1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.380 2006/07/21 16:48:42 jetmore 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.
114 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
115 but not after initializing Perl.
117 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
118 output them only if debugging. By default they are written stderr,
119 apparently, which is not desirable.
121 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
124 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
127 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
133 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
134 other effects) broke the use of negated acl sub-conditions.
136 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
139 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
140 "Deny" causes Exim to reject the incoming connection with a 554 error.
141 Unfortunately, if there is a major crisis, such as a disk failure,
142 tcp-wrappers gives "deny", whereas what one would like would be some
143 kind of temporary error. A kludge has been added to help with this.
144 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
145 554 error is used if errno is still zero or contains ENOENT (which occurs
146 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
149 PH/03 Add -lutil to the default FreeBSD LIBS setting.
151 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
152 errors. Otherwise a message that provokes a temporary error (when other
153 messages do not) can cause a whole host to time out.
155 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
156 addresses were routed directly to files or pipes from a redirect router.
157 File deliveries just didn't batch; pipe deliveries might have suffered
160 PH/06 A failure to get a lock for a hints database would erroneously always say
161 "Failed to get write lock", even when it was really a read lock.
163 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
164 of 0600. This has been changed to use the value of the lockfile_mode
165 option (which defaults to 0600).
167 PH/08 Applied small patch from the Sieve maintainer.
169 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
170 folder from quota calculations, a direct delivery into this folder messed
171 up the contents of the maildirsize file. This was because the regex was
172 used only to exclude .Trash (or whatever) when the size of the mailbox
173 was calculated. There was no check that a delivery was happening into an
174 excluded directory. This bug has been fixed by ignoring all quota
175 processing for deliveries into excluded directories.
177 PH/10 Added the maildirfolder_create_regex option to appendfile.
183 PH/01 The code for finding all the local interface addresses on a FreeBSD
184 system running IPv6 was broken. This may well have applied to all BSD
185 systems, as well as to others that have similar system calls. The broken
186 code found IPv4 interfaces correctly, but gave incorrect values for the
187 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
188 that it would not match correctly against @[] and not recognize the IPv6
191 PH/02 The ipliteral router was not recognizing addresses of the form user@
192 [ipv6:....] because it didn't know about the "ipv6:" prefix.
194 PH/03 Added disable_ipv6.
196 PH/04 Changed $reply_address to use the raw form of the headers instead of the
197 decoded form, because it is most often used to construct To: headers
198 lines in autoreplies, and the decoded form may well be syntactically
199 invalid. However, $reply_address has leading white space removed, and all
200 newlines turned into spaces so that the autoreply transport does not
203 PH/05 If group was specified without a user on a router, and no group or user
204 was specified on a transport, the group from the router was ignored.
206 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
207 for visible compile-time settings that can be used to change these
208 numbers, for those that want even more. Backwards compatibility with old
209 spool files has been maintained. However, going back to a previous Exim
210 release will lost any variables that are in spool files.
212 PH/07 Two small changes when running in the test harness: increase delay when
213 passing a TCP/IP connection to a new process, in case the original
214 process has to generate a bounce, and remove special handling of
215 127.0.0.2 (sic), which is no longer necessary.
217 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
218 be the same on different OS.
220 PH/09 Moved a debug statement in filter processing to avoid a race problem when
223 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
224 whether --show-vars was specified or not
226 JJ/02 exipick: Added support for new ACL variable spool format introduced
229 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
230 syntactically invalid From: or Reply-to: line, and a filter used this to
231 generate an autoreply, and therefore failed to obtain an address for the
232 autoreply, Exim could try to deliver to a non-existent relative file
233 name, causing unrelated and misleading errors. What now happens is that
234 it logs this as a hard delivery error, but does not attempt to create a
237 PH/11 The exinext utility has a -C option for testing purposes, but although
238 the given file was scanned by exinext itself; it wasn't being passed on
239 when Exim was called.
241 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
242 an end-of-file indication when reading a command response.
244 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
245 compiled. In many other places in Exim, IPv6 addresses are always
246 recognized, so I have changed this. It also means that IPv4 domain
247 literals of the form [IPV4:n.n.n.n] are now always recognized.
249 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
250 used if the router is not running as root, for example, when verifying at
251 ACL time, or when using -bh. The debugging output from this situation was
252 non-existent - all you got was a failure to exec. I have made two
255 (a) Failures to set uid/gid, the current directory, or a process leader
256 in a subprocess such as that created by queryprogram now generate
257 suitable debugging ouput when -d is set.
259 (b) The queryprogram router detects when it is not running as root,
260 outputs suitable debugging information if -d is set, and then runs
261 the subprocess without attempting to change uid/gid.
263 PH/15 Minor change to Makefile for building test_host (undocumented testing
266 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
267 additional section of a DNS packet that returns MX or SRV records.
268 Instead, it always explicitly searches for A/AAAA records. This avoids
269 major problems that occur when a DNS server includes only records of one
270 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
271 fixed another bug: if SRV records were looked up and the corresponding
272 address records were *not* found in the additional section, the port
273 values from the SRV records were lost.
275 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
276 using the correct key (the original address) when searching the retry
277 rules in order to find which one to use for generating the retry hint.
279 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
280 adding the default one. Similarly, if it contains a Reply-To: header, the
281 errors_reply_to option, if set, is not used.
283 PH/19 When calculating a retry time, Exim used to measure the "time since
284 failure" by looking at the "first failed" field in the retry record. Now
285 it does not use this if it is later than than the arrival time of the
286 message. Instead it uses the arrival time. This makes for better
287 behaviour in cases where some deliveries succeed, thus re-setting the
288 "first failed" field. An example is a quota failure for a huge message
289 when small messages continue to be delivered. Without this change, the
290 "time since failure" will always be short, possible causing more frequent
291 delivery attempts for the huge message than are intended.
292 [Note: This change was subsequently modified - see PH/04 for 4.62.]
294 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
295 $1, $2, $3) because the numerical variables can be reset during some
296 expansion items (e.g. "match"), thereby losing the authentication data.
298 PH/21 Make -bV show the size of off_t variables so that the test suite can
299 decide whether to run tests for quotas > 2G.
301 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
302 mailbox_size, and mailbox_filecount in the appendfile transport. If a
303 filecount value is greater than 2G or if a quota value is greater than 2G
304 on a system where the size of off_t is not greater than 4, a panic error
307 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
308 never match. The debug and -bh output now contains an explicit error
309 message indicating a malformed IPv4 address or mask.
311 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
312 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
315 PH/25 Do not write to syslog when running in the test harness. The only
316 occasion when this arises is a failure to open the main or panic logs
317 (for which there is an explicit test).
319 PH/26 Added the /no_tell option to "control=freeze".
321 PH/27 If a host name lookup failed very early in a connection, for example, if
322 the IP address matched host_lookup and the reverse lookup yielded a name
323 that did not have a forward lookup, an error message of the form "no IP
324 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
325 could be logged. Now it outputs the IP address instead of "NULL".
327 PH/28 An enabling patch from MH: add new function child_open_exim2() which
328 allows the sender and the authenticated sender to be set when
329 submitting a message from within Exim. Since child_open_exim() is
330 documented for local_scan(), the new function should be too.
332 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
333 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
334 results in an empty string is now treated as unset.
336 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
338 PH/31 Added sender_verify_fail logging option.
340 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
341 needed by RFC 822 but not by RFC 2822 was commented out. I have now
342 tidied the source and removed it altogether.
344 PH/33 When a queue run was abandoned because the load average was too high, a
345 log line was always written; now it is written only if the queue_run log
346 selector is set. In addition, the log line for abandonment now contains
347 information about the queue run such as the pid. This is always present
348 in "start" and "stop" lines but was omitted from the "abandon" line.
350 PH/34 Omit spaces between a header name and the colon in the error message that
351 is given when verify = headers_syntax fails (if there are lots of them,
352 the message gets confusing).
354 PH/35 Change the default for dns_check_names_pattern to allow slashes within
355 names, as there are now some PTR records that contain slashes. This check
356 is only to protect against broken name servers that fall over on strange
357 characters, so the fact that it applies to all lookups doesn't matter.
359 PH/36 Now that the new test suite is complete, we can remove some of the
360 special code in Exim that was needed for the old test suite. For example,
361 sorting DNS records because real resolvers return them in an arbitrary
362 order. The new test suite's fake resolver always returns records in the
365 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
366 bounces) except when queue_only is set, to avoid logging races between
367 the different processes.
369 PH/38 Panic-die if .include specifies a non-absolute path.
371 PH/39 A tweak to the "H" retry rule from its user.
373 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
374 a label. They prevented compilation on older perls.
376 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
377 a warning to be raised on newish perls.
379 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
380 on queue. Changes to match documented behaviour of showing count of
381 messages matching specified criteria.
383 PH/40 Changed the default ident timeout from 30s to 5s.
385 PH/41 Added support for the use of login_cap features, on those BSD systems
386 that have them, for controlling the resources used by pipe deliveries.
388 PH/42 The content-scanning code uses fopen() to create files in which to put
389 message data. Previously it was not paying any attention to the mode of
390 the files. Exim runs with umask(0) because the rest of the code creates
391 files with open(), and sets the required mode explicitly. Thus, these
392 files were ending up world-writeable. This was not a big issue, because,
393 being within the spool directory, they were not world-accessible. I have
394 created a function called modefopen, which takes an additional mode
395 argument. It sets umask(777), creates the file, chmods it to the required
396 mode, then resets the umask. All the relevant calls to fopen() in the
397 content scanning code have been changed to use this function.
399 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
400 to 24 hours. This avoids potential overflow problems when processing G
401 and H retry rules. I suspect nobody ever tinkers with this value.
403 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
405 PH/45 When the plaintext authenticator is running as a client, the server's
406 challenges are checked to ensure they are valid base64 strings. By
407 default, the authentication attempt is cancelled if an invalid string is
408 received. Setting client_ignore_invalid_base64 true ignores these errors.
409 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
410 they are received. Thus, the responses can be made to depend on the
411 challenges. If an invalid string is ignored, an empty string is placed in
414 PH/46 Messages that are created by the autoreply transport now contains a
415 References: header, in accordance with RFCs 2822 and 3834.
417 PH/47 Added authenticated_sender_force to the smtp transport.
419 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
421 PH/49 Installed latest patch from the Sieve maintainer.
423 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
424 was also set, the appendfile transport was unnecessarily scanning a
425 directory of message files (e.g. for maildir delivery) to find the count
426 of files (along with the size), even though it did not need this
427 information. It now does the scan only if it needs to find either the
428 size of the count of files.
430 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
432 PH/52 Two bugs concerned with error handling when the smtp transport is
435 (i) Exim was not creating retry information for temporary errors given
436 for individual recipients after the DATA command when the smtp transport
437 was used in LMTP mode. This meant that they could be retried too
438 frequently, and not timed out correctly.
440 (ii) Exim was setting the flag that allows error details to be returned
441 for LMTP errors on RCPT commands, but not for LMTP errors for individual
442 recipients that were returned after the DATA command.
444 PH/53 This is related to PH/52, but is more general: for any failing address,
445 when detailed error information was permitted to be returned to the
446 sender, but the error was temporary, then after the final timeout, only
447 "retry timeout exceeded" was returned. Now it returns the full error as
448 well as "retry timeout exceeded".
450 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
451 do this, and (what is worse) MTAs that accept it.
453 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
454 will now be deprecated.
456 PH/56 New os.c-cygwin from the Cygwin maintainer.
458 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
459 formats (previously only available in exim formats via -bpr, -bpru,
460 and -bpra. Now also available in native and exiqgrep formats)
462 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
463 with very large, slow to parse queues
465 JJ/08 exipick: added ! as generic prefix to negate any criteria format
467 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
469 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
470 responses to authentication challenges, though it was showing the
471 challenges; (ii) I've removed the CR characters from the debug output for
474 PH/58 Allow for the insertion of a newline as well as a space when a string
475 is turned into more than one encoded-word during RFC 2047 encoding. The
476 Sieve code now uses this.
478 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
479 data_4xx, lost_connection, tls_required.
481 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
482 message was being sent as an SMTP response.
484 PH/61 Add -l and -k options to exicyclog.
486 PH/62 When verifying, if an address was redirected to one new address, so that
487 verification continued, and the new address failed or deferred after
488 having set something in $address_data, the value of $address_data was not
489 passed back to the ACL. This was different to the case when no
490 redirection occurred. The value is now passed back in both cases.
492 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
493 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
494 use login_cap.h, so on its own it isn't the distinguishing feature. The
495 new name refers directly to the setclassresources() function.
497 PH/65 Added configuration files for NetBSD3.
499 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
501 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
502 is preferred over IPv4.
504 PH/68 The bounce_return_message and bounce_return_body options were not being
505 honoured for bounces generated during the reception of non-SMTP messages.
506 In particular, this applied to messages rejected by the ACL. This bug has
507 been fixed. However, if bounce_return_message is true and bounce_return_
508 body is false, the headers that are returned for a non-SMTP message
509 include only those that have been read before the error was detected.
510 (In the case of an ACL rejection, they have all been read.)
512 PH/69 The HTML version of the specification is now built in a directory called
513 spec_html instead of spec.html, because the latter looks like a path with
514 a MIME-type, and this confuses some software.
516 PH/70 Catch two compiler warnings in sieve.c.
518 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
519 function verify_get_ident() calls ip_connect() to connect a socket, but
520 if the "connect()" function timed out, ip_connect() used to close the
521 socket. However, verify_get_ident() also closes the socket later, and in
522 between Exim writes to the log, which may get opened at this point. When
523 the socket was closed in ip_connect(), the log could get the same file
524 descriptor number as the socket. This naturally causes chaos. The fix is
525 not to close the socket in ip_connect(); the socket should be closed by
526 the function that creates it. There was only one place in the code where
527 this was missing, in the iplookup router, which I don't think anybody now
528 uses, but I've fixed it anyway.
530 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
531 well as to direct DNS lookups. Otherwise the handling of names in host
532 lists is inconsistent and therefore confusing.
538 PH/01 Two changes to the default runtime configuration:
540 (1) Move the checks for relay_from_hosts and authenticated clients from
541 after to before the (commented out) DNS black list checks.
543 (2) Add control=submission to the relay_from_hosts and authenticated
544 clients checks, on the grounds that messages accepted by these
545 statements are most likely to be submissions.
547 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
549 (1) Generate an error if the third argument for the ${prvs expansion is
552 (2) Treat a missing third argument of ${prvscheck as if it were an empty
555 (3) Reset the variables that are obtained from the first argument of
556 ${prvscheck and used in the second argument before leaving the code,
557 because their memory is reclaimed, so using them afterwards may do
560 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
561 one (it's much easier than Tom thought :-).
563 (5) Because of (4), we can now allow for the use of $prvscheck_result
564 inside the third argument.
566 PH/03 For some reason, the default setting of PATH when running a command from
567 a pipe transport was just "/usr/bin". I have changed it to
570 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
571 anything to be listed in the output from -bV.
573 PH/05 When a filter generated an autoreply, the entire To: header line was
574 quoted in the delivery log line, like this:
576 => >A.N.Other <ano@some.domain> <original@ddress> ...
578 This has been changed so that it extracts the operative address. There
579 may be more than one such address. If so, they are comma-separated, like
582 => >ano@some.domain,ona@other.domain <original@ddress> ...
584 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
585 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
586 not being looked up in the rDNS to get a host name, Exim was showing the
587 IP address twice in Received: lines, even though the IP addresses were
588 identical. For example:
590 Received: from [1.2.3.4] (helo=[1.2.3.4])
592 However, if the real host name was known, it was omitting the HELO data
593 if it matched the actual IP address. This has been tidied up so that it
594 doesn't show the same IP address twice.
596 PH/07 When both +timestamp and +memory debugging was on, the value given by
597 $tod_xxx expansions could be wrong, because the tod_stamp() function was
598 called by the debug printing, thereby overwriting the timestamp buffer.
599 Debugging no longer uses the tod_stamp() function when +timestamp is set.
601 PH/08 When the original message was included in an autoreply transport, it
602 always said "this is a copy of the message, including all the headers",
603 even if body_only or headers_only was set. It now gives an appropriate
606 PH/09 Applied a patch from the Sieve maintainer which:
608 o fixes some comments
609 o adds the (disabled) notify extension core
610 o adds some debug output for the result of if/elsif tests
611 o points to the current vacation draft in the documentation
612 and documents the missing references header update
616 o fixes a bug in processing the envelope test (when testing
617 multiple envelope elements, the last element determinted the
620 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
621 Electronic Mail") by including:
623 Auto-submitted: auto-generated
625 in the messages that it generates (bounce messages and others, such as
626 warnings). In the case of bounce messages for non-SMTP mesages, there was
627 also a typo: it was using "Auto_submitted" (underscore instead of
628 hyphen). Since every message generated by Exim is necessarily in response
629 to another message, thes have all been changed to:
631 Auto-Submitted: auto-replied
633 in accordance with these statements in the RFC:
635 The auto-replied keyword:
637 - SHOULD be used on messages sent in direct response to another
638 message by an automatic process,
640 - MUST NOT be used on manually-generated messages,
642 - MAY be used on Delivery Status Notifications (DSNs) and Message
643 Disposition Notifications (MDNs),
645 - MUST NOT be used on messages generated by automatic or periodic
646 processes, except for messages which are automatic responses to
649 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
650 to the default Received: header definition.
652 PH/12 Added log selector acl_warn_skipped (default on).
654 PH/13 After a successful wildlsearch lookup, discard the values of numeric
655 variables because (a) they are in the wrong storage pool and (b) even if
656 they were copied, it wouldn't work properly because of the caching.
658 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
659 checking when decoding. Apparently there are clients that generate
660 overlong encoded strings. Why am I not surprised?
662 PH/15 If the first argument of "${if match_address" was not empty, but did not
663 contain an "@" character, Exim crashed. Now it writes a panic log message
664 and treats the condition as false.
666 PH/16 In autoreply, treat an empty string for "once" the same as unset.
668 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
669 extension "envelope-auth". The code is finished and in agreement with
670 other implementations, but there is no documentation so far and in fact,
671 nobody wrote the draft yet. This extension is currently #undef'ed, thus
672 not changing the active code.
674 Print executed "if" and "elsif" statements when debugging is used. This
675 helps a great deal to understand what a filter does.
677 Document more things not specified clearly in RFC3028. I had all this
678 sorted out, when out of a sudden new issues came to my mind. Oops."
680 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
683 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
684 canonical form (as documented). However, after a host name lookup from
685 the IP address, check_host() was doing a simple string comparison with
686 addresses acquired from the DNS when checking that the found name did
687 have the original IP as one of its addresses. Since any found IPv6
688 addresses are likely to be in abbreviated form, the comparison could
689 fail. Luckily, there already exists a function for doing the comparison
690 by converting both addresses to binary, so now that is used instead of
693 PH/20 There was another similar case to PH/19, when a complete host name was
694 given in a host list; looking up its IP address could give an abbreviated
695 form, whereas the current host's name might or might not be abbreviated.
696 The same fix has been applied.
702 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
703 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
706 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
707 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
709 PH/03 Typo: missing ".o" in src/pcre/Makefile.
711 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
712 header line, restrict the check to what is listed in RFCs 2369 and 2929.
713 Also, for "Auto-Submitted", treat anything other than "no" as
714 non-personal, in accordance with RFC 3834. (Previously it treated
715 anything starting "auto-" as non-personal.)
717 TF/01 The control=submission/name=... option had a problem with syntax
718 errors if the name included a slash character. The /name= option
719 now slurps the rest of the string, so it can include any characters
720 but it must come last in the list of options (after /sender_retain
723 PH/05 Some modifications to the interface to the fake nameserver for the new
731 TK/01 Added the "success_on_redirect" address verification option. See
732 NewStuff for rationale and an example.
734 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
736 PH/02 Patch to exigrep to allow it to work on syslog lines.
738 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
739 fread() to skip over the body file's header line, because in Cygwin the
740 header line is locked and is inaccessible.
742 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
743 co-exist for some time) to make it clear that it is the Exim ID that is
744 referenced, not the Message-ID: header line.
746 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
747 string_format() function, because snprintf() does not exist on all
750 PH/06 The use of forbid_filter_existstest now also locks out the use of the
751 ${stat: expansion item.
753 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
754 protocol synchronization error", to keep the pedants happy.
756 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
757 well as for IRIX systems, when gcc is being used. See the host.c source
760 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
762 PH/10 Named domain lists were not working if used in a queue_smtp_domains
765 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
766 transport and to the smtp transport in LMTP mode.
768 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
770 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
771 run a filter in a subprocess. This could lead to confusion in subsequent
772 lookups in the parent process. There should also be a search_tidyup() at
773 the end of the subprocess.
775 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
776 only if the host matched helo_try_verify_hosts, which caused the
777 verification to occur when the EHLO/HELO command was issued. The ACL just
778 tested the remembered result. Now, if a previous verification attempt has
779 not happened, "verify = helo" does it there and then.
781 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
783 TK/03 Fix log output including CR from clamd.
785 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
786 exist provoked a memory error which could cause a segfault.
788 PH/15 Installed PCRE 6.2
790 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
792 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
793 of the problem. Specifically, suggested +O2 rather than +O1 for the
796 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
798 PH/20 If a delivery was routed to a non-standard port by means of an SRV
799 record, the port was not correctly logged when the outgoing_port log
800 selector was set (it logged the transort's default port).
802 PH/21 Added support for host-specific ports to manualroute, queryprogram,
803 fallback_hosts, and "hosts" in the smtp transport.
805 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
806 host errors such as "Connection refused".
808 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
809 authentication with radiusclient 0.4.9:
811 - Error returned from rc_read_config was caught wrongly
812 - Username/password not passed on to radius server due to wrong length.
814 The presumption is that some radiusclient API changes for 4.51/PH/17
815 were not taken care of correctly. The code is still untested by me (my
816 Linux distribution still has 0.3.2 of radiusclient), but it was
817 contributed by a Radius user.
819 PH/24 When doing a callout, the value of $domain wasn't set correctly when
820 expanding the "port" option of the smtp transport.
822 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
823 while reading a MIME header. Thanks to Tom Hughes for a patch.
825 PH/24 Include config.h inside local_scan.h so that configuration settings are
828 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
829 that in an ACL for RCPT (for example), you can examine exactly what was
832 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
833 commands, but it was not correctly comparing the address with the actual
834 client host address. Thus, it would show the EHLO address in Received:
835 header lines when this was not necessary.
837 PH/27 Added the % operator to ${eval:}.
839 PH/28 Exim tries to create and chdir to its spool directory when it starts;
840 it should be ignoring failures (because with -C, for example, it has lost
841 privilege). It wasn't ignoring creation failures other than "already
844 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
847 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
848 because an input line was too long, either on its own, or by virtue of
849 too many continuations, the temporary file was not being removed, and the
850 return code was incorrect.
852 PH/31 Missing "BOOL" in function definition in filtertest.c.
854 PH/32 Applied Sieve patches from the maintainer.
856 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
858 PH/33 Added "verify = not_blind".
860 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
861 Local/Makefile (with some defaults set). These are used in built scripts
862 such as exicyclog, but they have never been used in the exim_install
863 script (though there are many overriding facilities there). I have
864 arranged that the exim_install script now takes note of these two
867 PH/35 Installed configuration files for Dragonfly.
869 PH/36 When a locally submitted message by a trusted user did not contain a
870 From: header, and the sender address was obtained from -f or from an SMTP
871 MAIL command, and the trusted user did not use -F to supply a sender
872 name, $originator_name was incorrectly used when constructing a From:
873 header. Furthermore, $originator_name was used for submission mode
874 messages from external hosts without From: headers in a similar way,
875 which is clearly wrong.
877 PH/37 Added control=suppress_local_fixups.
879 PH/38 When log_selector = +received_sender was set, and the addition of the
880 sender made the log line's construction buffer exactly full, or one byte
881 less than full, an overflow happened when the terminating "\n" was
884 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
885 when the result of a list match is failure because a DNS lookup failed.
887 PH/40 RM_COMMAND is now used in the building process.
889 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
890 the "build-* directories that it finds.
892 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
893 address in a domain literal was a prefix of an interface address.
895 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
896 when verifying a sender address, unless rewrite_headers is false.
898 PH/44 Wrote a long comment about why errors_to addresses are verified as
899 recipients, not senders.
901 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
902 the ratelimit ACL was added.
904 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
906 PH/46 Added extra information about PostgreSQL errors to the error string.
908 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
909 avoiding the need to install special zones in a real server. This is
910 backwards compatible; if it can't find the fake resolver, it drops back.
911 Thus, both old and new test suites can be run.
913 TF/02 Added util/ratelimit.pl
915 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
916 clock is set back in time.
918 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
919 Candler <B.Candler@pobox.com>.
921 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
922 OK for addresses that are the result of redirections.
924 PH/48 A number of further additions for the benefit of the new test suite,
925 including a fake gethostbyname() that interfaces to the fake DNS resolver
928 TF/06 The fix for widen_domains has also been applied to qualify_single and
929 search_parents which are the other dnslookup options that can cause
932 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
935 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
937 TF/07 Exim produced the error message "an SRV record indicated no SMTP
938 service" if it encountered an MX record with an empty target hostname.
939 The message is now "an MX or SRV record indicated no SMTP service".
941 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
942 if the DNS of the sending site is misconfigured. This is quite a
943 common situation. This change restores the behaviour of treating a
944 helo verification defer as a failure.
946 PH/51 If self=fail was set on a router, the bounce message did not include the
947 actual error message.
953 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
955 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
956 command itself ended in error, the underlying message about the transport
957 filter timeout was being overwritten with the pipe command error. Now the
958 underlying error message should be appended to the second error message.
960 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
962 PH/02 Reduce the amount of output that "make" produces by default. Full output
963 can still be requested.
965 PH/03 The warning log line about a condition test deferring for a "warn" verb
966 was being output only once per connection, rather than after each
967 occurrence (because it was using the same function as for successful
968 "warn" verbs). This seems wrong, so I have changed it.
970 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
971 it should not have, which might have caused a crash in the right
972 circumstances, but probably never did.
974 PH/04 Installed a modified version of Tony Finch's patch to make submission
975 mode fix the return path as well as the Sender: header line, and to
976 add a /name= option so that you can make the user's friendly name appear
979 TF/03 Added the control = fakedefer ACL modifier.
981 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
982 Mark Lowes for thorough testing.
984 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
986 TK/03 Merged latest SRS patch from Miles Wilton.
988 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
989 with the definition in sysexits.h (which is #included earlier).
990 Fortunately, Exim does not actually use EX_OK. The code used to try to
991 preserve the sysexits.h value, by assumimg that macro definitions were
992 scanned for macro replacements. I have been disabused of this notion,
993 so now the code just undefines EX_OK before #including unistd.h.
995 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
996 in the smtp transport. When a block could not be written in a single
997 write() function, the timeout was being re-applied to each part-write.
998 This seems wrong - if the receiver was accepting one byte at a time it
999 would take for ever. The timeout is now adjusted when this happens. It
1000 doesn't have to be particularly precise.
1002 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1003 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1005 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1006 without a domain if the check to <postmaster@domain> fails.
1008 SC/01 Eximstats: added -xls and the ability to specify output files
1009 (patch written by Frank Heydlauf).
1011 SC/02 Eximstats: use FileHandles for outputing results.
1013 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1015 SC/04 Eximstats: fixed display of large numbers with -nvr option
1017 SC/05 Eximstats: fixed merging of reports with empty tables.
1019 SC/06 Eximstats: added the -include_original_destination flag
1021 SC/07 Eximstats: removed tabs and trailing whitespace.
1023 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1025 TK/06 MBOX spool code: Add real "From " MBOX separator line
1026 so the .eml file is really in mbox format (even though
1027 most programs do not really care). Patch from Alex Miller.
1029 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1030 The latter is generated from $received_to and is only set if the
1031 message has one envelope recipient. SA can use these headers,
1032 obviously out-of-the-box. Patch from Alex Miller.
1034 PH/08 The ${def test on a variable was returning false if the variable's
1035 value was "0", contrary to what the specification has always said!
1036 The result should be true unless the variable is empty.
1038 PH/09 The syntax error of a character other than { following "${if
1039 def:variable_name" (after optional whitespace) was not being diagnosed.
1040 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1041 accidental colon was present, for example, could give incorrect results.
1043 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1044 result is used (not including appendfile, where other changes are about
1047 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1048 This involved changing a lot of size variables from int to off_t. It
1049 should work with maildirs and everything.
1051 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1052 spamd dying while we are connected to it.
1054 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1057 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1058 function for BDB 4.3.
1060 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1062 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1063 the sources for PCRE have been re-arranged and more files are now
1066 PH/15 The code I had for printing potentially long long variables in PH/11
1067 above was not the best (it lost precision). The length of off_t variables
1068 is now inspected at build time, and an appropriate printing format (%ld
1069 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1070 to be "long long int" or "long int". This is needed for the internal
1071 formatting function string_vformat().
1073 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1074 the configuration file to be ":syslog", then the script "guesses" where
1075 the logs files are, rather than using the compiled in default. In our
1076 case the guess is not the same as the compiled default, so the script
1077 suddenly stopped working when I started to use syslog. The patch checks
1078 to see if log_file_path is "". If so, it attempts to read it from exim
1079 with no configuration file to get the compiled in version, before it
1080 falls back to the previous guessing code."
1082 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1083 implementing BATV in an Exim configuration. See NewStuff for the gory
1086 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1087 Makefile that are specific to HP-UX.
1089 PH/18 If the "use_postmaster" option was set for a recipient callout together
1090 with the "random" option, the postmaster address was used as the MAIL
1091 FROM address for the random test, but not for the subsequent recipient
1092 test. It is now used for both.
1094 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1095 patch removes a few documentation additions to RFC 3028, because the
1096 latest draft now contains them. It adds the new en;ascii-case comparator
1097 and a new error check for 8bit text in MIME parts. Comparator and
1098 require names are now matched exactly. I enabled the subaddress
1099 extension, but it is not well tested yet (read: it works for me)."
1101 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1102 rework some of the code of TK/09 above to avoid the hardwired use of
1103 "%lld" and "long long". Replaced the call to snprintf() with a call to
1106 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1107 records point to non-existent hosts", "retry timeout exceeded", and
1108 "retry time not reached for any host after a long failure period".
1110 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1111 experimental DomainKeys support:
1113 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1114 (2) On an error such as an illegally used "control", the wrong name for
1115 the control was given.
1117 These problems did NOT occur unless DomainKeys support was compiled.
1119 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1121 PH/24 Added ${if match_ip condition.
1123 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1124 databases so that it will be absolutely obvious if a crash occurs in the
1125 DB library. This is a regular occurrence (often caused by mis-matched
1128 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1129 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1130 picked up on a user's system that detects such things. There doesn't seem
1131 to be a gcc warning option for this - only an attribute that has to be
1132 put on the function's prototype. It seems that in Fedora Core 4 they have
1133 set this on a number of new functions. No doubt there will be more in due
1136 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1137 specify a transport. However, if an address that was verified by such a
1138 router was the subject of a callout, Exim crashed because it tried to
1139 read the rcpt_include_affixes from the non-existent transport. Now it
1140 just assumes that the setting of that option is false. This bug was
1141 introduced by 4.51/PH/31.
1143 PH/28 Changed -d+all to exclude +memory, because that information is very
1144 rarely of interest, but it makes the output a lot bigger. People tend to
1145 do -d+all out of habit.
1147 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1148 code in os-type was giving problems when libc.so lives in lib64, like on
1151 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1152 aren't the modern standard, and it seems that some systems' include files
1153 don't always have them. Exim was already checking for some of the newer
1154 ones like T_AAAA, and defining it itself. I've added checks for all the
1155 record types that Exim uses.
1157 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1158 not automatically generating a new one, as it is supposed to. This
1159 prevented TLS from working. If the file did exist, but contained invalid
1160 data, a new version was generated, as expected. It was only the case of a
1161 non-existent file that was broken.
1163 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1164 with a change in libdomainkeys > 0.64.
1166 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1167 from DNS. If the selector record carries the flag, it now has
1168 precedence over the domain-wide flag.
1170 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1172 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1173 the use of a port name that isn't defined in /etc/services) occurred, the
1174 message was deferred as in a normal delivery, and thus remained on the
1175 spool, instead of being failed because of the mua_wrapper setting. This
1176 is now fixed, and I tidied up some of the mua_wrapper messages at the
1179 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1180 the messages in a hash of arrays rather than using individual hashes.
1181 This is a bit cleaner and results in dramatic memory savings, albeit
1182 at a slight CPU cost.
1184 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1185 as requested by Marc Sherman.
1187 SC/10 Eximstats: added histograms for user specified patterns as requested
1190 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1192 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1193 fopen() in the content-scanning modules that did not already have it.
1199 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1200 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1202 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1204 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1206 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1207 to test Sieve filters that use "vacation".
1209 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1210 that changes the way the GnuTLS parameters are stored in the cache file.
1211 The new format can be generated externally. For backward compatibility,
1212 if the data in the cache doesn't make sense, Exim assumes it has read an
1213 old-format file, and it generates new data and writes a new file. This
1214 means that you can't go back to an older release without removing the
1217 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1218 work if there are any delivery delays because "one_time" forces the
1219 parent to be marked "delivered", so its unseen clone is never tried
1220 again. For this reason, Exim now forbids the simultaneous setting of
1223 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1224 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1225 entry if you want to know the details. The fix, however, neglected to
1226 consider the case when local delivery batching is involved. The test for
1227 "previously delivered" was not happening when checking to see if an
1228 address could be batched with a previous (undelivered) one; under
1229 certain circumstances this could lead to multiple deliveries to the same
1232 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1233 in its include files, and this causes problems building Exim.
1235 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1236 header_syntax) but Exim was just ignoring anything given after a slash.
1237 In particular, this caused confusion with an attempt to use "verify =
1238 reverse_host_lookup/defer_ok". An error is now given when options are
1239 supplied for verify items that do not have them. (Maybe reverse_host_
1240 lookup should have a defer_ok option, but that's a different point.)
1242 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1243 defined by RFC 821) to 2048, because there were problems with some AUTH
1244 commands, and RFC 1869 says the size should be increased for extended
1245 SMTP commands that take arguments.
1247 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1250 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1251 "unknown" error; now it says that the functionality isn't in the binary.
1253 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1254 an address' error message when a string expansion fails (syntax or
1255 whatever). Otherwise the password may appear in the log. Following change
1256 PH/42 below, there is no longer a chance of it appearing in a bounce
1259 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1261 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1262 of its addresses was ever tried. (Bugzilla bug #2.)
1264 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1265 the result incorrectly in the debug output. (It correctly added a newline
1266 to what was transported.)
1268 TF/01 Added $received_time.
1270 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1271 commented out examples of how to interface to a virus scanner and to
1272 SpamAssassin. Also added commented examples of av_scanner and
1273 spamd_address settings.
1275 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1276 and controls are allowed in which ACLs. There were a couple of minor
1277 errors. Some of the entries in the conditions table (which is a table of
1278 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1279 negation of where the condition IS allowed.
1281 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1283 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1284 header file does not have a version number, so I've had to invent a new
1285 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1286 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1287 radiusclient), but it was contributed by a Radius user.
1289 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1290 files or directories, for OpenSSL.
1292 PH/19 When an Exim process that is running as root has to create an Exim log
1293 file, it does so in a subprocess that runs as exim:exim so as to get the
1294 ownership right at creation (otherwise, other Exim processes might see
1295 the file with the wrong ownership). There was no test for failure of this
1296 fork() call, which would lead to the process getting stuck as it waited
1297 for a non-existent subprocess. Forks do occasionally fail when resources
1298 run out. I reviewed all the other calls to fork(); they all seem to check
1301 PH/20 When checking for unexpected SMTP input at connect time (before writing
1302 the banner), Exim was not dealing correctly with a non-positive return
1303 from the read() function. If the client had disconnected by this time,
1304 the result was a log entry for a synchronization error with an empty
1305 string after "input=" when read() returned zero. If read() returned -1
1306 (an event I could not check), uninitialized data bytes were printed.
1307 There were reports of junk text (parts of files, etc) appearing after
1310 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1312 PH/22 Added support for macro redefinition, and (re)definition in between
1313 driver and ACL definitions.
1315 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1316 forgetting to use the resulting value; it was using the unexpanded value.
1318 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1319 hadn't been configured. The fix is from Juergen Kreileder, who
1320 understands it better than I do:
1322 "Here's what I see happening with three configured cyrus_sasl
1323 authenticators configured (plain, login, cram-md5):
1325 On startup auth_cyrus_sasl_init() gets called for each of these.
1326 This means three calls to sasl_listmech() without a specified mech_list.
1327 => SASL tests which mechs of all available mechs actually work
1328 => three warnings about OTP not working
1329 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1331 With the patch, sasl_listmech() also gets called three times. But now
1332 SASL's mech_list option is set to the server_mech specified in the the
1333 authenticator. Or in other words, the answer from sasl_listmech()
1334 gets limited to just the mech you're testing for (which is different
1336 => the return list contains just 'plain' or 'login', 'cram-md5' or
1337 nothing depending on the value of ob->server_mech.
1339 I've just tested the patch: Authentication still works fine,
1340 unavailable mechs specified in the exim configuration are still
1341 caught, and the auth.log warnings about OTP are gone."
1343 PH/25 When debugging is enabled, the contents of the command line are added
1344 to the debugging output, even when log_selector=+arguments is not
1347 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1348 answer is "GNU", and only if the return is "GNU/something" is the answer
1351 PH/27 $acl_verify_message is now set immediately after the failure of a
1352 verification in an ACL, and so is available in subsequent modifiers. In
1353 particular, the message can be preserved by coding like this:
1355 warn !verify = sender
1356 set acl_m0 = $acl_verify_message
1358 Previously, $acl_verify_message was set only while expanding "message"
1359 and "log_message" when a very denied access.
1361 PH/28 Modified OS/os.c-Linux with
1363 -#ifndef OS_LOAD_AVERAGE
1364 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1366 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1367 nomenclature these days.)
1369 PH/29 Installed patch from the Sieve maintainer that adds the options
1370 sieve_useraddress and sieve_subaddress to the redirect router.
1372 PH/30 In these circumstances:
1373 . Two addresses routed to the same list of hosts;
1374 . First host does not offer TLS;
1375 . First host accepts first address;
1376 . First host gives temporary error to second address;
1377 . Second host offers TLS and a TLS session is established;
1378 . Second host accepts second address.
1379 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1380 and peerdn, if requested) that were in fact used only for the second
1383 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1384 attention to any local part prefix or suffix that was matched by the
1385 router that accepted the address. It now behaves in the same way as it
1386 does for delivery: the affixes are removed from the local part unless
1387 rcpt_include_affixes is set on the transport.
1389 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1390 timeout during the DATA phase of an incoming message.
1392 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1393 applied a patch sanctioned by the Sieve maintainer.
1395 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1396 the uid or gid is negative. A case of a negative gid caused this to be
1397 noticed. The fix allows for either to be negative.
1399 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1400 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1403 PH/36 The change PH/12 above was broken. Fixed it.
1405 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1406 the grounds that routing the same address twice would always produce the
1407 same answer. This might have been true once, but it is certainly no
1408 longer true now. Routing a child address may depend on the previous
1409 routing that produced that child. Some complicated redirection strategies
1410 went wrong when messages had multiple recipients, and made Exim's
1411 behaviour dependent on the order in which the addresses were given.
1413 I have moved the duplicate checking until after the routing is complete.
1414 Exim scans the addresses that are assigned to local and remote
1415 transports, and removes any duplicates. This means that more work will be
1416 done, as duplicates will always all be routed, but duplicates are
1417 presumably rare, so I don't expect this is of any significance.
1419 For deliveries to pipes, files, and autoreplies, the duplicate checking
1420 still happens during the routing process, since they are not going to be
1423 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1424 It corrects a timeout issue with spamd. This is Ian's comment: "The
1425 background is that sometimes spamd either never reads data from a
1426 connection it has accepted, or it never writes response data. The exiscan
1427 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1428 blindly assumes that writes won't block so it may never time out."
1430 PH/39 Allow G after quota size as well as K and M.
1432 PH/40 The value set for $authenticated_id in an authenticator may not contain
1433 binary zeroes or newlines because the value is written to log lines and
1434 to spool files. There was no check on this. Now the value is run through
1435 the string_printing() function so that such characters are converted to
1436 printable escape sequences.
1438 PH/41 $message_linecount is a new variable that contains the total number of
1439 lines in the message. Compare $body_linecount, which is the count for the
1442 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1443 bounce and delay warning messages, except in certain special cases, which
1446 (a) An SMTP error message from a remote host;
1447 (b) A message specified in a :fail: redirection;
1448 (c) A message specified in a "fail" command in a system filter;
1449 (d) A message specified in a FAIL return from the queryprogram router;
1450 (e) A message specified by the cannot_route_message router option.
1452 In these cases only, Exim does include the error details in bounce and
1453 warning messages. There are also a few cases where bland messages such
1454 as "unrouteable address" or "local delivery error" are given.
1456 PH/43 $value is now also set for the "else" part of a ${run expansion.
1458 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1459 being worked on, but at least Exim now implements the latest version to
1462 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1463 process to complete was treated as a delivery failure, a timeout while
1464 writing the message to the pipe was logged, but erroneously treated as a
1465 successful delivery. Such timeouts include transport filter timeouts. For
1466 consistency with the overall process timeout, these timeouts are now
1467 treated as errors, giving rise to delivery failures by default. However,
1468 there is now a new Boolean option for the pipe transport called
1469 timeout_defer, which, if set TRUE, converts the failures into defers for
1470 both kinds of timeout. A transport filter timeout is now identified in
1473 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1474 systems where "make" and "gmake" are different, calling "gmake" at top
1475 level broke things. I've arranged for the value of $(MAKE) to be passed
1476 from the Makefile to this script so that it can call the same version of
1480 A note about Exim versions 4.44 and 4.50
1481 ----------------------------------------
1483 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1484 changes of various kinds. As a consequence, a big documentation update was
1485 needed. This delayed the release for rather longer than seemed good, especially
1486 in the light of a couple of (minor) security issues. Therefore, the changes
1487 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1488 release. So 4.44 and 4.50 are in effect two different branches that both start
1491 I have left the 4.50 change log unchanged; it contains all the changes since
1492 4.43. The change log for 4.44 is below; many of its items are identical to
1493 those for 4.50. This seems to be the most sensible way to preserve the
1494 historical information.
1500 1. Minor wording change to the doc/README.SIEVE file.
1502 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1503 computation of the current number of files was incorrect.
1505 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1506 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1509 4. Give more explanation in the error message when the command for a transport
1510 filter fails to execute.
1512 5. There are several places where Exim runs a non-Exim command in a
1513 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1514 was being done only for the command run by the queryprogram router. It is
1515 now done for all such subprocesses. The other cases are: ${run, transport
1516 filters, and the commands run by the lmtp and pipe transports.
1518 6. Added CONFIGURE_GROUP build-time option.
1520 7. Some older OS have a limit of 256 on the maximum number of file
1521 descriptors. Exim was using setrlimit() to set 1000 as a large value
1522 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1523 systems. I've change it so that if it can't get 1000, it tries for 256.
1525 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1526 was an oversight, and furthermore, ever since the addition of extra
1527 controls (e.g. 4.43/32), the checks on when to allow different forms of
1528 "control" were broken. There should now be diagnostics for all cases when a
1529 control that does not make sense is encountered.
1531 9. Added the /retain_sender option to "control=submission".
1533 10. $recipients is now available in the predata ACL (oversight).
1535 11. Tidy the search cache before the fork to do a delivery from a message
1536 received from the command line. Otherwise the child will trigger a lookup
1537 failure and thereby defer the delivery if it tries to use (for example) a
1538 cached ldap connection that the parent has called unbind on.
1540 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1541 of $address_data from the recipient verification was clobbered by the
1542 sender verification.
1544 13. The value of address_data from a sender verification is now available in
1545 $sender_address_data in subsequent conditions in the ACL statement.
1547 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1549 15. Added a new option "connect=<time>" to callout options, to set a different
1552 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1553 was its contents. (It was OK if the option was not defined at all.)
1555 17. A "Completed" log line is now written for messages that are removed from
1556 the spool by the -Mrm option.
1558 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1559 information about exactly what failed.
1561 19. Added -dd to debug only the daemon process.
1563 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1564 handles timeouts, both on the server side and network timeouts. Renamed the
1565 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1567 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1568 It is now set to "smtps".
1570 22. $host_address is now set to the target address during the checking of
1571 ignore_target_hosts.
1573 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1574 being passed; this would have caused $sender_host_name to have been used if
1575 matching the list had actually called for a host name (not very likely,
1576 since this list is usually IP addresses). A host name is now passed as
1579 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1580 code that specifies a non-restarting handler (typically sigaction() in
1581 modern systems) in an attempt to fix a rare and obscure crash bug.
1583 25. Narrowed the window for a race in the daemon that could cause it to ignore
1584 SIGCHLD signals. This is not a major problem, because they are used only to
1585 wake it up if nothing else does.
1587 26. A malformed maildirsize file could cause Exim to calculate negative values
1588 for the mailbox size or file count. Odd effects could occur as a result.
1589 The maildirsize information is now recalculated if the size or filecount
1592 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1593 support for a long time. Removed HAVE_SYS_VFS_H.
1595 28. Installed the latest version of exipick from John Jetmore.
1597 29. In an address list, if the pattern was not a regular expression, an empty
1598 subject address (from a bounce message) matched only if the pattern was an
1599 empty string. Non-empty patterns were not even tested. This was the wrong
1600 because it is perfectly reasonable to use an empty address as part of a
1601 database query. An empty address is now tested by patterns that are
1602 lookups. However, all the other forms of pattern expect the subject to
1603 contain a local part and a domain, and therefore, for them, an empty
1604 address still always fails if the pattern is not itself empty.
1606 30. Exim went into a mad DNS loop when attempting to do a callout where the
1607 host was specified on an smtp transport, and looking it up yielded more
1608 than one IP address.
1610 31. Re-factored the code for checking spool and log partition space into a
1611 function that finds that data and another that does the check. The former
1612 is then used to implement four new variables: $spool_space, $log_space,
1613 $spool_inodes, and $log_inodes.
1615 32. The RFC2047 encoding function was originally intended for short strings
1616 such as real names; it was not keeping to the 75-character limit for
1617 encoded words that the RFC imposes. It now respects the limit, and
1618 generates multiple encoded words if necessary. To be on the safe side, I
1619 have increased the buffer size for the ${rfc2047: expansion operator from
1622 33. It is now permitted to omit both strings after an "if" condition; if the
1623 condition is true, the result is "true". As before, when the second string
1624 is omitted, a false condition yields an empty string. This makes it less
1625 cumbersome to write custom ACL and router conditions.
1627 34. Failure to deliver a bounce message always caused it to be frozen, even if
1628 there was an errors_to setting on the router. The errors_to setting is now
1631 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1632 canonical form (fully expanded) before being placed in
1633 $sender_host_address.
1635 36. The table in the code that translates DNS record types into text (T_A to
1636 "A" for instance) was missing entries for NS and CNAME. It is just possible
1637 that this could have caused confusion if both these types were looked up
1638 for the same domain, because the text type is used as part of Exim's
1639 per-process caching. But the chance of anyone hitting this buglet seems
1642 37. The dnsdb lookup has been extended in a number of ways.
1644 (1) There is a new type, "zns", which walks up the domain tree until it
1645 finds some nameserver records. It should be used with care.
1647 (2) There is a new type, "mxh", which is like "mx" except that it returns
1648 just the host names, not the priorities.
1650 (3) It is now possible to give a list of domains (or IP addresses) to be
1651 looked up. The behaviour when one of the lookups defers can be
1652 controlled by a keyword.
1654 (4) It is now possible to specify the separator character for use when
1655 multiple records are returned.
1657 38. The dnslists ACL condition has been extended: it is now possible to supply
1658 a list of IP addresses and/or domains to be looked up in a particular DNS
1661 39. Added log_selector=+queue_time_overall.
1663 40. When running the queue in the test harness, wait just a tad after forking a
1664 delivery process, to get repeatability of debugging output.
1666 41. Include certificate and key file names in error message when GnuTLS fails
1667 to set them up, because the GnuTLS error message doesn't include the name
1668 of the failing file when there is a problem reading it.
1670 42. Allow both -bf and -bF in the same test run.
1672 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1674 44. The "Exiscan patch" is now merged into the mainline Exim source.
1676 45. Sometimes the final signoff response after QUIT could fail to get
1677 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1678 < 0 before doing a fflush(). This bug looks as though it goes back to the
1679 introduction of TLS in release 3.20, but "sometimes" must have been rare
1680 because the tests only now provoked it.
1682 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1683 (this can affect the format of dates).
1685 47. exim_tidydb, when checking for the continued existence of a message for
1686 which it has found a message-specific retry record, was not finding
1687 messages that were in split spool directories. Consequently, it was
1688 deleting retry records that should have stayed in existence.
1690 48. Steve fixed some bugs in eximstats.
1692 49. The SPA authentication driver was not abandoning authentication and moving
1693 on to the next authenticator when an expansion was forced to fail,
1694 contradicting the general specification for all authenticators. Instead it
1695 was generating a temporary error. It now behaves as specified.
1697 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1698 (the order specifies the preference for clients). The order is now AES256,
1699 AES128, 3DES, ARCFOUR128.
1701 51. Small patch to Sieve code - explicitly set From: when generating an
1704 52. Exim crashed if a remote delivery caused a very long error message to be
1705 recorded - for instance if somebody sent an entire SpamAssassin report back
1706 as a large number of 550 error lines. This bug was coincidentally fixed by
1707 increasing the size of one of Exim's internal buffers (big_buffer) that
1708 happened as part of the Exiscan merge. However, to be on the safe side, I
1709 have made the code more robust (and fixed the comments that describe what
1712 53. Now that there can be additional text after "Completed" in log lines (if
1713 the queue_time_overall log selector is set), a one-byte patch to exigrep
1714 was needed to allow it to recognize "Completed" as not the last thing in
1717 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1718 patch that reportedly fixes this has been added. I am not expert enough to
1719 create a test for it. This is what the patch creator wrote:
1721 "I found a little strange behaviour of ldap code when working with
1722 Windows 2003 AD Domain, where users was placed in more than one
1723 Organization Units. When I tried to give exim partial DN, the exit code
1724 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1725 But simultaneously result of request was absolutely normal ldap result,
1726 so I produce this patch..."
1728 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1729 REFERENCE, so I have modified the code to exclude the patch when that macro
1732 55. Some experimental protocols are using DNS PTR records for new purposes. The
1733 keys for these records are domain names, not reversed IP addresses. The
1734 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1735 leaves it alone. Component reversal etc. now happens only for IP addresses.
1738 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1740 57. Double the size of the debug message buffer (to 2048) so that more of very
1741 long debug lines gets shown.
1743 58. The exicyclog utility now does better if the number of log files to keep
1744 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1746 59. Two changes related to the smtp_active_hostname option:
1748 (1) $smtp_active_hostname is now available as a variable.
1749 (2) The default for smtp_banner uses $smtp_active_hostname instead
1750 of $primary_hostname.
1752 60. The host_aton() function is supposed to be passed a string that is known
1753 to be a valid IP address. However, in the case of IPv6 addresses, it was
1754 not checking this. This is a hostage to fortune. Exim now panics and dies
1755 if the condition is not met. A case was found where this could be provoked
1756 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1757 components; fortuitously, this particular loophole had already been fixed
1758 by change 4.50/55 above.
1760 If there are any other similar loopholes, the new check in host_aton()
1761 itself should stop them being exploited. The report I received stated that
1762 data on the command line could provoke the exploit when Exim was running as
1763 exim, but did not say which command line option was involved. All I could
1764 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1765 running as the user.
1768 61. There was a buffer overflow vulnerability in the SPA authentication code
1769 (which came originally from the Samba project). I have added a test to the
1770 spa_base64_to_bits() function which I hope fixes it.
1773 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1774 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1776 63. The daemon start-up calls getloadavg() while still root for those OS that
1777 need the first call to be done as root, but it missed one case: when
1778 deliver_queue_load_max is set with deliver_drop_privilege. This is
1779 necessary for the benefit of the queue runner, because there is no re-exec
1780 when deliver_drop_privilege is set.
1782 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1783 This has been fixed.
1785 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1786 was in use, was not putting the data itself into the right store pool;
1787 consequently, it could be overwritten for a subsequent message in the same
1788 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1791 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1793 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1794 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1795 the function were treating the return as a boolean value, which happened to
1796 work because 0=false and not-0=true, but is not correct code.
1798 68. The host_aton() function was not handling scoped IPv6 addresses (those
1799 with, for example, "%eth0" on the end) correctly.
1801 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1802 negated items (that is, ~something) in unsigned ints. Some compilers
1803 apparently mutter when there is no cast.
1805 70. If an address verification called from an ACL failed, and did not produce a
1806 user-specific message (i.e. there was only a "system" message), nothing was
1807 put in $acl_verify_message. In this situation, it now puts the system
1810 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1811 session; change 4.31/43 added the unwanted input to the log line - except
1812 that it did not do this in the start of session case. It now does.
1814 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1815 This is wrong and can cause the other end to generate a synchronization
1816 error if it is another Exim or anything else that does the synchronization
1817 check. A QUIT command is no longer sent after a timeout.
1819 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1820 during host lookups.
1822 74. The defer_ok option of callout verification was not working if it was used
1823 when verifying addresses in header lines, that is, for this case:
1825 verify = header_sender/callout=defer_ok
1827 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1828 those file descriptors could be used for SMTP connections. If anything
1829 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1830 could be sent to the SMTP client, causing chaos. The daemon now opens
1831 stdin, stdout, and stderr to /dev/null when it puts itself into the
1834 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1835 log by default. The user can override this with suitable Perl magic.
1837 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1838 the log message when discard triggers, was not working for the DATA ACL or
1839 for the non-SMTP ACL.
1841 78. Error message wording change in sieve.c.
1843 79. If smtp_accept_max_per_host was set, the number of connections could be
1844 restricted to fewer than expected, because the daemon was trying to set up
1845 a new connection before checking whether the processes handling previous
1846 connections had finished. The check for completed processes is now done
1847 earlier. On busy systems, this bug wouldn't be noticed because something
1848 else would have woken the daemon, and it would have reaped the completed
1851 80. If a message was submitted locally by a user whose login name contained one
1852 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1853 It caused a spool format error. I have fixed the spool reading code. A
1854 related problem was that the "from" clause in the Received: line became
1855 illegal because of the space(s). It is now covered by ${quote_local_part.
1857 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1860 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1862 83. Patch from Sieve maintainer to fix unterminated string problem in
1863 "vacation" handling.
1865 84. Some minor changes to the Linux configuration files to help with other
1866 OS variants using glibc.
1868 85. One more patch for Sieve to update vacation handling to latest spec.
1871 ----------------------------------------------------
1872 See the note above about the 4.44 and 4.50 releases.
1873 ----------------------------------------------------
1879 1. Change 4.43/35 introduced a bug that caused file counts to be
1880 incorrectly computed when quota_filecount was set in an appendfile
1883 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1884 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1887 3. Give more explanation in the error message when the command for a transport
1888 filter fails to execute.
1890 4. There are several places where Exim runs a non-Exim command in a
1891 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1892 was being done only for the command run by the queryprogram router. It is
1893 now done for all such subprocesses. The other cases are: ${run, transport
1894 filters, and the commands run by the lmtp and pipe transports.
1896 5. Some older OS have a limit of 256 on the maximum number of file
1897 descriptors. Exim was using setrlimit() to set 1000 as a large value
1898 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1899 systems. I've change it so that if it can't get 1000, it tries for 256.
1901 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1902 was an oversight, and furthermore, ever since the addition of extra
1903 controls (e.g. 4.43/32), the checks on when to allow different forms of
1904 "control" were broken. There should now be diagnostics for all cases when a
1905 control that does not make sense is encountered.
1907 7. $recipients is now available in the predata ACL (oversight).
1909 8. Tidy the search cache before the fork to do a delivery from a message
1910 received from the command line. Otherwise the child will trigger a lookup
1911 failure and thereby defer the delivery if it tries to use (for example) a
1912 cached ldap connection that the parent has called unbind on.
1914 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1915 of $address_data from the recipient verification was clobbered by the
1916 sender verification.
1918 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1919 was its contents. (It was OK if the option was not defined at all.)
1921 11. A "Completed" log line is now written for messages that are removed from
1922 the spool by the -Mrm option.
1924 12. $host_address is now set to the target address during the checking of
1925 ignore_target_hosts.
1927 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1928 being passed; this would have caused $sender_host_name to have been used if
1929 matching the list had actually called for a host name (not very likely,
1930 since this list is usually IP addresses). A host name is now passed as
1933 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1934 code that specifies a non-restarting handler (typically sigaction() in
1935 modern systems) in an attempt to fix a rare and obscure crash bug.
1937 15. Narrowed the window for a race in the daemon that could cause it to ignore
1938 SIGCHLD signals. This is not a major problem, because they are used only to
1939 wake it up if nothing else does.
1941 16. A malformed maildirsize file could cause Exim to calculate negative values
1942 for the mailbox size or file count. Odd effects could occur as a result.
1943 The maildirsize information is now recalculated if the size or filecount
1946 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1947 support for a long time. Removed HAVE_SYS_VFS_H.
1949 18. Updated exipick to current release from John Jetmore.
1951 19. Allow an empty sender to be matched against a lookup in an address list.
1952 Previously the only cases considered were a regular expression, or an
1955 20. Exim went into a mad DNS lookup loop when doing a callout where the
1956 host was specified on the transport, if the DNS lookup yielded more than
1959 21. The RFC2047 encoding function was originally intended for short strings
1960 such as real names; it was not keeping to the 75-character limit for
1961 encoded words that the RFC imposes. It now respects the limit, and
1962 generates multiple encoded words if necessary. To be on the safe side, I
1963 have increased the buffer size for the ${rfc2047: expansion operator from
1966 22. Failure to deliver a bounce message always caused it to be frozen, even if
1967 there was an errors_to setting on the router. The errors_to setting is now
1970 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1971 canonical form (fully expanded) before being placed in
1972 $sender_host_address.
1974 24. Updated eximstats to version 1.33
1976 25. Include certificate and key file names in error message when GnuTLS fails
1977 to set them up, because the GnuTLS error message doesn't include the name
1978 of the failing file when there is a problem reading it.
1980 26. Expand error message when OpenSSL has problems setting up cert/key files.
1983 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1984 (this can affect the format of dates).
1986 28. exim_tidydb, when checking for the continued existence of a message for
1987 which it has found a message-specific retry record, was not finding
1988 messages that were in split spool directories. Consequently, it was
1989 deleting retry records that should have stayed in existence.
1991 29. eximstats updated to version 1.35
1992 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1993 1.35 - bugfix such that pie charts by volume are generated correctly
1995 30. The SPA authentication driver was not abandoning authentication and moving
1996 on to the next authenticator when an expansion was forced to fail,
1997 contradicting the general specification for all authenticators. Instead it
1998 was generating a temporary error. It now behaves as specified.
2000 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2001 (the order specifies the preference for clients). The order is now AES256,
2002 AES128, 3DES, ARCFOUR128.
2004 31. Small patch to Sieve code - explicitly set From: when generating an
2007 32. Exim crashed if a remote delivery caused a very long error message to be
2008 recorded - for instance if somebody sent an entire SpamAssassin report back
2009 as a large number of 550 error lines. This bug was coincidentally fixed by
2010 increasing the size of one of Exim's internal buffers (big_buffer) that
2011 happened as part of the Exiscan merge. However, to be on the safe side, I
2012 have made the code more robust (and fixed the comments that describe what
2015 33. Some experimental protocols are using DNS PTR records for new purposes. The
2016 keys for these records are domain names, not reversed IP addresses. The
2017 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2018 leaves it alone. Component reversal etc. now happens only for IP addresses.
2021 34. The host_aton() function is supposed to be passed a string that is known
2022 to be a valid IP address. However, in the case of IPv6 addresses, it was
2023 not checking this. This is a hostage to fortune. Exim now panics and dies
2024 if the condition is not met. A case was found where this could be provoked
2025 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2026 components; fortuitously, this particular loophole had already been fixed
2027 by change 4.50/55 or 4.44/33 above.
2029 If there are any other similar loopholes, the new check in host_aton()
2030 itself should stop them being exploited. The report I received stated that
2031 data on the command line could provoke the exploit when Exim was running as
2032 exim, but did not say which command line option was involved. All I could
2033 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2034 running as the user.
2037 35. There was a buffer overflow vulnerability in the SPA authentication code
2038 (which came originally from the Samba project). I have added a test to the
2039 spa_base64_to_bits() function which I hope fixes it.
2042 36. The daemon start-up calls getloadavg() while still root for those OS that
2043 need the first call to be done as root, but it missed one case: when
2044 deliver_queue_load_max is set with deliver_drop_privilege. This is
2045 necessary for the benefit of the queue runner, because there is no re-exec
2046 when deliver_drop_privilege is set.
2048 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2049 was in use, was not putting the data itself into the right store pool;
2050 consequently, it could be overwritten for a subsequent message in the same
2051 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2054 38. Sometimes the final signoff response after QUIT could fail to get
2055 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2056 < 0 before doing a fflush(). This bug looks as though it goes back to the
2057 introduction of TLS in release 3.20, but "sometimes" must have been rare
2058 because the tests only now provoked it.
2064 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2065 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2066 Now it returns an integer. A number of calls were still expecting a T/F
2067 return. Fortuitously, in all cases, the tests worked in OK situations,
2068 which is the norm. However, things would have gone wrong on any write
2069 failures on the smtp file descriptor. This function is used when sending
2070 messages over SMTP and also when doing verify callouts.
2072 2. When Exim is called to do synchronous delivery of a locally submitted
2073 message (the -odf or -odi options), it no longer closes stderr before doing
2076 3. Implemented the mua_wrapper option.
2078 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2080 5. Implemented the functions header_remove(), header_testname(),
2081 header_add_at_position(), and receive_remove_recipient(), and exported them
2084 6. If an ACL "warn" statement specified the addition of headers, Exim already
2085 inserted X-ACL-Warn: at the start if there was no header name. However, it
2086 was not making this test for the second and subsequent header lines if
2087 there were newlines in the string. This meant that an invalid header could
2088 be inserted if Exim was badly configured.
2090 7. Allow an ACL "warn" statement to add header lines at the start or after all
2091 the Received: headers, as well as at the end.
2093 8. Added the rcpt_4xx retry error code.
2095 9. Added postmaster_mailfrom=xxx to callout verification option.
2097 10. Added mailfrom=xxxx to the callout verification option, for verify=
2100 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2101 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2103 12. Inserted some casts to stop certain compilers warning when using pointer
2104 differences as field lengths or precisions in printf-type calls (mostly
2105 affecting debugging statements).
2107 13. Added optional readline() support for -be (dynamically loaded).
2109 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2110 same clock tick as a message's arrival, so that its received time was the
2111 same as the "first fail" time on the retry record, and that message
2112 remained on the queue past the ultimate address timeout, every queue runner
2113 would try a delivery (because it was past the ultimate address timeout) but
2114 after another failure, the ultimate address timeout, which should have then
2115 bounced the address, did not kick in. This was a "< instead of <=" error;
2116 in most cases the first failure would have been in the next clock tick
2117 after the received time, and all would be well.
2119 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2120 being recognized when the domain list was tested by the match_domain
2121 condition in an expansion string.
2123 16. Added the ${str2b64: operator.
2125 17. Exim was always calling setrlimit() to set a large limit for the number of
2126 processes, without checking whether the existing limit was already
2127 adequate. (It did check for the limit on file descriptors.) Furthermore,
2128 errors from getrlimit() and setrlimit() were being ignored. Now they are
2129 logged to the main and panic logs, but Exim does carry on, to try to do its
2130 job under whatever limits there are.
2132 18. Imported PCRE 5.0.
2134 19. Trivial typo in log message " temporarily refused connection" (the leading
2137 20. If the log selector return_path_on_delivery was set and an address was
2138 redirected to /dev/null, the delivery process crashed because it assumed
2139 that a return path would always be set for a "successful" delivery. In this
2140 case, the whole delivery is bypassed as an optimization, and therefore no
2143 21. Internal re-arrangement: the function for sending a challenge and reading
2144 a response while authentication was assuming a zero-terminated challenge
2145 string. It's now changed to take a pointer and a length, to allow for
2146 binary data in such strings.
2148 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2150 23. Exim was not respecting finduser_retries when seeking the login of the
2151 uid under which it was called; it was always trying 10 times. (The default
2152 setting of finduser_retries is zero.) Also, it was sleeping after the final
2153 failure, which is pointless.
2155 24. Implemented tls_on_connect_ports.
2157 25. Implemented acl_smtp_predata.
2159 26. If the domain in control=submission is set empty, Exim assumes that the
2160 authenticated id is a complete email address when it generates From: or
2161 Sender: header lines.
2163 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2164 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2165 chown and chgrp in /bin and hostname in /usr/bin.
2167 28. Exim was keeping the "process log" file open after each use, just as it
2168 does for the main log. This opens the possibility of it remaining open for
2169 long periods when the USR1 signal hits a daemon. Occasional processlog
2170 errors were reported, that could have been caused by this. Anyway, it seems
2171 much more sensible not to leave this file open at all, so that is what now
2174 29. The long-running daemon process does not normally write to the log once it
2175 has entered its main loop, and it closes the log before doing so. This is
2176 so that log files can straightforwardly be renamed and moved. However,
2177 there are a couple of unusual error situations where the daemon does write
2178 log entries, and I had neglected to close the log afterwards.
2180 30. The text of an SMTP error response that was received during a remote
2181 delivery was being truncated at 512 bytes. This is too short for some of
2182 the long messages that one sometimes sees. I've increased the limit to
2185 31. It is now possible to make retry rules that apply only when a message has a
2186 specific sender, in particular, an empty sender.
2188 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2189 it possible to be selective about when SMTP synchronization is enforced.
2191 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2193 32. Implemented hosts_connection_nolog.
2195 33. Added an ACL for QUIT.
2197 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2200 35. Added mailbox_size and mailbox_filecount to appendfile.
2202 36. Added control = no_multiline_responses to ACLs.
2204 37. There was a bug in the logic of the code that waits for the clock to tick
2205 in the case where the clock went backwards by a substantial amount such
2206 that the microsecond fraction of "now" was more than the microsecond
2207 fraction of "then" (but the whole seconds number was less).
2209 38. Added support for the libradius Radius client library this is found on
2210 FreeBSD (previously only the radiusclient library was supported).
2216 1. When certain lookups returned multiple values in the form name=value, the
2217 quoting of the values was not always being done properly. Specifically:
2218 (a) If the value started with a double quote, but contained no whitespace,
2220 (b) If the value contained whitespace other than a space character (i.e.
2221 tabs or newlines or carriage returns) it was not quoted.
2222 This fix has been applied to the mysql and pgsql lookups by writing a
2223 separate quoting function and calling it from the lookup code. The fix
2224 should probably also be applied to nisplus, ibase and oracle lookups, but
2225 since I cannot test any of those, I have not disturbed their existing code.
2227 2. A hit in the callout cache for a specific address caused a log line with no
2228 reason for rejecting RCPT. Now it says "Previous (cached) callout
2229 verification failure".
2231 3. There was an off-by-one bug in the queryprogram router. An over-long
2232 return line was truncated at 256 instead of 255 characters, thereby
2233 overflowing its buffer with the terminating zero. As well as fixing this, I
2234 have increased the buffer size to 1024 (and made a note to document this).
2236 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2237 when Exim is waiting for an SMTP response from a remote server, Exim
2238 restarts its select() call on the socket, thereby resetting its timeout.
2239 This is not a problem when such interrupts are rare. Somebody set up a cron
2240 job to run exiwhat every 2 minutes, which is less than the normal select()
2241 timeout (5 or 10 minutes). This meant that the select() timeout never
2242 kicked in because it was always reset. I have fixed this by comparing the
2243 time when an interrupt arrives with the time at the start of the first call
2244 to select(). If more time than the timeout has elapsed, the interrupt is
2245 treated as a timeout.
2247 5. Some internal re-factoring in preparation for the addition of Sieve
2248 extensions (by MH). In particular, the "personal" test is moved to a
2249 separate function, and given an option for scanning Cc: and Bcc: (which is
2250 not set for Exim filters).
2252 6. When Exim created an email address using the login of the caller as the
2253 local part (e.g. when creating a From: or Sender: header line), it was not
2254 quoting the local part when it contained special characters such as @.
2256 7. Installed new OpenBSD configuration files.
2258 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2259 try to make them clearer.
2261 9. Callout options, other than the timeout value, were being ignored when
2262 verifying sender addresses in header lines. For example, when using
2264 verify = header_sender/callout=no_cache
2266 the cache was (incorrectly) being used.
2268 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2269 only the Cygwin environment.
2271 11. When return_path_on_delivery was set as a log selector, if different remote
2272 addresses in the same message used different return paths and parallel
2273 remote delivery occurred, the wrong values would sometimes be logged.
2274 (Whenever a remote delivery process finished, the return path value from
2275 the most recently started remote delivery process was logged.)
2277 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2278 lines when AUTH and/or TLS are in use. This is the "received protocol"
2279 field. Exim used to use "asmtp" for authenticated SMTP, without any
2280 indication (in the protocol name) for TLS use. Now it follows the RFC and
2281 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2282 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2283 names appear in log lines as well as in Received: header lines.
2285 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2286 extensions, and comparison tests, and to fix some bugs.
2288 14. Changes to the "personal" filter test:
2290 (1) The test was buggy in that it was just doing the equivalent of
2291 "contains" tests on header lines. For example, if a user's address was
2292 anne@some.where, the "personal" test would incorrectly be true for
2294 To: susanne@some.where
2296 This test is now done by extracting each address from the header in turn,
2297 and checking the entire address. Other tests that are part of "personal"
2298 are now done using regular expressions (for example, to check local parts
2299 of addresses in From: header lines).
2301 (2) The list of non-personal local parts in From: addresses has been
2302 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2303 taken from the Sieve specification recommendations.
2305 (3) If the message contains any header line starting with "List-" it is
2306 treated as non-personal.
2308 (4) The test for "circular" in the Subject: header line has been removed
2309 because it now seems ill-conceived.
2311 15. Minor typos in src/EDITME comments corrected.
2313 16. Installed latest exipick from John Jetmore.
2315 17. If headers_add on a router specified a text string that was too long for
2316 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2317 of string_sprintf() is now avoided.
2319 18. $message_body_size was not set (it was always zero) when running the DATA
2320 ACL and the local_scan() function.
2322 19. For the "mail" command in an Exim filter, no default was being set for
2323 the once_repeat time, causing a random time value to be used if "once" was
2324 specified. (If the value happened to be <= 0, no repeat happened.) The
2325 default is now 0s, meaning "never repeat". The "vacation" command was OK
2326 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2327 (I found it when inspecting the code).
2329 20. There is now an overall timeout for performing a callout verification. It
2330 defaults to 4 times the callout timeout, which applies to individual SMTP
2331 commands during the callout. The overall timeout applies when there is more
2332 than one host that can be tried. The timeout is checked before trying the
2333 next host. This prevents very long delays if there are a large number of
2334 hosts and all are timing out (e.g. when the network connections are timing
2335 out). The value of the overall timeout can be changed by specifying an
2336 additional sub-option for "callout", called "maxwait". For example:
2338 verify = sender/callout=5s,maxwait=20s
2340 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2341 to the end before writing, but this should make it even safer).
2343 22. Exim was forgetting that it had advertised PIPELINING for the second and
2344 subsequent messages on an SMTP connection. It was also not resetting its
2345 memory on STARTTLS and an internal HELO.
2347 23. When Exim logs an SMTP synchronization error within a session, it now
2348 records whether PIPELINING has been advertised or not.
2350 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2351 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2352 rather than long int.
2354 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2356 26. Added the never_mail option to autoreply.
2362 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2363 crash if the getsockname() call failed; this can happen if a connection is
2364 closed very soon after it is established. The problem was simply in the
2365 order in which certain operations were done, causing Exim to try to write
2366 to the SMTP stream before it had set up the file descriptor. The bug has
2367 been fixed by making things happen in the correct order.
2373 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2374 before the connection was closed, thus losing the rejection response.
2376 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2377 some early Solaris releases, but causes trouble in current releases where
2378 socklen_t is defined.
2380 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2383 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2386 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2387 that releasing the top of it at the end releases what was used for sub-
2388 expansions (unless the block got too big). However, discard this block if
2389 the first thing is a variable or header, so that we can use its block when
2390 it is dynamic (useful for very large $message_headers, for example).
2392 6. Lookups now cache *every* query, not just the most recent. A new, separate
2393 store pool is used for this. It can be recovered when all lookup caches are
2394 flushed. Lookups now release memory at the end of their result strings.
2395 This has involved some general refactoring of the lookup sources.
2397 7. Some code has been added to the store_xxx() functions to reduce the amount
2398 of flapping under certain conditions.
2400 8. log_incoming_interface used to affect only the <= reception log lines. Now
2401 it causes the local interface and port to be added to several more SMTP log
2402 lines, for example "SMTP connection from", and rejection lines.
2404 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2406 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2408 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2409 could be overwritten at the end of the current message (or the start of a
2410 new message if it was set in a HELO ACL). The value is now preserved for
2411 the duration of the SMTP connection.
2413 12. If a transport had a headers_rewrite setting, and a matching header line
2414 contained an unqualified address, that address was qualified, even if it
2415 did not match any rewriting rules. The underlying bug was that the values
2416 of the flags that permit the existence of unqualified sender and recipient
2417 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2418 non-local messages, and by -bnq for local messages) were not being
2419 preserved with the message after it was received.
2421 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2422 "next input=" as part of the text comprising the host identity instead of
2423 the correct text. The code was using the same buffer for two different
2424 strings. However, depending on which order the printing function evaluated
2425 its arguments, the bug did not always show up. Under Linux, for example, my
2426 test suite worked just fine.
2428 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2429 doesn't work with some older versions of Perl. It has been changed to "my",
2430 which in any case is probably the better facility to use.
2432 15. A really picky compiler found some instances of statements for creating
2433 error messages that either had too many or two few arguments for the format
2436 16. The size of the buffer for calls to the DNS resolver has been increased
2437 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2438 for addresses that have a lot of PTR records. This alleviates a problem; it
2439 does not fully solve it.
2441 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2442 buffer now truncates the list and logs the incident, which is the same
2443 action as happens when Exim is looking up a host name and its aliases.
2444 Previously in this situation something unpredictable would happen;
2445 sometimes it was "internal error: store_reset failed".
2447 18. If a server dropped the connection unexpectedly when an Exim client was
2448 using GnuTLS and trying to read a response, the client delivery process
2449 crashed while trying to generate an error log message.
2451 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2452 string, for example:
2454 warn message = H1: something\nH2: something
2456 the text was added as a single header line from Exim's point of view
2457 though it ended up OK in the delivered message. However, searching for the
2458 second and subsequent header lines using $h_h2: did not work. This has been
2459 fixed. Similarly, if a system filter added multiple headers in this way,
2460 the routers could not see them.
2462 20. Expanded the error message when iplsearch is called with an invalid key to
2463 suggest using net-iplsearch in a host list.
2465 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2466 ACLs are no longer actually imposed (and a message to that effect is
2469 22. If a "gecos" field in a passwd entry contained escaped characters, in
2470 particular, if it contained a \" sequence, Exim got it wrong when building
2471 a From: or a Sender: header from that name. A second bug also caused
2472 incorrect handling when an unquoted " was present following a character
2473 that needed quoting.
2475 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2476 was not being matched caselessly.
2478 24. Arranged for all hyphens in the exim.8 source to be escaped with
2481 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2482 properly. Recipient callout cache records were still being keyed to include
2483 the sender, even when use_sender was set false. This led to far more
2484 callouts that were necessary. The sender is no longer included in the key
2485 when use_sender is false.
2487 26. Added "control = submission" modifier to ACLs.
2489 27. Added the ${base62d: operator to decode base 62 numbers.
2491 28. dnsdb lookups can now access SRV records.
2493 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2494 the configuration file.
2496 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2497 (-v) mode. This makes the output for a verbose queue run more intelligible.
2499 31. Added a use_postmaster feature to recipient callouts.
2501 32. Added the $body_zerocount variable, containing the number of binary zero
2502 bytes in the message body.
2504 33. The time of last modification of the "new" subdirectory is now used as the
2505 "mailbox time last read" when there is a quota error for a maildir
2508 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2510 35. Added +ignore_unknown as a special item in host lists.
2512 36. Code for decoding IPv6 addresses in host lists is now included, even if
2513 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2514 address was recognized as an IP address, but was then not correctly decoded
2515 into binary, causing unexpected and incorrect effects when compared with
2522 1. Very minor rewording of debugging text in manualroute to say "list of
2523 hosts" instead of "hostlist".
2525 2. If verify=header_syntax was set, and a header line with an unqualified
2526 address (no domain) and a large number of spaces between the end of the
2527 name and the colon was received, the reception process suffered a buffer
2528 overflow, and (when I tested it) crashed. This was caused by some obsolete
2529 code that should have been removed. The fix is to remove it!
2531 3. When running in the test harness, delay a bit after writing a bounce
2532 message to get a bit more predictability in the log output.
2534 4. Added a call to search_tidyup() just before forking a reception process. In
2535 theory, someone could use a lookup in the expansion of smtp_accept_max_
2536 per_host which, without the tidyup, could leave open a database connection.
2538 5. Added the variables $recipient_data and $sender_data which get set from a
2539 lookup success in an ACL "recipients" or "senders" condition, or a router
2540 "senders" option, similar to $domain_data and $local_part_data.
2542 6. Moved the writing of debug_print from before to after the "senders" test
2545 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2546 problems for message scanning, either using a data ACL, or using
2547 local_scan() because the Received: header was not generated till after they
2548 were called (in order to set the time as the time of reception completion).
2549 I have revised the way this works. The header is now generated after the
2550 body is received, but before the ACL or local_scan() are called. After they
2551 are run, the timestamp in the header is updated.
2557 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2558 before starting a queue runner without re-exec. This happened only when
2559 deliver_drop_privilege was set or when the Exim user was set to root. The
2560 effect of the bug was that timeouts during subsequent deliveries caused
2561 crashes instead of being properly handled. The handler is now left at its
2562 default (and expected) setting.
2564 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2565 message, again when deliver_drop_privilege is set or Exim is run as root.
2566 The bug described in (1) was not present in this case, but the tidying up
2567 of the other signals was missing. I have made the two cases consistent.
2569 3. The ignore_target_hosts setting on a manualroute router was being ignored
2570 for hosts that were looked up using the /MX notation.
2572 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2575 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2576 operated on the sender address. After changing the $sender_address to <>
2577 for the sender address verify, Exim was re-instated it as the original
2578 (before rewriting) address, but remembering that it had rewritten it, so it
2579 wasn't rewriting it again. This bug also had the effect of breaking the
2580 sender address verification caching when the sender address was rewritten.
2582 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2583 This has been changed so that if the ip literal address matches
2584 ignore_target_hosts, the router declines.
2586 7. Added expansion conditions match_domain, match_address, and match_local_
2587 part (NOT match_host).
2589 8. The placeholder for the Received: header didn't have a length field set.
2591 9. Added code to Exim itself and to exim_lock to test for a specific race
2592 condition that could lead to file corruption when using MBX delivery. The
2593 issue is with the lockfile that is created in /tmp. If this file is removed
2594 after a process has opened it but before that process has acquired a lock,
2595 there is the potential for a second process to recreate the file and also
2596 acquire a lock. This could lead to two Exim processes writing to the file
2597 at the same time. The added code performs the same test as UW imapd; it
2598 checks after acquiring the lock that its file descriptor still refers to
2599 the same named file.
2601 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2602 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2605 11. Added the smtp_active_hostname option. If used, this will typically be made
2606 to depend on the incoming interface address. Because $interface_address is
2607 not set up until the daemon has forked a reception process, error responses
2608 that can happen earlier (such as "too many connections") no longer contain
2611 12. If an expansion in a condition on a "warn" statement fails because a lookup
2612 defers, the "warn" statement is abandoned, and the next ACL statement is
2613 processed. Previously this caused the whole ACL to be aborted.
2615 13. Added the iplsearch lookup type.
2617 14. Added ident_timeout as a log selector.
2619 15. Added tls_certificate_verified as a log selector.
2621 16. Added a global option tls_require_ciphers (compare the smtp transport
2622 option of the same name). This controls incoming TLS connections.
2624 17. I finally figured out how to make tls_require_ciphers do a similar thing
2625 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2626 before starting the TLS session.
2628 18. Tabs are now shown as \t in -bP output.
2630 19. If the log selector return_path_on_delivery was set, Exim crashed when
2631 bouncing a message because it had too many Received: header lines.
2633 20. If two routers both had headers_remove settings, and the first one included
2634 a superfluous trailing colon, the final name in the first list and the
2635 first name in the second list were incorrectly joined into one item (with a
2636 colon in the middle).
2642 1. Added -C and -D options to the exinext utility, mainly to make it easier
2643 to include in the automated testing, but these could be helpful when
2644 multiple configurations are in use.
2646 2. The exinext utility was not formatting the output nicely when there was
2647 an alternate port involved in the retry record key, nor when there was a
2648 message id as well (for retries that were specific to a specific message
2649 and a specific host). It was also confused by IPv6 addresses, because of
2650 the additional colons they contain. I have fixed the IPv4 problem, and
2651 patched it up to do a reasonable job for IPv6.
2653 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2654 delivery, the log line now contains "pipelined" if PIPELINING was used.
2656 4. An SMTP transport process used to panic and die if the bind() call to set
2657 an explicit outgoing interface failed. This has been changed; it is now
2658 treated in the same way as a connect() failure.
2660 5. A reference to $sender_host_name in the part of a conditional expansion
2661 that was being skipped was still causing a DNS lookup. This no longer
2664 6. The def: expansion condition was not recognizing references to header lines
2665 that used bh_ and bheader_.
2667 7. Added the _cache feature to named lists.
2669 8. The code for checking quota_filecount in the appendfile transport was
2670 allowing one more file than it should have been.
2672 9. For compatibility with Sendmail, the command line option
2680 and sets the incoming protocol and host name (for trusted callers). The
2681 host name and its colon can be omitted when only the protocol is to be set.
2682 Note the Exim already has two private options, -pd and -ps, that refer to
2683 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2684 "s", but I don't think that's a major issue.
2686 10. A number of refactoring changes to the code, none of which should affect
2689 (a) The number of logging options was getting close to filling up the
2690 32-bit word that was used as a bit map. I have split them into two classes:
2691 those that are passed in the argument to log_write(), and those that are
2692 only ever tested independently outside of that function. These are now in
2693 separate 32-bit words, so there is plenty of room for expansion again.
2694 There is no change in the user interface or the logging behaviour.
2696 (b) When building, for example, log lines, the code previously used a
2697 macro that called string_cat() twice, in order to add two strings. This is
2698 not really sufficiently general. Furthermore, there was one instance where
2699 it was actually wrong because one of the argument was used twice, and in
2700 one call a function was used. (As it happened, calling the function twice
2701 did not affect the overall behaviour.) The macro has been replaced by a
2702 function that can join an arbitrary number of extra strings onto a growing
2705 (c) The code for expansion conditions now uses a table and a binary chop
2706 instead of a serial search (which was left over from when there were very
2707 few conditions). Also, it now recognizes conditions like "pam" even when
2708 the relevant support is not compiled in: a suitably worded error message is
2709 given if an attempt is made to use such a condition.
2711 11. Added ${time_interval:xxxxx}.
2713 12. A bug was causing one of the ddress fields not to be passed back correctly
2714 from remote delivery subprocesses. The field in question was not being
2715 subsequently used, so this caused to problems in practice.
2717 13. Added new log selectors queue_time and deliver_time.
2719 14. Might have fixed a bug in maildirsizefile handling that threw up
2720 "unexpected character" debug warnings, and recalculated the data
2721 unnecessarily. In any case, I expanded the warning message to give more
2724 15. Added the message "Restricted characters in address" to the statements in
2725 the default ACL that block characters like @ and % in local parts.
2727 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2728 Three changes have been made:
2730 (a) There was a serious bug; a negative response to MAIL caused the whole
2731 recipient domain to be cached as invalid, thereby blocking all messages
2732 to all local parts at the same domain, from all senders. This bug has
2733 been fixed. The domain is no longer cached after a negative response to
2734 MAIL if the sender used is not empty.
2736 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2739 (c) A new callout option, "use_sender" has been added for people who want
2740 the modified behaviour.
2746 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2749 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2750 indeed breaks things for older releases.
2752 3. Added additional logging to the case where there is a problem reading data
2753 from a filter that is running in a subprocess using a pipe, in order to
2754 try to track down a specific problem.
2756 4. Testing facility fudge: when running in the test harness and attempting
2757 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2758 getting "No route to host". Convert this to a timeout.
2760 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2763 6. Some OS don't have socklen_t but use size_t instead. This affects the
2764 fifth argument of getsockopt() amongst other things. This is now
2765 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2766 can be set for individual OS. I have set it for SunOS5, OSF1, and
2767 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2768 some earlier ones do not.
2770 7. Change 4.30/15 was not doing the test caselessly.
2772 8. The standard form for an IPv6 address literal was being rejected by address
2773 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2774 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2775 this, as well as the form without the "IPv6" on the front (but only when
2776 address literals are enabled, of course).
2778 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2780 10. Exim crashed if a message with an empty sender address specified by -f
2781 encountered a router with an errors_to setting. This could be provoked only
2782 by a command such as
2786 where an empty string was supplied; "<>" did not hit this bug.
2788 11. Installed PCRE release 4.5.
2790 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2791 remained set. It is now erased.
2793 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2794 times from message ids (which are base 36 rather than the normal 62).
2796 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2797 were being counted as actual protocol errors, and logged if the log
2798 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2799 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2800 and DATA following a set of rejected RCPTs do not count as protocol errors.
2801 In other words, Exim assumes they were pipelined, though this may not
2802 actually be the case. Of course, in all cases the client gets an
2803 appropriate error code.
2805 15. If a lookup fails in an ACL condition, a message about the failure may
2806 be available; it is used if testing the ACL cannot continue, because most
2807 such messages specify what the cause of the deferral is. However, some
2808 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2809 that caused an old message to be retained and used if a later statement
2810 caused a defer, replacing the real cause of the deferral.
2812 16. If an IP address had so many PTR records that the DNS lookup buffer
2813 was not large enough to hold them, Exim could crash while trying to process
2814 the truncated data. It now detects and logs this case.
2816 17. Further to 4.21/58, another change has been made: if (and only if) the
2817 first line of a message (the first header line) ends with CRLF, a bare LF
2818 in a subsequent header line has a space inserted after it, so as not to
2819 terminate the header.
2821 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2822 unnecessarily, used atoi() instead of strtol(), and didn't check the
2823 termination when getting file sizes from file names by regex.
2825 19. Completely re-implemented the support for maildirsize files, in the light
2826 of a number of problems with the previous contributed implementation
2827 (4.30/29). In particular:
2829 . If the quota is zero, the maildirsize file is maintained, but no quota is
2832 . If the maildir directory does not exist, it is created before any attempt
2833 to write a maildirsize file.
2835 . The quota value in the file is just a cache; if the quota is changed in
2836 the transport, the new value overrides.
2838 . A regular expression is available for excluding directories from the
2841 20. The autoreply transport checks the characters in options that define the
2842 message's headers; it allows continued headers, but it was checking with
2843 isspace() after an embedded newline instead of explicitly looking for a
2846 21. If all the "regular" hosts to which an address was routed had passed their
2847 expiry times, and had not reached their retry times, the address was
2848 bounced, even if fallback hosts were defined. Now Exim should go on to try
2851 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2852 equivalent code in the SMTP transport. Some hosts send humungous responses
2853 to HELO/EHLO, more than 1024 it seems.
2855 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2856 gives compiler warnings in some environments. I've now done it "properly",
2859 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2860 (because of problems with the built-in one) was declared to return uschar *
2861 instead of char *, causing compiler failure.
2863 25. Fixed a file descriptor leak when processing alias/forward files.
2865 26. Fixed a minor format string issue in dbfn.c.
2867 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2869 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2870 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2873 29. When a sender address is verified, it is cached, to save repeating the test
2874 when there is more than one recipient in a message. However, when the
2875 verification involves a callout, it is possible for different callout
2876 options to be set for different recipients. It is too complicated to keep
2877 track of this in the cache, so now Exim always runs a verification when a
2878 callout is required, relying on the callout cache for the optimization.
2879 The overhead is duplication of the address routing, but this should not be
2882 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2883 to be verified with callout=postmaster, and the main callout worked but the
2884 postmaster check failed, the verification correctly failed. However, if a
2885 subsequent RCPT command asked for sender verification *without* the
2886 postmaster check, incorrect caching caused this verification also to fail,
2889 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2890 it was not caching the DNS options (qualify_single, search_parents) that
2891 were used when the lookup failed. A subsequent lookup with different
2892 options therefore always gave the same answer, though there were cases
2893 where it should not have. (Example: a "domains = !$mx_any" option on a
2894 dnslookup router: the "domains" option is always processed without any
2895 widening, but the router might have qualify_single set.) Now Exim uses the
2896 cached value only when the same options are set.
2898 32. Added John Jetmore's "exipick" utility to the distribution.
2900 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2901 than a timeout (e.g. a certificate is required, and is not provided), an
2902 Exim server now closes the connection immediately. Previously it waited for
2903 the client to close - but if the client is SSL, it seems that they each
2904 wait for each other, leading to a delay before one of them times out.
2906 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2907 maintained 0.8.x compatibility because I don't think many are using it, and
2908 it is clearly obsolete.
2910 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2913 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2914 client certificate was expired. A simple patch fixes this, though I don't
2915 understand the full logic of why the verify callback is called multiple
2918 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2919 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2920 which causes problems with some clients (such as the Certicom SSL Plus
2921 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2922 disables the coutermeasure allowing Eudora to connect."
2924 38. Exim was not checking that a write() to a log file succeeded. This could
2925 lead to Bad Things if a log got too big, in particular if it hit a file
2926 size limit. Exim now panics and dies if it cannot write to a log file, just
2927 as it does if it cannot open a log file.
2929 39. Modified OS/Makefile-Linux so that it now contains
2931 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2933 The two -D definitions ensure that Exim is compiled with large file
2934 support, which makes it possible to handle log files that are bigger than
2937 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2938 instance) a domain was checked against a named list that involved a lookup,
2939 causing $domain_data to be set, then another domain was checked against the
2940 same list, then the first domain was re-checked, the value of $domain_data
2941 after the final check could be wrong. In particular, if the second check
2942 failed, it could be set empty. This bug probably also applied to
2945 41. The strip_trailing_dot option was not being applied to the address given
2946 with the -f command-line option.
2948 42. The code for reading a message's header from the spool was incrementing
2949 $received_count, but never initializing it. This meant that the value was
2950 incorrect (doubled) while delivering a message in the same process in which
2951 it was received. In the most common configuration of Exim, this never
2952 happens - a fresh exec is done - but it can happen when
2953 deliver_drop_privilege is set.
2955 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2956 it now includes up to 150 characters of the unexpected data in the log
2959 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2960 and building data strings. The size of both of these buffers was 10 000
2961 bytes - far larger than anybody would *ever* want, thought I. Needless to
2962 say, somebody hit the limit. I have increased the maximum line length to
2963 20 000 and the maximum data length of concatenated lines to 100 000. I have
2964 also fixed two bugs, because there was no checking on these buffers. Tsk,
2965 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2966 buffer is too small.
2968 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2969 lsearch lookups. Now it does.
2971 46. When parsing a route_list item in a manualroute router, a fixed-length
2972 buffer was used for the list of hosts. I made this 1024 bytes long,
2973 thinking that nobody would ever have a list of hosts that long. Wrong.
2974 Somebody had a whole pile of complicated expansion conditions, and the
2975 string was silently truncated, leading to an expansion error. It turns out
2976 that it is easier to change to an unlimited length (owing to other changes
2977 that have happened since this code was originally written) than to build
2978 structure for giving a limitation error. The length of the item that
2979 expands into the list of hosts is now unlimited.
2981 47. The lsearch lookup could not handle data where the length of text line was
2982 more than 4095 characters. Such lines were truncated, leading to shortened
2983 data being returned. It should now handle lines of any length.
2985 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2986 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2989 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2990 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2991 no longer happens in comment lines. A list of replacements is now placed
2992 at the head of all of the source files, except those whose only change is
2993 to replace PERL_COMMAND in the very first #! line.
2995 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2996 messages on the queue, with a bottom-up merge sort, using code contributed
2997 by Michael Haardt. This should make operations like -bp somewhat faster on
2998 large queues. It won't affect queue runners, except when queue_run_in_order
3001 51. Installed eximstats 1.31 in the distribution.
3003 52. Added support for SRV lookups to the dnslookup router.
3005 53. If an ACL referred to $message_body or $message_body_end, the value was not
3006 reset for any messages that followed in the same SMTP session.
3008 54. The store-handling optimization for building very long strings was not
3009 differentiating between the different store pools. I don't think this
3010 actually made any difference in practice, but I've tidied it.
3012 55. While running the routers to verify a sender address, $sender_address
3013 was still set to the sender address. This is wrong, because when routing to
3014 send a bounce to the sender, it would be empty. Therefore, I have changed
3015 it so that, while verifying a sender address, $sender_address is set to <>.
3016 (There is no change to what happens when verifying a recipient address.)
3018 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3019 target A or AAAA records (if not already returned) without resetting the
3020 qualify_single or search_parents options of the DNS resolver. These are
3021 inappropriate in this case because the targets of MX and SRV records must
3022 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3023 target that, when qualified, matched something in the local domain. These
3024 two options are now turned off when doing these lookups.
3026 57. It seems that at least some releases of Reiserfs (which does not have the
3027 concept of a fixed number of inodes) returns zero and not -1 for the
3028 number of available inodes. This interacted badly with check_spool_inodes,
3029 which assumed that -1 was the "no such thing" setting. What I have done is
3030 to check that the total number of inodes is greater than zero before doing
3031 the test of how many are available.
3033 58. When a "warn" ACL statement has a log_message modifier, the message is
3034 remembered, and not repeated. This is to avoid a lot of repetition when a
3035 message has many recipients that cause the same warning to be written.
3036 Howewer, Exim was preserving the list of already written lines for an
3037 entire SMTP session, which doesn't seem right. The memory is now reset if a
3038 new message is started.
3040 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3041 debugging output unless log_rewrite was also set.
3043 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3044 of (int)(handle) when we know that handle contains (void *)(-1).
3046 61. The Exim daemon panic-logs an error return when it closes the incoming
3047 connection. However "connection reset by peer" seems to be common, and
3048 isn't really an error worthy of noting specially, so that particular error
3051 62. When Exim is trying to find all the local interfaces, it used to panic and
3052 die if the ioctl to get the interface flags failed. However, it seems that
3053 on at least one OS (Solaris 9) it is possible to have an interface that is
3054 included in the list of interfaces, but for which you get a failure error
3055 for this call. This happens when the interface is not "plumbed" into a
3056 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3057 failure of the "get flags" call assumes that the interface is down.
3059 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3060 makes life easier for people who are doing arithmetic on fields extracted
3061 from dates, where you often get leading zeros that should not be
3062 interpreted as octal.
3064 64. Added qualify_domain to the redirect router, to override the global
3067 65. If a pathologically long header line contained very many addresses (the
3068 report of this problem mentioned 10 000) and each of them was rewritten,
3069 Exim could use up a very large amount of memory. (It kept on making new
3070 copies of the header line as it rewrote, and never released the old ones.)
3071 At the expense of a bit more processing, the header rewriting function has
3072 been changed so that it no longer eats memory in this way.
3074 66. The generation of the Received: header has been moved from the time that a
3075 message starts to be received, to the time that it finishes. The timestamp
3076 in the Received: header should now be very close to that of the <= log
3077 line. There are two side-effects of this change:
3079 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3080 logged header lines no longer include the local Received: line, because
3081 it has not yet been created. The same applies to a copy of the message
3082 that is returned to a non-SMTP sender when a message is rejected.
3084 (b) When a filter file is tested using -bf, no additional Received: header
3085 is added to the test message. After some thought, I decided that this
3088 This change does not affect the value of $received_for. It is still set
3089 after address rewriting, but before local_scan() is called.
3091 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3093 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3094 gave an unhelpful panic error message, and a defer error. I have managed to
3095 change this behaviour so that it now rejects any supplied certificate,
3096 which seems right, as the list of acceptable certificates is empty.
3098 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3099 gave an unhelpful defer error. I have not managed to make this reject any
3100 supplied certificates, but the error message it gives is "no certificate
3101 supplied", which is not helpful.
3103 70. exigrep's output now also includes lines that are not associated with any
3104 message, but which match the given pattern. Implemented by a patch from
3105 Martin Sluka, which also tidied up the Perl a bit.
3107 71. Recipient callout verification, like sender verification, was using <> in
3108 the MAIL FROM command. This isn't really the right thing, since the actual
3109 sender may affect whether the remote host accepts the recipient or not. I
3110 have changed it to use the actual sender in the callout; this means that
3111 the cache record is now keyed on a recipient/sender pair, not just the
3112 recipient address. There doesn't seem to be a real danger of callout loops,
3113 since a callout by the remote host to check the sender would use <>.
3114 [SEE ABOVE: changed after hitting problems.]
3116 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3117 temporary errors. However, in the case of such a code being given after
3118 the end of a data transmission (i.e. after ".") Exim was failing to write
3119 a retry record for the message. (Yes, there was some broken host that was
3120 actually sending 8xx at this point.)
3122 73. An unknown lookup type in a host list could cause Exim to panic-die when
3123 the list was checked. (An example that provoked this was putting <; in the
3124 middle of a list instead of at the start.) If this happened during a DATA
3125 ACL check, a -D file could be left lying around. This kind of configuration
3126 error no longer causes Exim to die; instead it causes a defer errror. The
3127 incident is still logged to the main and panic logs.
3129 74. Buglet left over from Exim 3 conversion. The message "too many messages
3130 in one connection" was written to the rejectlog but not the mainlog, except
3131 when address rewriting (yes!) was being logged.
3133 75. Added write_rejectlog option.
3135 76. When a system filter was run not as root (that is, when system_filter_user
3136 was set), the values of the $n variables were not being returned to the
3137 main process; thus, they were not subsequently available in the $sn
3140 77. Added +return_path_on_delivery log selector.
3142 78. A connection timeout was being treated differently from recipients deferred
3143 when testing hosts_max_try with a message that was older than the host's
3144 retry timeout. (The host should not be counted, thus allowing all hosts to
3145 be tried at least once before bouncing.) This may have been the cause of an
3146 occasionally reported bug whereby a message would remain on the queue
3147 longer than the retry timeout, but would be bounced if a delivery was
3148 forced. I say "may" because I never totally pinned down the problem;
3149 setting up timeout/retry tests is difficult. See also the next item.
3151 79. The ultimate address timeout was not being applied to errors that involved
3152 a combination of host plus message (for example, a timeout on a MAIL
3153 command). When an address resolved to a number of possible hosts, and they
3154 were not all tried for each delivery (e.g. because of hosts_max_try), a
3155 message could remain on the queue longer than the retry timeout.
3157 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3160 81. Fixed an obscure SMTP outgoing bug which required at least the following
3161 conditions: (a) there was another message waiting for the same server;
3162 (b) the server returned 5xx to all RCPT commands in the first message so
3163 that the message was not completed; (c) the server dropped the connection
3164 or gave a negative response to the RSET that Exim sends to abort the
3165 transaction. The observed case was a dropped connection after DATA that had
3166 been sent in pipelining mode. That is, the server had advertised PIPELINING
3167 but was not implementing it correctly. The effect of the bug was incorrect
3168 behaviour, such as trying another host, and this could lead to a crash.
3174 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3175 and daemon.c were passed as pointers to ints; they should have been
3176 pointers to socklen_t variables (which are typically unsigned ints).
3178 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3181 3. Fixed a really odd bug that affected only the testing scheme; patching a
3182 certain fixed string in the binary changed the value of another string that
3183 happened to be identical to the end of the original first string.
3185 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3186 name", it returns that address as the IP address. On some operating
3187 systems (e.g. Solaris), it also passes back the IP address string as the
3188 "host name". However, on others (e.g. Linux), it passes back an empty
3189 string. Exim wasn't checking for this, and was changing the host name to an
3190 empty string, assuming it had been canonicized.
3192 5. Although rare, it is permitted to have more than one PTR record for a given
3193 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3194 all the names associated with an address, because they do in Solaris.
3195 However, it seems that they do not in Linux for data that comes from the
3196 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3197 I found this out when I moved to a new Linux workstation and tried to run
3198 the Exim test suite.
3200 To get round this problem I have changed the code so that it now does its
3201 own call to the DNS to look up PTR records when searching for a host name.
3202 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3203 addresses that are only in /etc/hosts are still found.
3205 This behaviour is, however, controlled by an option called host_lookup_
3206 order, which defaults to "bydns:byaddr". If people want to use the other
3207 order, or indeed, just use one or the other means of lookup, they can
3208 specify it in this variable.
3210 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3211 some operating systems, this comes back from gethostbyaddr() as an empty
3212 string, and this is what Exim used to test for. However, it seems that in
3213 other systems, "." is yielded. Exim now tests for this case too.
3215 7. The values of check_spool_space and check_log_space are now held internally
3216 as a number of kilobytes instead of an absolute number of bytes. If a
3217 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3218 kilobyte. This means that much larger values can be stored.
3220 8. Exim monitor: an attempt to get the action menu when not actually pointing
3221 at a message produces an empty menu entitled "No message selected". This
3222 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3223 no entries in it ("Shell widget menu has zero width and/or height"). So I
3224 have added a single, blank menu entry in this case.
3226 9. Added ${quote_local_part.
3228 10. MIME decoding is now applied to the contents of Subject: header lines when
3231 11. Now that a reference to $sender_host_address automatically causes a reverse
3232 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3233 host lookup before query-style lookups in lists that might use this
3234 variable. This has therefore been abolished, and the "net-" prefix is no
3235 longer necessary for query-style lookups.
3237 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3238 have been a typo for LFLAGS, so it has been changed.
3240 13. The install script calls Exim with "-C /dev/null" in order to find the
3241 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3242 to be output. Howeve, since Exim outputs its version number before the
3243 error, it didn't break the script. It just looked ugly. I fixed this by
3244 always allowing "-C /dev/null" if the caller is root.
3246 14. Ignore overlarge ACL variable number when reading spool file - insurance
3247 against a later release with more variables having written the file.
3249 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3250 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3251 this, as well as the form without the "IPv6" on the front.
3253 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3254 OS/Makefile-Darwin file.
3256 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3257 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3259 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3260 variable. However, some OS use ints for inodes. I've added cast to long int
3261 to get rid of the compiler warning.
3263 19. I had forgotten to lock out "/../" in configuration file names when
3264 ALT_CONFIG_PREFIX was set.
3266 20. Routers used for verification do not need to specify transports. However,
3267 if such a router generated a host list, and callout was configured, Exim
3268 crashed, because it could not find a port number from the (non-existent)
3269 transport. It now assumes port 25 in this circumstance.
3271 21. Added the -t option to exigrep.
3273 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3274 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3275 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3276 them is set, LOOKUP_LSEARCH is forced.
3278 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3280 24. Added sender and host information to the "rejected by local_scan()" log
3281 line; previously there was no indication of these.
3283 25. Added .include_if_exists.
3285 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3286 Linux. It turns out that not all file systems support this. Apparently some
3287 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3288 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3289 not supported on the file descriptor, is now ignored when Exim is trying to
3290 sync a directory. This applies only to Linux.
3292 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3294 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3295 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3298 29. Added support for maildirsize files from supplied patch (modified a bit).
3300 30. The use of :fail: followed by an empty string could lead Exim to respond to
3301 sender verification failures with (e.g.):
3303 550 Verification failed for <xxx>
3304 550 Sender verify failed
3306 where the first response line was missing the '-' that indicates it is not
3307 the final line of the response.
3309 31. The loop for finding the name of the user that called Exim had a hardwired
3310 limit of 10; it now uses the value of finduser_retries, which is used for
3311 all other user lookups.
3313 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3316 33. Exim was neglecting to zero errno before one call of strtol() when
3317 expanding a string and expecting an integer value. On some systems this
3318 resulted in spurious "integer overflow" errors. Also, it was casting the
3319 result into an int without checking.
3321 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3322 did not work. The code looks as if it has *never* worked, though it appears
3323 to have been documented since at least releast 1.62. I have made it work.
3325 35. The "timeout_DNS" error in retry rules, also documented since at least
3326 1.62, also never worked. As it isn't clear exactly what this means, and
3327 clearly it isn't a major issue, I have abolished the feature by treating it
3328 as "timeout", and writing a warning to the main and panic logs.
3330 36. The display of retry rules for -brt wasn't always showing the error code
3333 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3334 timeout_connect_A, timeout_connect_MX.
3336 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3337 to the empty sender.
3339 39. The daemon was not analysing the content of -oX till after it had closed
3340 stderr and disconnected from the controlling terminal. This meant that any
3341 syntax errors were only noted on the panic log, and the return code from
3342 the command was 0. By re-arranging the code a little, I've made the
3343 decoding happen first, so such errors now appear on stderr, and the return
3344 code is 1. However, the actual setting up of the sockets still happens in
3345 the disconnected process, so errors there are still only recorded on the
3348 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3349 connections (as happens on some IP stacks) was logged at start up time as
3350 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3351 it from "IPv6 and IPv4", which means that two separate sockets are being
3354 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3355 says whether AF_INET or AF_INET6 was passed as an argument.
3357 42. Exiwhat output was messed up when time zones were included in log
3360 43. Exiwhat now gives more information about the daemon's listening ports,
3361 and whether -tls-on-connect was used.
3363 44. The "port" option of the smtp transport is now expanded.
3365 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3366 silently ignored. Now an error message is written to the main and panic
3369 46. There's a new ACL modifier called "logwrite" which writes to a log file
3370 as soon as it is encountered.
3372 47. Added $local_user_uid and $local_user_gid at routing time.
3374 48. Exim crashed when trying to verify a sender address that was being
3377 49. Exim was recognizing only a space character after ".include". It now also
3378 recognizes a tab character.
3380 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3381 extracting the relevant information from the specification. The man page no
3382 longer contains scrambled data for the -d option, and I've added a section
3383 at the front about calling Exim under different names.
3385 51. Added "extra_headers" argument to the "mail" command in filter files.
3387 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3390 53. Installed eximstats 1.29.
3392 54. Added transport_filter_timeout as a generic transport option.
3394 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3395 Cc: header lines. This was required by RFC 822, but it not required by RFC
3398 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3399 incoming messages that did not have them. Now it does so only if the
3400 message originates locally, that is, if there is no associated remote host
3401 address. When Resent- header lines are present, this applies to the Resent-
3402 lines rather than the non-Resent- lines.
3404 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3405 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3407 58. Messages for configuration errors now include the name of the main
3408 configuration file - useful now that there may be more than one file in a
3409 list (.included file names were always shown).
3411 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3412 for those rare installations that do not start the daemon as root or run it
3413 setuid root. I've cut out the call to initgroups() if the daemon is not
3416 60. The Exim user and group can now be bound into the binary as text strings
3417 that are looked up at the start of Exim's processing.
3419 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3421 62. Added $mailstore_basename variable.
3423 63. Installed patch to sieve.c from Michael Haardt.
3425 64. When Exim failed to open the panic log after failing to open the main log,
3426 the original message it was trying to log was written to stderr and debug
3427 output, but if they were not available (the usual case in production), it
3428 was lost. Now it is written to syslog before the two lines that record the
3429 failures to open the logs.
3431 65. Users' Exim filters run in subprocesses under the user's uid. It is
3432 possible for a "deliver" command or an alias in a "personal" command to
3433 provoke an address rewrite. If logging of address rewriting is configured,
3434 this fails because the process is not running as root or exim. There may be
3435 a better way of dealing with this, but for the moment (because 4.30 needs
3436 to be released), I have disabled address rewrite logging when running a
3437 filter in a non-root, non-exim process.
3443 1. The buildconfig auxiliary program wasn't quoting the value set for
3444 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3445 not defined. This bug was masked in 4.22 by the effect that was fixed in
3448 2. Some messages that were rejected after a message id was allocated were
3449 shown as "incomplete" by exigrep. It no longer does this for messages that
3450 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3452 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3453 have allow_domain_literals set, the ID did not get logged in the <= line.
3454 Domain literals are now always recognized in Message-ID: header lines.
3456 4. The first argument for a ${extract expansion item is the key name or field
3457 number. Leading and trailing spaces in this item were not being ignored,
3458 causing some misleading effects.
3460 5. When deliver_drop_privilege was set, single queue runner processes started
3461 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3462 same command in the process it spins off) were not dropping privilege.
3464 6. When the daemon running as "exim" started a queue runner, it always
3465 re-executed Exim in the spun-off process. This is a waste of effort when
3466 deliver_drop_privilege is set. The new process now just calls the
3467 queue-runner function directly.
3473 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3476 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3477 ignored. Though the use of -oP was forcing the writing of a pid file, it
3478 was always written to the default place.
3480 3. If the message "no IP address found for host xxxx" is generated during
3481 incoming verification, it is now followed by identification of the incoming
3482 connection (so you can more easily find what provoked it).
3484 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3486 5. Added some features to "harden" Exim a bit more against certain attacks:
3488 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3489 be put in Local/Makefile. This is like the never_users runtime option,
3490 but it cannot be overridden. The default setting is "root".
3492 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3493 prefix string with which any file named in a -C command line option
3496 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3497 is retained for -C and -D only if the caller of Exim is root. Without
3498 it, the exim user may also use -C and -D and retain privilege.
3500 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3501 command line option is disabled.
3503 6. Macro names set by the -D option must start with an upper case letter, just
3504 like macro names defined in the configuration file.
3506 7. Added "dereference=" facility to LDAP.
3508 8. Two instances of the typo "uknown" in the source files are fixed.
3510 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3511 the Configure-Makefile script screwed up while processing it.
3513 10. Incorporated PCRE 4.4.
3515 11. The SMTP synchronization check was not operating right at the start of an
3516 SMTP session. For example, it could not catch a HELO sent before the client
3517 waited for the greeting. There is now a check for outstanding input at the
3518 point when the greeting is written. Because of the duplex, asynchronous
3519 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3520 way, but not yet received, when the check is performed.
3522 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3523 on TCP/IP sockets, because this apparently causes some broken clients to
3526 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3527 unchanged) from the Cygwin maintainer.
3529 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3530 was supported instead of testing for mbx. Effectively a typo.
3532 15. The spa authenticator server code was not checking that the input it
3533 received was valid base64.
3535 16. The debug output line for the "set" modifier in ACLs was not showing the
3536 name of the variable that was being set.
3538 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3540 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3541 lookup on its own. Something else was needed to trigger the lookup. For
3542 example, a match in host_lookup or the need for a host name in a host list.
3543 Now, if $sender_host_name is referenced and the host name has not yet been
3544 looked up, a lookup is performed. If the lookup fails, the variable remains
3545 empty, and $host_lookup_failed is set to "1".
3547 19. Added "eqi" as a case-independent comparison operator.
3549 20. The saslauthd authentication condition could segfault if neither service
3550 nor realm was specified.
3552 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3553 error message that was logged was misleading, and incoming SMTP
3554 connections were dropped. The message is now more accurate, and temporary
3555 errors are given to SMTP connections.
3557 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3558 (except RSET) with a 421 error, until QUIT is received. However, it was
3559 failing to send a response to QUIT.
3561 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3562 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3563 ACL is now run after the helo_try_verify_hosts code.
3565 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3566 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3567 case-independent, so other case variants are also recognized). Apparently
3568 some systems use these upper case variants.
3570 25. If more than two messages were waiting for the same host, and a transport
3571 filter was specified for the transport, Exim sent two messages over the
3572 same TCP/IP connection, and then failed with "socket operation on non-
3573 socket" when it tried to send the third.
3575 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3577 27. The extern definition of crypt16() in expand.c was not being excluded when
3578 the OS had its own crypt16() function.
3580 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3581 as a preferred synonym for return_size_limit, both as an option and as an
3584 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3586 30. Changed the default configuration ACL to relax the local part checking rule
3587 for addresses that are not in any local domains. For these addresses,
3588 slashes and pipe symbols are allowed within local parts, but the sequence
3589 /../ is explicitly forbidden.
3591 31. SPA server authentication was not clearing the challenge buffer before
3594 32. log_message in a "warn" ACL statement was writing to the reject log as
3595 well as to the main log, which contradicts the documentation and doesn't
3596 seem right (because no rejection is happening). So I have stopped it.
3598 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3599 However, I am unable to do any testing of this.
3601 34. Fixed an infelicity in the appendfile transport. When checking directories
3602 for a mailbox, to see if any needed to be created, it was accidentally
3603 using path names with one or more superfluous leading slashes; tracing
3604 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3606 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3607 contents are added to the log line that is written for every discarded
3608 recipient. (Previously a log_message setting was ignored.)
3610 36. The ${quote: operator now quotes the string if it is empty.
3612 37. The install script runs exim in order to find its version number. If for
3613 some reason other than non-existence or emptiness, which it checks, it
3614 could not run './exim', it was installing it with an empty version number,
3615 i.e. as "exim-". This error state is now caught, and the installation is
3618 38. An argument was missing from the function that creates an error message
3619 when Exim fails to connect to the socket for saslauthd authentication.
3620 This could cause Exim to crash, or give a corrupted message.
3622 39. Added isip, isip4, and isip6 to ${if conditions.
3624 40. The ACL variables $acl_xx are now saved with the message, and can be
3625 accessed later in routers, transports, and filters.
3627 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3628 strings in the file are not string-expanded.
3630 42. If a MAIL command specified a SIZE value that was too large to fit into an
3631 int variable, the check against message_size_limit failed. Such values are
3632 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3633 day this will have to be increased, but I don't think I want to be around
3634 when emails are that large.
3641 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3642 iconv() is not standard in FreeBSD.
3644 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3645 IPv6 enabled. The observed symptom was a segmentation fault on return from
3646 the function os_common_find_running_interfaces() in src/os.c.
3648 3. In the check_special_case() function in daemon.c I had used "errno" as an
3649 argument name, which causes warnings on some systems. This was basically a
3650 typo, since it was named "eno" in the comments!
3652 4. The code that waits for the clock to tick (at a resolution of some fraction
3653 of a second) so as to ensure message-id uniqueness was always waiting for
3654 at least one whole tick, when it could have waited for less. [This is
3655 almost certainly not relevant at current processor speeds, where it is
3656 unlikely to ever wait at all. But we try to future-proof.]
3658 5. The function that sleeps for a time interval that includes fractions of a
3659 second contained a race. It did not block SIGALRM between setting the
3660 timer, and suspending (a couple of lines later). If the interval was short
3661 and the sigsuspend() was delayed until after it had expired, the suspension
3662 never ended. On busy systems this could lead to processes getting stuck for
3665 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3666 process, before it forks any delivery processes. The open lookup caching
3667 mechanism meant that the open file or database connection was passed into
3668 the delivery process. The problem was that delivery processes always tidy
3669 up cached lookup data. This could cause a problem for the next delivery
3670 process started by the queue runner, because the external queue runner
3671 process does not know about the closure. So the next delivery process
3672 still has data in the lookup cache. In the case of a file lookup, there was
3673 no problem because closing a file descriptor in a subprocess doesn't affect
3674 the parent. However, if the lookup was caching a connection to a database,
3675 the connection was closed, and the second delivery process was likely to
3676 see errors such as "PGSQL: query failed: server closed the connection
3677 unexpectedly". The problem has been fixed by closing all cached lookups
3678 in a queue runner before running a delivery process.
3680 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3681 seem to have the "const" qualifier which it has on other OS. I've
3684 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3685 *of the same type* (client or server) with the same public name that an
3686 error should be diagnosed.
3688 9. When Exim looked up a host name for an IP address, but failed to find the
3689 original IP address when looking up the host name (a safety check), it
3690 output the message "<ip address> does not match any IP for NULL", which was
3691 confusing, to say the least. The bug was that the host name should have
3692 appeared instead of "NULL".
3694 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3695 user that is built into the binary, and the -C or -D options is used, root
3696 privilege is dropped before the configuration file is read. In addition,
3697 logging is switched to stderr instead of the normal log files. If the
3698 configuration then re-defines the Exim user, the unprivileged environment
3699 is probably not what is expected, so Exim logs a panic warning message (but
3702 However, if deliver_drop_privilege is set, the unprivileged state may well
3703 be exactly what is intended, so the warning has been cut out in that case,
3704 and Exim is allowed to try to write to its normal log files.
3710 1. smtp_return_error_details was not giving details for temporary sender
3711 or receiver verification errors.
3713 2. Diagnose a configuration error if two authenticators have the same public
3716 3. Exim used not to create the message log file for a message until the first
3717 delivery attempt. This could be confusing when incoming messages were held
3718 for policy or load reasons. The message log file is now created at the time
3719 the message is received, and an initial "Received" line is written to it.
3721 4. The automatically generated man page for command line options had a minor
3722 bug that caused no ill effects; however, a more serious problem was that
3723 the procedure for building the man page automatically didn't always
3724 operate. Consequently, release 4.20 contains an out-of-date version. This
3725 shouldn't happen again.
3727 5. When building Exim with embedded Perl support, the script that builds the
3728 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3729 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3731 6. The freeze_tell option was not being used for messages that were frozen on
3732 arrival, either by an ACL or by local_scan().
3734 7. Added the smtp_incomplete_transaction log selector.
3736 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3737 was accepting AUTH without a new EHLO.
3739 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3740 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3743 10. Logging of TCP/IP connections (when configured) now happens in the main
3744 daemon process instead of the child process, so that the TCP/IP connection
3745 count is more accurate (but it can never be perfect).
3747 11. The use of "drop" in a nested ACL was not being handled correctly in the
3748 outer ACL. Now, if condition failure induced by the nested "drop" causes
3749 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3750 or "require"), the connection is dropped.
3752 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3753 that yield "discard" can now be used with an "accept" or a "discard" verb,
3754 but an error is generated for any others (because I can't see a useful way
3755 to define what should happen).
3757 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3758 are now processed in the same way as lines in the Exim configuration file.
3759 In particular, continuation lines are supported.
3761 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3763 15. Added -ti meaning -t -i.
3765 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3766 domains, and warn by logging if others are found.
3768 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3769 returned from GIFCONF when Exim is trying to find the list of interfaces on
3770 a host. The code in os.c has been modified to copy each ifreq to an aligned
3771 structure in all cases.
3773 Also, in some cases, the returned ifreq's were being copied to a 'struct
3774 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3775 means the last couple of bytes of an IPv6 address could be chopped if the
3776 ifreq contained only a normal sockaddr (14 bytes storage).
3778 18. Named domain lists were not supported in the hosts_treat_as_local option.
3779 An entry such as +xxxx was not recognized, and was treated as a literal
3782 19. Ensure that header lines added by a DATA ACL are included in the reject log
3783 if the ACL subsequently rejects the message.
3785 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3786 MD5 (which is deprecated).
3788 21. When testing a filter file using -bf, Exim was writing a message when it
3789 took the sender from a "From " line in the message, but it was not doing so
3790 when it took $return_path from a Return-Path: header line. It now does.
3792 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3793 with a valid header line field name (a series of printing characters
3794 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3796 23. Changed "disc" in the source to "disk" to conform to the documentation and
3797 the book and for uniformity.
3799 24. Ignore Sendmail's -Ooption=value command line item.
3801 25. When execve() failed while trying to run a command in a pipe transport,
3802 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3803 could be confused with a return value of 69 from the command itself. This
3804 has been changed to 127, the value the shell returns if it is asked to run
3805 a non-existent command. The wording for the related log line suggests a
3806 non-existent command as the problem.
3808 26. If received_header_text expands to an empty string, do not add a Received:
3809 header line to the message. (Well, it adds a token one on the spool, but
3810 marks it "old" so that it doesn't get used or transmitted.)
3812 27. Installed eximstats 1.28 (addition of -nt option).
3814 28. There was no check for failure on the call to getsockname() in the daemon
3815 code. This can fail if there is a shortage of resources on the system, with
3816 ENOMEM, for example. A temporary error is now given on failure.
3818 29. Contrary to the C standard, it seems that in some environments, the
3819 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3820 program. Exim now does this explicitly; it affects the formatting of
3821 timestamps using strftime().
3823 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3824 complaints. I've now initialized all the variables, to avoid this.
3826 32. Header lines added by a system filter were not being "seen" during
3827 transport-time rewrites.
3829 33. The info_callback() function passed to OpenSSL is set up with type void
3830 (*)(SSL *, int, int), as described somewhere. However, when calling the
3831 function (actually a macro) that sets it up, the type void(*)() is
3832 expected. I've put in a cast to prevent warnings from picky compilers.
3834 34. If a DNS black list lookup found a CNAME record, but there were no A
3835 records associated with the domain it pointed at, Exim crashed.
3837 35. If a DNS black list lookup returned more than one A record, Exim ignored
3838 all but the first. It now scans all returned addresses if a particular IP
3839 value is being sought. In this situation, the contents of the
3840 $dnslist_value variable are a list of all the addresses, separated by a
3843 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3844 to accept a host name and all its aliases if the forward lookup for any of
3845 them yielded the IP address of the incoming connection. Now it accepts only
3846 those names whose forward lookup yields the correct IP address. Any other
3847 names are discarded. This closes a loophole whereby a rogue DNS
3848 administrator could create reverse DNS records to break through a
3849 wildcarded host restriction in an ACL.
3851 37. If a user filter or a system filter that ran in a subprocess used any of
3852 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3853 the wrong values were passed to the pipe command ($thisaddress had the
3854 value of $0, $0 had the value of $1, etc). This bug was introduced by
3855 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3857 38. Improved the line breaking for long SMTP error messages from ACLs.
3858 Previously, if there was no break point between 40 and 75 characters, Exim
3859 left the rest of the message alone. Two changes have been made: (a) I've
3860 reduced the minimum length to 35 characters; (b) if it can't find a break
3861 point between 35 and 75 characters, it looks ahead and uses the first one
3862 that it finds. This may give the occasional overlong line, but at least the
3863 remaining text gets split now.
3865 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3866 file descriptors might be low, and that setting 1000 would always raise it.
3867 It turns out that in some environments, the limit is already over 1000 and
3868 that lowering it causes trouble. So now Exim takes care not to decrease it.
3870 40. When delivering a message, the value of $return_path is set to $sender_
3871 address at the start of routing (routers may change the value). By an
3872 oversight, this default was not being set up when an address was tested by
3873 -bt or -bv, which affected the outcome if any router or filter referred to
3876 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3877 log only when "message" or "log_message" are set. However, if one of the
3878 conditions was an address verification, or a call to a nested ACL, the
3879 messages generated by the underlying test were being passed through. This
3880 no longer happens. The underlying message is available in $acl_verify_
3881 message for both "message" and "log_message" expansions, so it can be
3882 passed through if needed.
3884 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3885 new expansion $bh_ to give the encoded byte string without charset
3886 translation. Translation happens only if iconv() is available; HAVE_ICONV
3887 indicates this at build time. HEADERS_CHARSET gives the charset to
3888 translate to; headers_charset can change it in the configuration, and
3889 "headers charset" can change it in an individual filter file.
3891 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3892 that creates RFC 2047 encoded "words" labels them as that charset instead
3893 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3894 expansion operator; (ii) when Exim creates a From: line for a local
3895 message; (iii) when a header line is rewritten to include a "phrase" part.
3897 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3898 buggy, causing it to skip the first lines of messages whose message ID
3899 ended in 'D'. This would not have bitten before Exim release 4.14, because
3900 message IDs were unlikely to end in 'D' before then. The effect was to have
3901 incorrect size information for certain domains.
3903 45. #include "config.h" was missing at the start of the crypt16.c module. This
3904 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3907 46. If there was a timeout during a "random" callout check, Exim treated it as
3908 a failure of the random address, and carried on sending RSET and the real
3909 address. If the delay was just some slowness somewhere, the response to the
3910 original RCPT would be taken as a response to RSET and so on, causing
3911 mayhem of various kinds.
3913 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3914 when I implemented it. It didn't allow for the fact that some option values
3915 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3916 the right test for positive values.
3918 48. Domain names in DNS records are case-independent. Exim always looks them up
3919 in lower case. Some resolvers return domain names in exactly the case they
3920 appear in the zone file, that is, they may contain uppercase letters. Not
3921 all resolvers do this - some return always lower case. Exim was treating a
3922 change of case by a resolver as a change of domain, similar to a widening
3923 of a domain abbreviation. This triggered its re-routing code and so it was
3924 trying to route what was effectively the same domain again. This normally
3925 caused routing to fail (because the router wouldn't handle the domain
3926 twice). Now Exim checks for this case specially, and just changes the
3927 casing of the domain that it ultimately uses when it transmits the message
3930 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3933 50. If a filter generated a file delivery with a non-absolute name (possible if
3934 no home directory exists for the router), the forbid_file option was not
3937 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3938 the existing equality matching.
3940 52. Exim was using ints instead of ino_t variables in some places where it was
3941 dealing with inode numbers.
3943 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3944 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3945 called TMPDIR, and if it finds it is different, it changes its value.
3947 54. The smtp_printf() function is now made available to local_scan() so
3948 additional output lines can be written before returning. There is also an
3949 smtp_fflush() function to enable the detection of a dropped connection.
3950 The variables smtp_input and smtp_batched_input are exported to
3953 55. Changed the default runtime configuration: the message "Unknown user"
3954 has been removed from the ACL, and instead placed on the localuser router,
3955 using the cannot_route_message feature. This means that any verification
3956 failures that generate their own messages won't get overridden. Similarly,
3957 the "Unrouteable address" message that was in the ACL for unverifiable
3958 relay addresses has also been removed.
3960 56. Added hosts_avoid_esmtp to the smtp transport.
3962 57. The exicyclog script was not checking for the esoteric option
3963 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3964 will work only if exicyclog is run under the appropriate euid.
3966 58. Following a discussion on the list, the rules by which Exim recognises line
3967 endings on incoming messages have been changed. The -dropcr and drop_cr
3968 options are now no-ops, retained only for backwards compatibility. The
3969 following line terminators are recognized: LF CRLF CR. However, special
3970 processing applies to CR:
3972 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3973 nor a local message in the state where . is a terminator.
3975 (ii) If a bare CR is encountered in a header line, an extra space is added
3976 after the line terminator so as not to end the header. The reasoning
3977 behind this is that bare CRs in header lines are most likely either
3978 to be mistakes, or people trying to play silly games.
3980 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3981 was being incorrectly given as 18 bytes larger than it should have been.
3982 This is a VOB (very old bug).
3984 60. This may never have affected anything current, but just in case it has:
3985 When the local host is found other than at the start of a list of hosts,
3986 the local host, those with the same MX, and any that follow, are discarded.
3987 When the list in question was part of a longer list of hosts, the following
3988 hosts (not currently being processed) were also being discarded. This no
3989 longer happens. I'm not sure if this situation could ever has previously
3992 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3995 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3996 header. This is something that is recommended in a new Internet Draft, and
3997 is something that is documented as being done by Sendmail. There are two
3998 possible values. For messages generated by the autoreply transport, Exim
4001 Auto-Submitted: auto-replied
4003 whereas for all other generated messages (e.g. bounces) it adds
4005 Auto-Submitted: auto-generated
4007 63. The "personal" condition in filters now includes a test for the
4008 Auto-Submitted: header. If it contains the string "auto-" the message it
4009 not considered personal.
4011 64. Added rcpt_include_affixes as a generic transport option.
4013 65. Added queue_only_override (default true).
4015 66. Added the syslog_duplication option.
4017 67. If what should have been the first header line of a message consisted of
4018 a space followed by a colon, Exim was mis-interpreting it as a header line.
4019 It isn't of course - it is syntactically invalid and should therefore be
4020 treated as the start of the message body. The misbehaviour could have
4021 caused a number of strange effects, including loss of data in subsequent
4022 header lines, and spool format errors.
4024 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4025 client host had authenticated. This control can now be exercised by an ACL
4026 for more flexibility.
4028 69. By default, callouts do not happen when testing with -bh. There is now a
4029 variant, -bhc, which does actually run the callout code, including
4030 consulting and updating the callout cache.
4032 70. Added support for saslauthd authentication, courtesy of Alexander
4035 71. If statvfs() failed on the spool or log directories while checking their
4036 size for availability, Exim confusingly gave the error "space shortage".
4037 Furthermore, in debugging mode it crashed with a floating point exception.
4038 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4039 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4040 serious problem, Exim now writes to the main and panic logs when this
4041 happens, with details of the failure. It then refuses to accept the
4042 incoming message, giving the message "spool directory problem" or "log
4043 directory problem" with a 421 code for SMTP messages.
4045 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4046 0, 1, and 2 exist, because some OS complain for execs without them (see
4047 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4048 descriptors. However, the code omitted to check that the open succeeded,
4049 causing mysterious errors if for some reason the permissions on /dev/null
4050 got screwed. Now Exim writes a message to the main and panic logs, and
4051 bombs out if it can't open /dev/null.
4053 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4054 interact so that it is all more flexible. It is supposed to remain
4055 backwards compatible. Also added extra_local_interfaces.
4057 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4058 to bomb out with an assertion failure - to the client this appears as a
4059 connection drop. This problem occurs in the part of the code that was taken
4060 from the Samba project. Fortunately, the assertion is in a very simple
4061 function, so I have fixed this by reproducing the function inline in the
4062 one place where it is called, and arranging for authentication to fail
4063 instead of killing the process with assert().
4065 75. The SPA client code was not working when the server requested OEM rather
4066 than Unicode encoding.
4068 76. Added code to make require_files with a specific uid setting more usable in
4069 the case where statting the file as root fails - usually a non-root-mounted
4070 NFS file system. When this happens and the failure is EACCES, Exim now
4071 forks a subprocess and does the per-uid checking as the relevant uid.
4073 77. Added process_log_path.
4075 78. If log_file_path was not explicitly set, a setting of check_log_space or
4076 check_log_inodes was ignored.
4078 79. If a space check for the spool or log partitions fails, the incident is now
4079 logged. Of course, in the latter case the data may get lost...
4081 80. Added the %p formatting code to string_format() so that it can be used to
4082 print addresses in debug_print(). Adjusted all the address printing in the
4083 debugging in store.c to use %p rather than %d.
4085 81. There was a concern that a line of code in smtp_in.c could overflow a
4086 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4087 initially expressed, the concern was not well-founded, because trailing
4088 spaces are removed early. However, if the trailing spaces were followed by
4089 a NULL, they did not get removed, so the overflow was possible. Two fixes
4092 (a) I re-wrote the offending code in a cleaner fashion.
4093 (b) If an incoming SMTP command contains a NULL character, it is rejected
4096 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4097 runs initgroups(), so that if the Exim user is in any additional groups,
4098 they will be used during message reception.
4104 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob