
3249 ALT_CONFIG_PREFIX was set.
3251 20. Routers used for verification do not need to specify transports. However,
3252 if such a router generated a host list, and callout was configured, Exim
3253 crashed, because it could not find a port number from the (non-existent)
3254 transport. It now assumes port 25 in this circumstance.
3256 21. Added the -t option to exigrep.
3258 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3259 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3260 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3261 them is set, LOOKUP_LSEARCH is forced.
3263 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3265 24. Added sender and host information to the "rejected by local_scan()" log
3266 line; previously there was no indication of these.
3268 25. Added .include_if_exists.
3270 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3271 Linux. It turns out that not all file systems support this. Apparently some
3272 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3273 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3274 not supported on the file descriptor, is now ignored when Exim is trying to
3275 sync a directory. This applies only to Linux.
3277 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3279 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3280 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3283 29. Added support for maildirsize files from supplied patch (modified a bit).
3285 30. The use of :fail: followed by an empty string could lead Exim to respond to
3286 sender verification failures with (e.g.):
3288 550 Verification failed for <xxx>
3289 550 Sender verify failed
3291 where the first response line was missing the '-' that indicates it is not
3292 the final line of the response.
3294 31. The loop for finding the name of the user that called Exim had a hardwired
3295 limit of 10; it now uses the value of finduser_retries, which is used for
3296 all other user lookups.
3298 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3301 33. Exim was neglecting to zero errno before one call of strtol() when
3302 expanding a string and expecting an integer value. On some systems this
3303 resulted in spurious "integer overflow" errors. Also, it was casting the
3304 result into an int without checking.
3306 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3307 did not work. The code looks as if it has *never* worked, though it appears
3308 to have been documented since at least releast 1.62. I have made it work.
3310 35. The "timeout_DNS" error in retry rules, also documented since at least
3311 1.62, also never worked. As it isn't clear exactly what this means, and
3312 clearly it isn't a major issue, I have abolished the feature by treating it
3313 as "timeout", and writing a warning to the main and panic logs.
3315 36. The display of retry rules for -brt wasn't always showing the error code
3318 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3319 timeout_connect_A, timeout_connect_MX.
3321 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3322 to the empty sender.
3324 39. The daemon was not analysing the content of -oX till after it had closed
3325 stderr and disconnected from the controlling terminal. This meant that any
3326 syntax errors were only noted on the panic log, and the return code from
3327 the command was 0. By re-arranging the code a little, I've made the
3328 decoding happen first, so such errors now appear on stderr, and the return
3329 code is 1. However, the actual setting up of the sockets still happens in
3330 the disconnected process, so errors there are still only recorded on the
3333 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3334 connections (as happens on some IP stacks) was logged at start up time as
3335 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3336 it from "IPv6 and IPv4", which means that two separate sockets are being
3339 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3340 says whether AF_INET or AF_INET6 was passed as an argument.
3342 42. Exiwhat output was messed up when time zones were included in log
3345 43. Exiwhat now gives more information about the daemon's listening ports,
3346 and whether -tls-on-connect was used.
3348 44. The "port" option of the smtp transport is now expanded.
3350 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3351 silently ignored. Now an error message is written to the main and panic
3354 46. There's a new ACL modifier called "logwrite" which writes to a log file
3355 as soon as it is encountered.
3357 47. Added $local_user_uid and $local_user_gid at routing time.
3359 48. Exim crashed when trying to verify a sender address that was being
3362 49. Exim was recognizing only a space character after ".include". It now also
3363 recognizes a tab character.
3365 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3366 extracting the relevant information from the specification. The man page no
3367 longer contains scrambled data for the -d option, and I've added a section
3368 at the front about calling Exim under different names.
3370 51. Added "extra_headers" argument to the "mail" command in filter files.
3372 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3375 53. Installed eximstats 1.29.
3377 54. Added transport_filter_timeout as a generic transport option.
3379 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3380 Cc: header lines. This was required by RFC 822, but it not required by RFC
3383 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3384 incoming messages that did not have them. Now it does so only if the
3385 message originates locally, that is, if there is no associated remote host
3386 address. When Resent- header lines are present, this applies to the Resent-
3387 lines rather than the non-Resent- lines.
3389 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3390 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3392 58. Messages for configuration errors now include the name of the main
3393 configuration file - useful now that there may be more than one file in a
3394 list (.included file names were always shown).
3396 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3397 for those rare installations that do not start the daemon as root or run it
3398 setuid root. I've cut out the call to initgroups() if the daemon is not
3401 60. The Exim user and group can now be bound into the binary as text strings
3402 that are looked up at the start of Exim's processing.
3404 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3406 62. Added $mailstore_basename variable.
3408 63. Installed patch to sieve.c from Michael Haardt.
3410 64. When Exim failed to open the panic log after failing to open the main log,
3411 the original message it was trying to log was written to stderr and debug
3412 output, but if they were not available (the usual case in production), it
3413 was lost. Now it is written to syslog before the two lines that record the
3414 failures to open the logs.
3416 65. Users' Exim filters run in subprocesses under the user's uid. It is
3417 possible for a "deliver" command or an alias in a "personal" command to
3418 provoke an address rewrite. If logging of address rewriting is configured,
3419 this fails because the process is not running as root or exim. There may be
3420 a better way of dealing with this, but for the moment (because 4.30 needs
3421 to be released), I have disabled address rewrite logging when running a
3422 filter in a non-root, non-exim process.
3428 1. The buildconfig auxiliary program wasn't quoting the value set for
3429 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3430 not defined. This bug was masked in 4.22 by the effect that was fixed in
3433 2. Some messages that were rejected after a message id was allocated were
3434 shown as "incomplete" by exigrep. It no longer does this for messages that
3435 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3437 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3438 have allow_domain_literals set, the ID did not get logged in the <= line.
3439 Domain literals are now always recognized in Message-ID: header lines.
3441 4. The first argument for a ${extract expansion item is the key name or field
3442 number. Leading and trailing spaces in this item were not being ignored,
3443 causing some misleading effects.
3445 5. When deliver_drop_privilege was set, single queue runner processes started
3446 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3447 same command in the process it spins off) were not dropping privilege.
3449 6. When the daemon running as "exim" started a queue runner, it always
3450 re-executed Exim in the spun-off process. This is a waste of effort when
3451 deliver_drop_privilege is set. The new process now just calls the
3452 queue-runner function directly.
3458 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3461 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3462 ignored. Though the use of -oP was forcing the writing of a pid file, it
3463 was always written to the default place.
3465 3. If the message "no IP address found for host xxxx" is generated during
3466 incoming verification, it is now followed by identification of the incoming
3467 connection (so you can more easily find what provoked it).
3469 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3471 5. Added some features to "harden" Exim a bit more against certain attacks:
3473 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3474 be put in Local/Makefile. This is like the never_users runtime option,
3475 but it cannot be overridden. The default setting is "root".
3477 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3478 prefix string with which any file named in a -C command line option
3481 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3482 is retained for -C and -D only if the caller of Exim is root. Without
3483 it, the exim user may also use -C and -D and retain privilege.
3485 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3486 command line option is disabled.
3488 6. Macro names set by the -D option must start with an upper case letter, just
3489 like macro names defined in the configuration file.
3491 7. Added "dereference=" facility to LDAP.
3493 8. Two instances of the typo "uknown" in the source files are fixed.
3495 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3496 the Configure-Makefile script screwed up while processing it.
3498 10. Incorporated PCRE 4.4.
3500 11. The SMTP synchronization check was not operating right at the start of an
3501 SMTP session. For example, it could not catch a HELO sent before the client
3502 waited for the greeting. There is now a check for outstanding input at the
3503 point when the greeting is written. Because of the duplex, asynchronous
3504 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3505 way, but not yet received, when the check is performed.
3507 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3508 on TCP/IP sockets, because this apparently causes some broken clients to
3511 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3512 unchanged) from the Cygwin maintainer.
3514 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3515 was supported instead of testing for mbx. Effectively a typo.
3517 15. The spa authenticator server code was not checking that the input it
3518 received was valid base64.
3520 16. The debug output line for the "set" modifier in ACLs was not showing the
3521 name of the variable that was being set.
3523 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3525 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3526 lookup on its own. Something else was needed to trigger the lookup. For
3527 example, a match in host_lookup or the need for a host name in a host list.
3528 Now, if $sender_host_name is referenced and the host name has not yet been
3529 looked up, a lookup is performed. If the lookup fails, the variable remains
3530 empty, and $host_lookup_failed is set to "1".
3532 19. Added "eqi" as a case-independent comparison operator.
3534 20. The saslauthd authentication condition could segfault if neither service
3535 nor realm was specified.
3537 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3538 error message that was logged was misleading, and incoming SMTP
3539 connections were dropped. The message is now more accurate, and temporary
3540 errors are given to SMTP connections.
3542 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3543 (except RSET) with a 421 error, until QUIT is received. However, it was
3544 failing to send a response to QUIT.
3546 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3547 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3548 ACL is now run after the helo_try_verify_hosts code.
3550 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3551 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3552 case-independent, so other case variants are also recognized). Apparently
3553 some systems use these upper case variants.
3555 25. If more than two messages were waiting for the same host, and a transport
3556 filter was specified for the transport, Exim sent two messages over the
3557 same TCP/IP connection, and then failed with "socket operation on non-
3558 socket" when it tried to send the third.
3560 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3562 27. The extern definition of crypt16() in expand.c was not being excluded when
3563 the OS had its own crypt16() function.
3565 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3566 as a preferred synonym for return_size_limit, both as an option and as an
3569 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3571 30. Changed the default configuration ACL to relax the local part checking rule
3572 for addresses that are not in any local domains. For these addresses,
3573 slashes and pipe symbols are allowed within local parts, but the sequence
3574 /../ is explicitly forbidden.
3576 31. SPA server authentication was not clearing the challenge buffer before
3579 32. log_message in a "warn" ACL statement was writing to the reject log as
3580 well as to the main log, which contradicts the documentation and doesn't
3581 seem right (because no rejection is happening). So I have stopped it.
3583 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3584 However, I am unable to do any testing of this.
3586 34. Fixed an infelicity in the appendfile transport. When checking directories
3587 for a mailbox, to see if any needed to be created, it was accidentally
3588 using path names with one or more superfluous leading slashes; tracing
3589 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3591 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3592 contents are added to the log line that is written for every discarded
3593 recipient. (Previously a log_message setting was ignored.)
3595 36. The ${quote: operator now quotes the string if it is empty.
3597 37. The install script runs exim in order to find its version number. If for
3598 some reason other than non-existence or emptiness, which it checks, it
3599 could not run './exim', it was installing it with an empty version number,
3600 i.e. as "exim-". This error state is now caught, and the installation is
3603 38. An argument was missing from the function that creates an error message
3604 when Exim fails to connect to the socket for saslauthd authentication.
3605 This could cause Exim to crash, or give a corrupted message.
3607 39. Added isip, isip4, and isip6 to ${if conditions.
3609 40. The ACL variables $acl_xx are now saved with the message, and can be
3610 accessed later in routers, transports, and filters.
3612 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3613 strings in the file are not string-expanded.
3615 42. If a MAIL command specified a SIZE value that was too large to fit into an
3616 int variable, the check against message_size_limit failed. Such values are
3617 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3618 day this will have to be increased, but I don't think I want to be around
3619 when emails are that large.
3626 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3627 iconv() is not standard in FreeBSD.
3629 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3630 IPv6 enabled. The observed symptom was a segmentation fault on return from
3631 the function os_common_find_running_interfaces() in src/os.c.
3633 3. In the check_special_case() function in daemon.c I had used "errno" as an
3634 argument name, which causes warnings on some systems. This was basically a
3635 typo, since it was named "eno" in the comments!
3637 4. The code that waits for the clock to tick (at a resolution of some fraction
3638 of a second) so as to ensure message-id uniqueness was always waiting for
3639 at least one whole tick, when it could have waited for less. [This is
3640 almost certainly not relevant at current processor speeds, where it is
3641 unlikely to ever wait at all. But we try to future-proof.]
3643 5. The function that sleeps for a time interval that includes fractions of a
3644 second contained a race. It did not block SIGALRM between setting the
3645 timer, and suspending (a couple of lines later). If the interval was short
3646 and the sigsuspend() was delayed until after it had expired, the suspension
3647 never ended. On busy systems this could lead to processes getting stuck for
3650 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3651 process, before it forks any delivery processes. The open lookup caching
3652 mechanism meant that the open file or database connection was passed into
3653 the delivery process. The problem was that delivery processes always tidy
3654 up cached lookup data. This could cause a problem for the next delivery
3655 process started by the queue runner, because the external queue runner
3656 process does not know about the closure. So the next delivery process
3657 still has data in the lookup cache. In the case of a file lookup, there was
3658 no problem because closing a file descriptor in a subprocess doesn't affect
3659 the parent. However, if the lookup was caching a connection to a database,
3660 the connection was closed, and the second delivery process was likely to
3661 see errors such as "PGSQL: query failed: server closed the connection
3662 unexpectedly". The problem has been fixed by closing all cached lookups
3663 in a queue runner before running a delivery process.
3665 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3666 seem to have the "const" qualifier which it has on other OS. I've
3669 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3670 *of the same type* (client or server) with the same public name that an
3671 error should be diagnosed.
3673 9. When Exim looked up a host name for an IP address, but failed to find the
3674 original IP address when looking up the host name (a safety check), it
3675 output the message "<ip address> does not match any IP for NULL", which was
3676 confusing, to say the least. The bug was that the host name should have
3677 appeared instead of "NULL".
3679 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3680 user that is built into the binary, and the -C or -D options is used, root
3681 privilege is dropped before the configuration file is read. In addition,
3682 logging is switched to stderr instead of the normal log files. If the
3683 configuration then re-defines the Exim user, the unprivileged environment
3684 is probably not what is expected, so Exim logs a panic warning message (but
3687 However, if deliver_drop_privilege is set, the unprivileged state may well
3688 be exactly what is intended, so the warning has been cut out in that case,
3689 and Exim is allowed to try to write to its normal log files.
3695 1. smtp_return_error_details was not giving details for temporary sender
3696 or receiver verification errors.
3698 2. Diagnose a configuration error if two authenticators have the same public
3701 3. Exim used not to create the message log file for a message until the first
3702 delivery attempt. This could be confusing when incoming messages were held
3703 for policy or load reasons. The message log file is now created at the time
3704 the message is received, and an initial "Received" line is written to it.
3706 4. The automatically generated man page for command line options had a minor
3707 bug that caused no ill effects; however, a more serious problem was that
3708 the procedure for building the man page automatically didn't always
3709 operate. Consequently, release 4.20 contains an out-of-date version. This
3710 shouldn't happen again.
3712 5. When building Exim with embedded Perl support, the script that builds the
3713 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3714 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3716 6. The freeze_tell option was not being used for messages that were frozen on
3717 arrival, either by an ACL or by local_scan().
3719 7. Added the smtp_incomplete_transaction log selector.
3721 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3722 was accepting AUTH without a new EHLO.
3724 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3725 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3728 10. Logging of TCP/IP connections (when configured) now happens in the main
3729 daemon process instead of the child process, so that the TCP/IP connection
3730 count is more accurate (but it can never be perfect).
3732 11. The use of "drop" in a nested ACL was not being handled correctly in the
3733 outer ACL. Now, if condition failure induced by the nested "drop" causes
3734 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3735 or "require"), the connection is dropped.
3737 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3738 that yield "discard" can now be used with an "accept" or a "discard" verb,
3739 but an error is generated for any others (because I can't see a useful way
3740 to define what should happen).
3742 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3743 are now processed in the same way as lines in the Exim configuration file.
3744 In particular, continuation lines are supported.
3746 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3748 15. Added -ti meaning -t -i.
3750 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3751 domains, and warn by logging if others are found.
3753 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3754 returned from GIFCONF when Exim is trying to find the list of interfaces on
3755 a host. The code in os.c has been modified to copy each ifreq to an aligned
3756 structure in all cases.
3758 Also, in some cases, the returned ifreq's were being copied to a 'struct
3759 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3760 means the last couple of bytes of an IPv6 address could be chopped if the
3761 ifreq contained only a normal sockaddr (14 bytes storage).
3763 18. Named domain lists were not supported in the hosts_treat_as_local option.
3764 An entry such as +xxxx was not recognized, and was treated as a literal
3767 19. Ensure that header lines added by a DATA ACL are included in the reject log
3768 if the ACL subsequently rejects the message.
3770 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3771 MD5 (which is deprecated).
3773 21. When testing a filter file using -bf, Exim was writing a message when it
3774 took the sender from a "From " line in the message, but it was not doing so
3775 when it took $return_path from a Return-Path: header line. It now does.
3777 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3778 with a valid header line field name (a series of printing characters
3779 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3781 23. Changed "disc" in the source to "disk" to conform to the documentation and
3782 the book and for uniformity.
3784 24. Ignore Sendmail's -Ooption=value command line item.
3786 25. When execve() failed while trying to run a command in a pipe transport,
3787 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3788 could be confused with a return value of 69 from the command itself. This
3789 has been changed to 127, the value the shell returns if it is asked to run
3790 a non-existent command. The wording for the related log line suggests a
3791 non-existent command as the problem.
3793 26. If received_header_text expands to an empty string, do not add a Received:
3794 header line to the message. (Well, it adds a token one on the spool, but
3795 marks it "old" so that it doesn't get used or transmitted.)
3797 27. Installed eximstats 1.28 (addition of -nt option).
3799 28. There was no check for failure on the call to getsockname() in the daemon
3800 code. This can fail if there is a shortage of resources on the system, with
3801 ENOMEM, for example. A temporary error is now given on failure.
3803 29. Contrary to the C standard, it seems that in some environments, the
3804 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3805 program. Exim now does this explicitly; it affects the formatting of
3806 timestamps using strftime().
3808 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3809 complaints. I've now initialized all the variables, to avoid this.
3811 32. Header lines added by a system filter were not being "seen" during
3812 transport-time rewrites.
3814 33. The info_callback() function passed to OpenSSL is set up with type void
3815 (*)(SSL *, int, int), as described somewhere. However, when calling the
3816 function (actually a macro) that sets it up, the type void(*)() is
3817 expected. I've put in a cast to prevent warnings from picky compilers.
3819 34. If a DNS black list lookup found a CNAME record, but there were no A
3820 records associated with the domain it pointed at, Exim crashed.
3822 35. If a DNS black list lookup returned more than one A record, Exim ignored
3823 all but the first. It now scans all returned addresses if a particular IP
3824 value is being sought. In this situation, the contents of the
3825 $dnslist_value variable are a list of all the addresses, separated by a
3828 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3829 to accept a host name and all its aliases if the forward lookup for any of
3830 them yielded the IP address of the incoming connection. Now it accepts only
3831 those names whose forward lookup yields the correct IP address. Any other
3832 names are discarded. This closes a loophole whereby a rogue DNS
3833 administrator could create reverse DNS records to break through a
3834 wildcarded host restriction in an ACL.
3836 37. If a user filter or a system filter that ran in a subprocess used any of
3837 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3838 the wrong values were passed to the pipe command ($thisaddress had the
3839 value of $0, $0 had the value of $1, etc). This bug was introduced by
3840 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3842 38. Improved the line breaking for long SMTP error messages from ACLs.
3843 Previously, if there was no break point between 40 and 75 characters, Exim
3844 left the rest of the message alone. Two changes have been made: (a) I've
3845 reduced the minimum length to 35 characters; (b) if it can't find a break
3846 point between 35 and 75 characters, it looks ahead and uses the first one
3847 that it finds. This may give the occasional overlong line, but at least the
3848 remaining text gets split now.
3850 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3851 file descriptors might be low, and that setting 1000 would always raise it.
3852 It turns out that in some environments, the limit is already over 1000 and
3853 that lowering it causes trouble. So now Exim takes care not to decrease it.
3855 40. When delivering a message, the value of $return_path is set to $sender_
3856 address at the start of routing (routers may change the value). By an
3857 oversight, this default was not being set up when an address was tested by
3858 -bt or -bv, which affected the outcome if any router or filter referred to
3861 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3862 log only when "message" or "log_message" are set. However, if one of the
3863 conditions was an address verification, or a call to a nested ACL, the
3864 messages generated by the underlying test were being passed through. This
3865 no longer happens. The underlying message is available in $acl_verify_
3866 message for both "message" and "log_message" expansions, so it can be
3867 passed through if needed.
3869 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3870 new expansion $bh_ to give the encoded byte string without charset
3871 translation. Translation happens only if iconv() is available; HAVE_ICONV
3872 indicates this at build time. HEADERS_CHARSET gives the charset to
3873 translate to; headers_charset can change it in the configuration, and
3874 "headers charset" can change it in an individual filter file.
3876 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3877 that creates RFC 2047 encoded "words" labels them as that charset instead
3878 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3879 expansion operator; (ii) when Exim creates a From: line for a local
3880 message; (iii) when a header line is rewritten to include a "phrase" part.
3882 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3883 buggy, causing it to skip the first lines of messages whose message ID
3884 ended in 'D'. This would not have bitten before Exim release 4.14, because
3885 message IDs were unlikely to end in 'D' before then. The effect was to have
3886 incorrect size information for certain domains.
3888 45. #include "config.h" was missing at the start of the crypt16.c module. This
3889 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3892 46. If there was a timeout during a "random" callout check, Exim treated it as
3893 a failure of the random address, and carried on sending RSET and the real
3894 address. If the delay was just some slowness somewhere, the response to the
3895 original RCPT would be taken as a response to RSET and so on, causing
3896 mayhem of various kinds.
3898 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3899 when I implemented it. It didn't allow for the fact that some option values
3900 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3901 the right test for positive values.
3903 48. Domain names in DNS records are case-independent. Exim always looks them up
3904 in lower case. Some resolvers return domain names in exactly the case they
3905 appear in the zone file, that is, they may contain uppercase letters. Not
3906 all resolvers do this - some return always lower case. Exim was treating a
3907 change of case by a resolver as a change of domain, similar to a widening
3908 of a domain abbreviation. This triggered its re-routing code and so it was
3909 trying to route what was effectively the same domain again. This normally
3910 caused routing to fail (because the router wouldn't handle the domain
3911 twice). Now Exim checks for this case specially, and just changes the
3912 casing of the domain that it ultimately uses when it transmits the message
3915 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3918 50. If a filter generated a file delivery with a non-absolute name (possible if
3919 no home directory exists for the router), the forbid_file option was not
3922 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3923 the existing equality matching.
3925 52. Exim was using ints instead of ino_t variables in some places where it was
3926 dealing with inode numbers.
3928 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3929 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3930 called TMPDIR, and if it finds it is different, it changes its value.
3932 54. The smtp_printf() function is now made available to local_scan() so
3933 additional output lines can be written before returning. There is also an
3934 smtp_fflush() function to enable the detection of a dropped connection.
3935 The variables smtp_input and smtp_batched_input are exported to
3938 55. Changed the default runtime configuration: the message "Unknown user"
3939 has been removed from the ACL, and instead placed on the localuser router,
3940 using the cannot_route_message feature. This means that any verification
3941 failures that generate their own messages won't get overridden. Similarly,
3942 the "Unrouteable address" message that was in the ACL for unverifiable
3943 relay addresses has also been removed.
3945 56. Added hosts_avoid_esmtp to the smtp transport.
3947 57. The exicyclog script was not checking for the esoteric option
3948 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3949 will work only if exicyclog is run under the appropriate euid.
3951 58. Following a discussion on the list, the rules by which Exim recognises line
3952 endings on incoming messages have been changed. The -dropcr and drop_cr
3953 options are now no-ops, retained only for backwards compatibility. The
3954 following line terminators are recognized: LF CRLF CR. However, special
3955 processing applies to CR:
3957 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3958 nor a local message in the state where . is a terminator.
3960 (ii) If a bare CR is encountered in a header line, an extra space is added
3961 after the line terminator so as not to end the header. The reasoning
3962 behind this is that bare CRs in header lines are most likely either
3963 to be mistakes, or people trying to play silly games.
3965 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3966 was being incorrectly given as 18 bytes larger than it should have been.
3967 This is a VOB (very old bug).
3969 60. This may never have affected anything current, but just in case it has:
3970 When the local host is found other than at the start of a list of hosts,
3971 the local host, those with the same MX, and any that follow, are discarded.
3972 When the list in question was part of a longer list of hosts, the following
3973 hosts (not currently being processed) were also being discarded. This no
3974 longer happens. I'm not sure if this situation could ever has previously
3977 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3980 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3981 header. This is something that is recommended in a new Internet Draft, and
3982 is something that is documented as being done by Sendmail. There are two
3983 possible values. For messages generated by the autoreply transport, Exim
3986 Auto-Submitted: auto-replied
3988 whereas for all other generated messages (e.g. bounces) it adds
3990 Auto-Submitted: auto-generated
3992 63. The "personal" condition in filters now includes a test for the
3993 Auto-Submitted: header. If it contains the string "auto-" the message it
3994 not considered personal.
3996 64. Added rcpt_include_affixes as a generic transport option.
3998 65. Added queue_only_override (default true).
4000 66. Added the syslog_duplication option.
4002 67. If what should have been the first header line of a message consisted of
4003 a space followed by a colon, Exim was mis-interpreting it as a header line.
4004 It isn't of course - it is syntactically invalid and should therefore be
4005 treated as the start of the message body. The misbehaviour could have
4006 caused a number of strange effects, including loss of data in subsequent
4007 header lines, and spool format errors.
4009 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4010 client host had authenticated. This control can now be exercised by an ACL
4011 for more flexibility.
4013 69. By default, callouts do not happen when testing with -bh. There is now a
4014 variant, -bhc, which does actually run the callout code, including
4015 consulting and updating the callout cache.
4017 70. Added support for saslauthd authentication, courtesy of Alexander
4020 71. If statvfs() failed on the spool or log directories while checking their
4021 size for availability, Exim confusingly gave the error "space shortage".
4022 Furthermore, in debugging mode it crashed with a floating point exception.
4023 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4024 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4025 serious problem, Exim now writes to the main and panic logs when this
4026 happens, with details of the failure. It then refuses to accept the
4027 incoming message, giving the message "spool directory problem" or "log
4028 directory problem" with a 421 code for SMTP messages.
4030 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4031 0, 1, and 2 exist, because some OS complain for execs without them (see
4032 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4033 descriptors. However, the code omitted to check that the open succeeded,
4034 causing mysterious errors if for some reason the permissions on /dev/null
4035 got screwed. Now Exim writes a message to the main and panic logs, and
4036 bombs out if it can't open /dev/null.
4038 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4039 interact so that it is all more flexible. It is supposed to remain
4040 backwards compatible. Also added extra_local_interfaces.
4042 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4043 to bomb out with an assertion failure - to the client this appears as a
4044 connection drop. This problem occurs in the part of the code that was taken
4045 from the Samba project. Fortunately, the assertion is in a very simple
4046 function, so I have fixed this by reproducing the function inline in the
4047 one place where it is called, and arranging for authentication to fail
4048 instead of killing the process with assert().
4050 75. The SPA client code was not working when the server requested OEM rather
4051 than Unicode encoding.
4053 76. Added code to make require_files with a specific uid setting more usable in
4054 the case where statting the file as root fails - usually a non-root-mounted
4055 NFS file system. When this happens and the failure is EACCES, Exim now
4056 forks a subprocess and does the per-uid checking as the relevant uid.
4058 77. Added process_log_path.
4060 78. If log_file_path was not explicitly set, a setting of check_log_space or
4061 check_log_inodes was ignored.
4063 79. If a space check for the spool or log partitions fails, the incident is now
4064 logged. Of course, in the latter case the data may get lost...
4066 80. Added the %p formatting code to string_format() so that it can be used to
4067 print addresses in debug_print(). Adjusted all the address printing in the
4068 debugging in store.c to use %p rather than %d.
4070 81. There was a concern that a line of code in smtp_in.c could overflow a
4071 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4072 initially expressed, the concern was not well-founded, because trailing
4073 spaces are removed early. However, if the trailing spaces were followed by
4074 a NULL, they did not get removed, so the overflow was possible. Two fixes
4077 (a) I re-wrote the offending code in a cleaner fashion.
4078 (b) If an incoming SMTP command contains a NULL character, it is rejected
4081 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4082 runs initgroups(), so that if the Exim user is in any additional groups,
4083 they will be used during message reception.
4089 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob