1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.379 2006/07/17 09:18:09 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 SC/01 Use a glob alias rather than an array ref in eximstats generated
10 parser. This improves both readability and performance.
12 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
13 Don't display local sender or destination tables in eximstats unless
14 there is data to show.
15 Added average volumes into the eximstats top table text output.
17 SC/03 Collect data on the number of addresses (recipients) as well
18 as the number of messages in eximstats.
20 TF/01 Correct an error in the documentation for the redirect router. Exim
21 does (usually) call initgroups() when daemonizing.
23 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
24 with consistent privilege compared to when running as a daemon.
26 TF/03 Note in the spec that $authenticated_id is not set for local
27 submissions from trusted users.
29 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
30 Thanks to Dean Brooks <dean@iglou.com> for the patch.
32 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
33 by adding some example configuration directives to the default
34 configuration file. A little bit of work is required to uncomment the
35 directives and define how usernames and passwords are checked, but
36 there is now a framework to start from.
38 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
39 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
40 without this. I don't know how relevant this is to other LDAP libraries.
42 PH/02 Add the verb name to the "unknown ACL verb" error.
44 PH/03 Magnus Holmgren's patch for filter_prepend_home.
46 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
48 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
49 directory not expanded when it should be if an expanded home directory
50 was set for the address (which is overridden by the transport).
52 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
55 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
56 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
57 because it is too late at that time, and has no effect.
59 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
60 security issue with \' (bugzilla #107). I could not use the
61 PQescapeStringConn() function, because it needs a PGconn value as one of
64 PH/08 When testing addresses using -bt, indicate those final addresses that
65 are duplicates that would not cause an additional delivery. At least one
66 person was confused, thinking that -bt output corresponded to deliveries.
67 (Suppressing duplicates isn't a good idea as you lose the information
68 about possibly different redirections that led to the duplicates.)
70 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
71 systems where poll() doesn't work, in particular OS X.
73 PH/10 Added more information to debugging output for retry time not reached.
75 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
76 operations in malware.c.
78 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
81 PH/13 If write_rejectlog was set false when logging was sent to syslog with
82 syslog_duplication set false, log lines that would normally be written
83 both the the main log and to the reject log were not written to syslog at
86 PH/14 In the default configuration, change the use of "message" in ACL warn
87 statements to "add_header".
89 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
90 not followed by a command (e.g. "seen endif").
92 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
93 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
96 PH/17 Added an extra condition to the default value of delay_warning_condition
100 { match{$h_precedence:}{(?i)bulk|list|junk} }
101 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} }
104 The Auto-Submitted: header is standardised, whereas I don't think
105 Precedence: ever was.
107 PH/18 Refactored debugging code in route_finduser() to show more information,
108 in particular, the error code if getpwnam() issues one.
110 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
111 This is apparently needed in addition to the PH/07 change above to avoid
112 any possible encoding problems.
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
128 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
129 other effects) broke the use of negated acl sub-conditions.
131 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
134 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
135 "Deny" causes Exim to reject the incoming connection with a 554 error.
136 Unfortunately, if there is a major crisis, such as a disk failure,
137 tcp-wrappers gives "deny", whereas what one would like would be some
138 kind of temporary error. A kludge has been added to help with this.
139 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
140 554 error is used if errno is still zero or contains ENOENT (which occurs
141 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
144 PH/03 Add -lutil to the default FreeBSD LIBS setting.
146 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
147 errors. Otherwise a message that provokes a temporary error (when other
148 messages do not) can cause a whole host to time out.
150 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
151 addresses were routed directly to files or pipes from a redirect router.
152 File deliveries just didn't batch; pipe deliveries might have suffered
155 PH/06 A failure to get a lock for a hints database would erroneously always say
156 "Failed to get write lock", even when it was really a read lock.
158 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
159 of 0600. This has been changed to use the value of the lockfile_mode
160 option (which defaults to 0600).
162 PH/08 Applied small patch from the Sieve maintainer.
164 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
165 folder from quota calculations, a direct delivery into this folder messed
166 up the contents of the maildirsize file. This was because the regex was
167 used only to exclude .Trash (or whatever) when the size of the mailbox
168 was calculated. There was no check that a delivery was happening into an
169 excluded directory. This bug has been fixed by ignoring all quota
170 processing for deliveries into excluded directories.
172 PH/10 Added the maildirfolder_create_regex option to appendfile.
178 PH/01 The code for finding all the local interface addresses on a FreeBSD
179 system running IPv6 was broken. This may well have applied to all BSD
180 systems, as well as to others that have similar system calls. The broken
181 code found IPv4 interfaces correctly, but gave incorrect values for the
182 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
183 that it would not match correctly against @[] and not recognize the IPv6
186 PH/02 The ipliteral router was not recognizing addresses of the form user@
187 [ipv6:....] because it didn't know about the "ipv6:" prefix.
189 PH/03 Added disable_ipv6.
191 PH/04 Changed $reply_address to use the raw form of the headers instead of the
192 decoded form, because it is most often used to construct To: headers
193 lines in autoreplies, and the decoded form may well be syntactically
194 invalid. However, $reply_address has leading white space removed, and all
195 newlines turned into spaces so that the autoreply transport does not
198 PH/05 If group was specified without a user on a router, and no group or user
199 was specified on a transport, the group from the router was ignored.
201 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
202 for visible compile-time settings that can be used to change these
203 numbers, for those that want even more. Backwards compatibility with old
204 spool files has been maintained. However, going back to a previous Exim
205 release will lost any variables that are in spool files.
207 PH/07 Two small changes when running in the test harness: increase delay when
208 passing a TCP/IP connection to a new process, in case the original
209 process has to generate a bounce, and remove special handling of
210 127.0.0.2 (sic), which is no longer necessary.
212 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
213 be the same on different OS.
215 PH/09 Moved a debug statement in filter processing to avoid a race problem when
218 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
219 whether --show-vars was specified or not
221 JJ/02 exipick: Added support for new ACL variable spool format introduced
224 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
225 syntactically invalid From: or Reply-to: line, and a filter used this to
226 generate an autoreply, and therefore failed to obtain an address for the
227 autoreply, Exim could try to deliver to a non-existent relative file
228 name, causing unrelated and misleading errors. What now happens is that
229 it logs this as a hard delivery error, but does not attempt to create a
232 PH/11 The exinext utility has a -C option for testing purposes, but although
233 the given file was scanned by exinext itself; it wasn't being passed on
234 when Exim was called.
236 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
237 an end-of-file indication when reading a command response.
239 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
240 compiled. In many other places in Exim, IPv6 addresses are always
241 recognized, so I have changed this. It also means that IPv4 domain
242 literals of the form [IPV4:n.n.n.n] are now always recognized.
244 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
245 used if the router is not running as root, for example, when verifying at
246 ACL time, or when using -bh. The debugging output from this situation was
247 non-existent - all you got was a failure to exec. I have made two
250 (a) Failures to set uid/gid, the current directory, or a process leader
251 in a subprocess such as that created by queryprogram now generate
252 suitable debugging ouput when -d is set.
254 (b) The queryprogram router detects when it is not running as root,
255 outputs suitable debugging information if -d is set, and then runs
256 the subprocess without attempting to change uid/gid.
258 PH/15 Minor change to Makefile for building test_host (undocumented testing
261 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
262 additional section of a DNS packet that returns MX or SRV records.
263 Instead, it always explicitly searches for A/AAAA records. This avoids
264 major problems that occur when a DNS server includes only records of one
265 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
266 fixed another bug: if SRV records were looked up and the corresponding
267 address records were *not* found in the additional section, the port
268 values from the SRV records were lost.
270 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
271 using the correct key (the original address) when searching the retry
272 rules in order to find which one to use for generating the retry hint.
274 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
275 adding the default one. Similarly, if it contains a Reply-To: header, the
276 errors_reply_to option, if set, is not used.
278 PH/19 When calculating a retry time, Exim used to measure the "time since
279 failure" by looking at the "first failed" field in the retry record. Now
280 it does not use this if it is later than than the arrival time of the
281 message. Instead it uses the arrival time. This makes for better
282 behaviour in cases where some deliveries succeed, thus re-setting the
283 "first failed" field. An example is a quota failure for a huge message
284 when small messages continue to be delivered. Without this change, the
285 "time since failure" will always be short, possible causing more frequent
286 delivery attempts for the huge message than are intended.
287 [Note: This change was subsequently modified - see PH/04 for 4.62.]
289 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
290 $1, $2, $3) because the numerical variables can be reset during some
291 expansion items (e.g. "match"), thereby losing the authentication data.
293 PH/21 Make -bV show the size of off_t variables so that the test suite can
294 decide whether to run tests for quotas > 2G.
296 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
297 mailbox_size, and mailbox_filecount in the appendfile transport. If a
298 filecount value is greater than 2G or if a quota value is greater than 2G
299 on a system where the size of off_t is not greater than 4, a panic error
302 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
303 never match. The debug and -bh output now contains an explicit error
304 message indicating a malformed IPv4 address or mask.
306 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
307 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
310 PH/25 Do not write to syslog when running in the test harness. The only
311 occasion when this arises is a failure to open the main or panic logs
312 (for which there is an explicit test).
314 PH/26 Added the /no_tell option to "control=freeze".
316 PH/27 If a host name lookup failed very early in a connection, for example, if
317 the IP address matched host_lookup and the reverse lookup yielded a name
318 that did not have a forward lookup, an error message of the form "no IP
319 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
320 could be logged. Now it outputs the IP address instead of "NULL".
322 PH/28 An enabling patch from MH: add new function child_open_exim2() which
323 allows the sender and the authenticated sender to be set when
324 submitting a message from within Exim. Since child_open_exim() is
325 documented for local_scan(), the new function should be too.
327 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
328 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
329 results in an empty string is now treated as unset.
331 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
333 PH/31 Added sender_verify_fail logging option.
335 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
336 needed by RFC 822 but not by RFC 2822 was commented out. I have now
337 tidied the source and removed it altogether.
339 PH/33 When a queue run was abandoned because the load average was too high, a
340 log line was always written; now it is written only if the queue_run log
341 selector is set. In addition, the log line for abandonment now contains
342 information about the queue run such as the pid. This is always present
343 in "start" and "stop" lines but was omitted from the "abandon" line.
345 PH/34 Omit spaces between a header name and the colon in the error message that
346 is given when verify = headers_syntax fails (if there are lots of them,
347 the message gets confusing).
349 PH/35 Change the default for dns_check_names_pattern to allow slashes within
350 names, as there are now some PTR records that contain slashes. This check
351 is only to protect against broken name servers that fall over on strange
352 characters, so the fact that it applies to all lookups doesn't matter.
354 PH/36 Now that the new test suite is complete, we can remove some of the
355 special code in Exim that was needed for the old test suite. For example,
356 sorting DNS records because real resolvers return them in an arbitrary
357 order. The new test suite's fake resolver always returns records in the
360 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
361 bounces) except when queue_only is set, to avoid logging races between
362 the different processes.
364 PH/38 Panic-die if .include specifies a non-absolute path.
366 PH/39 A tweak to the "H" retry rule from its user.
368 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
369 a label. They prevented compilation on older perls.
371 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
372 a warning to be raised on newish perls.
374 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
375 on queue. Changes to match documented behaviour of showing count of
376 messages matching specified criteria.
378 PH/40 Changed the default ident timeout from 30s to 5s.
380 PH/41 Added support for the use of login_cap features, on those BSD systems
381 that have them, for controlling the resources used by pipe deliveries.
383 PH/42 The content-scanning code uses fopen() to create files in which to put
384 message data. Previously it was not paying any attention to the mode of
385 the files. Exim runs with umask(0) because the rest of the code creates
386 files with open(), and sets the required mode explicitly. Thus, these
387 files were ending up world-writeable. This was not a big issue, because,
388 being within the spool directory, they were not world-accessible. I have
389 created a function called modefopen, which takes an additional mode
390 argument. It sets umask(777), creates the file, chmods it to the required
391 mode, then resets the umask. All the relevant calls to fopen() in the
392 content scanning code have been changed to use this function.
394 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
395 to 24 hours. This avoids potential overflow problems when processing G
396 and H retry rules. I suspect nobody ever tinkers with this value.
398 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
400 PH/45 When the plaintext authenticator is running as a client, the server's
401 challenges are checked to ensure they are valid base64 strings. By
402 default, the authentication attempt is cancelled if an invalid string is
403 received. Setting client_ignore_invalid_base64 true ignores these errors.
404 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
405 they are received. Thus, the responses can be made to depend on the
406 challenges. If an invalid string is ignored, an empty string is placed in
409 PH/46 Messages that are created by the autoreply transport now contains a
410 References: header, in accordance with RFCs 2822 and 3834.
412 PH/47 Added authenticated_sender_force to the smtp transport.
414 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
416 PH/49 Installed latest patch from the Sieve maintainer.
418 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
419 was also set, the appendfile transport was unnecessarily scanning a
420 directory of message files (e.g. for maildir delivery) to find the count
421 of files (along with the size), even though it did not need this
422 information. It now does the scan only if it needs to find either the
423 size of the count of files.
425 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
427 PH/52 Two bugs concerned with error handling when the smtp transport is
430 (i) Exim was not creating retry information for temporary errors given
431 for individual recipients after the DATA command when the smtp transport
432 was used in LMTP mode. This meant that they could be retried too
433 frequently, and not timed out correctly.
435 (ii) Exim was setting the flag that allows error details to be returned
436 for LMTP errors on RCPT commands, but not for LMTP errors for individual
437 recipients that were returned after the DATA command.
439 PH/53 This is related to PH/52, but is more general: for any failing address,
440 when detailed error information was permitted to be returned to the
441 sender, but the error was temporary, then after the final timeout, only
442 "retry timeout exceeded" was returned. Now it returns the full error as
443 well as "retry timeout exceeded".
445 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
446 do this, and (what is worse) MTAs that accept it.
448 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
449 will now be deprecated.
451 PH/56 New os.c-cygwin from the Cygwin maintainer.
453 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
454 formats (previously only available in exim formats via -bpr, -bpru,
455 and -bpra. Now also available in native and exiqgrep formats)
457 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
458 with very large, slow to parse queues
460 JJ/08 exipick: added ! as generic prefix to negate any criteria format
462 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
464 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
465 responses to authentication challenges, though it was showing the
466 challenges; (ii) I've removed the CR characters from the debug output for
469 PH/58 Allow for the insertion of a newline as well as a space when a string
470 is turned into more than one encoded-word during RFC 2047 encoding. The
471 Sieve code now uses this.
473 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
474 data_4xx, lost_connection, tls_required.
476 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
477 message was being sent as an SMTP response.
479 PH/61 Add -l and -k options to exicyclog.
481 PH/62 When verifying, if an address was redirected to one new address, so that
482 verification continued, and the new address failed or deferred after
483 having set something in $address_data, the value of $address_data was not
484 passed back to the ACL. This was different to the case when no
485 redirection occurred. The value is now passed back in both cases.
487 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
488 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
489 use login_cap.h, so on its own it isn't the distinguishing feature. The
490 new name refers directly to the setclassresources() function.
492 PH/65 Added configuration files for NetBSD3.
494 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
496 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
497 is preferred over IPv4.
499 PH/68 The bounce_return_message and bounce_return_body options were not being
500 honoured for bounces generated during the reception of non-SMTP messages.
501 In particular, this applied to messages rejected by the ACL. This bug has
502 been fixed. However, if bounce_return_message is true and bounce_return_
503 body is false, the headers that are returned for a non-SMTP message
504 include only those that have been read before the error was detected.
505 (In the case of an ACL rejection, they have all been read.)
507 PH/69 The HTML version of the specification is now built in a directory called
508 spec_html instead of spec.html, because the latter looks like a path with
509 a MIME-type, and this confuses some software.
511 PH/70 Catch two compiler warnings in sieve.c.
513 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
514 function verify_get_ident() calls ip_connect() to connect a socket, but
515 if the "connect()" function timed out, ip_connect() used to close the
516 socket. However, verify_get_ident() also closes the socket later, and in
517 between Exim writes to the log, which may get opened at this point. When
518 the socket was closed in ip_connect(), the log could get the same file
519 descriptor number as the socket. This naturally causes chaos. The fix is
520 not to close the socket in ip_connect(); the socket should be closed by
521 the function that creates it. There was only one place in the code where
522 this was missing, in the iplookup router, which I don't think anybody now
523 uses, but I've fixed it anyway.
525 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
526 well as to direct DNS lookups. Otherwise the handling of names in host
527 lists is inconsistent and therefore confusing.
533 PH/01 Two changes to the default runtime configuration:
535 (1) Move the checks for relay_from_hosts and authenticated clients from
536 after to before the (commented out) DNS black list checks.
538 (2) Add control=submission to the relay_from_hosts and authenticated
539 clients checks, on the grounds that messages accepted by these
540 statements are most likely to be submissions.
542 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
544 (1) Generate an error if the third argument for the ${prvs expansion is
547 (2) Treat a missing third argument of ${prvscheck as if it were an empty
550 (3) Reset the variables that are obtained from the first argument of
551 ${prvscheck and used in the second argument before leaving the code,
552 because their memory is reclaimed, so using them afterwards may do
555 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
556 one (it's much easier than Tom thought :-).
558 (5) Because of (4), we can now allow for the use of $prvscheck_result
559 inside the third argument.
561 PH/03 For some reason, the default setting of PATH when running a command from
562 a pipe transport was just "/usr/bin". I have changed it to
565 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
566 anything to be listed in the output from -bV.
568 PH/05 When a filter generated an autoreply, the entire To: header line was
569 quoted in the delivery log line, like this:
571 => >A.N.Other <ano@some.domain> <original@ddress> ...
573 This has been changed so that it extracts the operative address. There
574 may be more than one such address. If so, they are comma-separated, like
577 => >ano@some.domain,ona@other.domain <original@ddress> ...
579 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
580 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
581 not being looked up in the rDNS to get a host name, Exim was showing the
582 IP address twice in Received: lines, even though the IP addresses were
583 identical. For example:
585 Received: from [1.2.3.4] (helo=[1.2.3.4])
587 However, if the real host name was known, it was omitting the HELO data
588 if it matched the actual IP address. This has been tidied up so that it
589 doesn't show the same IP address twice.
591 PH/07 When both +timestamp and +memory debugging was on, the value given by
592 $tod_xxx expansions could be wrong, because the tod_stamp() function was
593 called by the debug printing, thereby overwriting the timestamp buffer.
594 Debugging no longer uses the tod_stamp() function when +timestamp is set.
596 PH/08 When the original message was included in an autoreply transport, it
597 always said "this is a copy of the message, including all the headers",
598 even if body_only or headers_only was set. It now gives an appropriate
601 PH/09 Applied a patch from the Sieve maintainer which:
603 o fixes some comments
604 o adds the (disabled) notify extension core
605 o adds some debug output for the result of if/elsif tests
606 o points to the current vacation draft in the documentation
607 and documents the missing references header update
611 o fixes a bug in processing the envelope test (when testing
612 multiple envelope elements, the last element determinted the
615 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
616 Electronic Mail") by including:
618 Auto-submitted: auto-generated
620 in the messages that it generates (bounce messages and others, such as
621 warnings). In the case of bounce messages for non-SMTP mesages, there was
622 also a typo: it was using "Auto_submitted" (underscore instead of
623 hyphen). Since every message generated by Exim is necessarily in response
624 to another message, thes have all been changed to:
626 Auto-Submitted: auto-replied
628 in accordance with these statements in the RFC:
630 The auto-replied keyword:
632 - SHOULD be used on messages sent in direct response to another
633 message by an automatic process,
635 - MUST NOT be used on manually-generated messages,
637 - MAY be used on Delivery Status Notifications (DSNs) and Message
638 Disposition Notifications (MDNs),
640 - MUST NOT be used on messages generated by automatic or periodic
641 processes, except for messages which are automatic responses to
644 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
645 to the default Received: header definition.
647 PH/12 Added log selector acl_warn_skipped (default on).
649 PH/13 After a successful wildlsearch lookup, discard the values of numeric
650 variables because (a) they are in the wrong storage pool and (b) even if
651 they were copied, it wouldn't work properly because of the caching.
653 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
654 checking when decoding. Apparently there are clients that generate
655 overlong encoded strings. Why am I not surprised?
657 PH/15 If the first argument of "${if match_address" was not empty, but did not
658 contain an "@" character, Exim crashed. Now it writes a panic log message
659 and treats the condition as false.
661 PH/16 In autoreply, treat an empty string for "once" the same as unset.
663 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
664 extension "envelope-auth". The code is finished and in agreement with
665 other implementations, but there is no documentation so far and in fact,
666 nobody wrote the draft yet. This extension is currently #undef'ed, thus
667 not changing the active code.
669 Print executed "if" and "elsif" statements when debugging is used. This
670 helps a great deal to understand what a filter does.
672 Document more things not specified clearly in RFC3028. I had all this
673 sorted out, when out of a sudden new issues came to my mind. Oops."
675 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
678 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
679 canonical form (as documented). However, after a host name lookup from
680 the IP address, check_host() was doing a simple string comparison with
681 addresses acquired from the DNS when checking that the found name did
682 have the original IP as one of its addresses. Since any found IPv6
683 addresses are likely to be in abbreviated form, the comparison could
684 fail. Luckily, there already exists a function for doing the comparison
685 by converting both addresses to binary, so now that is used instead of
688 PH/20 There was another similar case to PH/19, when a complete host name was
689 given in a host list; looking up its IP address could give an abbreviated
690 form, whereas the current host's name might or might not be abbreviated.
691 The same fix has been applied.
697 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
698 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
701 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
702 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
704 PH/03 Typo: missing ".o" in src/pcre/Makefile.
706 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
707 header line, restrict the check to what is listed in RFCs 2369 and 2929.
708 Also, for "Auto-Submitted", treat anything other than "no" as
709 non-personal, in accordance with RFC 3834. (Previously it treated
710 anything starting "auto-" as non-personal.)
712 TF/01 The control=submission/name=... option had a problem with syntax
713 errors if the name included a slash character. The /name= option
714 now slurps the rest of the string, so it can include any characters
715 but it must come last in the list of options (after /sender_retain
718 PH/05 Some modifications to the interface to the fake nameserver for the new
726 TK/01 Added the "success_on_redirect" address verification option. See
727 NewStuff for rationale and an example.
729 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
731 PH/02 Patch to exigrep to allow it to work on syslog lines.
733 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
734 fread() to skip over the body file's header line, because in Cygwin the
735 header line is locked and is inaccessible.
737 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
738 co-exist for some time) to make it clear that it is the Exim ID that is
739 referenced, not the Message-ID: header line.
741 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
742 string_format() function, because snprintf() does not exist on all
745 PH/06 The use of forbid_filter_existstest now also locks out the use of the
746 ${stat: expansion item.
748 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
749 protocol synchronization error", to keep the pedants happy.
751 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
752 well as for IRIX systems, when gcc is being used. See the host.c source
755 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
757 PH/10 Named domain lists were not working if used in a queue_smtp_domains
760 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
761 transport and to the smtp transport in LMTP mode.
763 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
765 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
766 run a filter in a subprocess. This could lead to confusion in subsequent
767 lookups in the parent process. There should also be a search_tidyup() at
768 the end of the subprocess.
770 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
771 only if the host matched helo_try_verify_hosts, which caused the
772 verification to occur when the EHLO/HELO command was issued. The ACL just
773 tested the remembered result. Now, if a previous verification attempt has
774 not happened, "verify = helo" does it there and then.
776 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
778 TK/03 Fix log output including CR from clamd.
780 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
781 exist provoked a memory error which could cause a segfault.
783 PH/15 Installed PCRE 6.2
785 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
787 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
788 of the problem. Specifically, suggested +O2 rather than +O1 for the
791 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
793 PH/20 If a delivery was routed to a non-standard port by means of an SRV
794 record, the port was not correctly logged when the outgoing_port log
795 selector was set (it logged the transort's default port).
797 PH/21 Added support for host-specific ports to manualroute, queryprogram,
798 fallback_hosts, and "hosts" in the smtp transport.
800 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
801 host errors such as "Connection refused".
803 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
804 authentication with radiusclient 0.4.9:
806 - Error returned from rc_read_config was caught wrongly
807 - Username/password not passed on to radius server due to wrong length.
809 The presumption is that some radiusclient API changes for 4.51/PH/17
810 were not taken care of correctly. The code is still untested by me (my
811 Linux distribution still has 0.3.2 of radiusclient), but it was
812 contributed by a Radius user.
814 PH/24 When doing a callout, the value of $domain wasn't set correctly when
815 expanding the "port" option of the smtp transport.
817 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
818 while reading a MIME header. Thanks to Tom Hughes for a patch.
820 PH/24 Include config.h inside local_scan.h so that configuration settings are
823 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
824 that in an ACL for RCPT (for example), you can examine exactly what was
827 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
828 commands, but it was not correctly comparing the address with the actual
829 client host address. Thus, it would show the EHLO address in Received:
830 header lines when this was not necessary.
832 PH/27 Added the % operator to ${eval:}.
834 PH/28 Exim tries to create and chdir to its spool directory when it starts;
835 it should be ignoring failures (because with -C, for example, it has lost
836 privilege). It wasn't ignoring creation failures other than "already
839 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
842 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
843 because an input line was too long, either on its own, or by virtue of
844 too many continuations, the temporary file was not being removed, and the
845 return code was incorrect.
847 PH/31 Missing "BOOL" in function definition in filtertest.c.
849 PH/32 Applied Sieve patches from the maintainer.
851 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
853 PH/33 Added "verify = not_blind".
855 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
856 Local/Makefile (with some defaults set). These are used in built scripts
857 such as exicyclog, but they have never been used in the exim_install
858 script (though there are many overriding facilities there). I have
859 arranged that the exim_install script now takes note of these two
862 PH/35 Installed configuration files for Dragonfly.
864 PH/36 When a locally submitted message by a trusted user did not contain a
865 From: header, and the sender address was obtained from -f or from an SMTP
866 MAIL command, and the trusted user did not use -F to supply a sender
867 name, $originator_name was incorrectly used when constructing a From:
868 header. Furthermore, $originator_name was used for submission mode
869 messages from external hosts without From: headers in a similar way,
870 which is clearly wrong.
872 PH/37 Added control=suppress_local_fixups.
874 PH/38 When log_selector = +received_sender was set, and the addition of the
875 sender made the log line's construction buffer exactly full, or one byte
876 less than full, an overflow happened when the terminating "\n" was
879 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
880 when the result of a list match is failure because a DNS lookup failed.
882 PH/40 RM_COMMAND is now used in the building process.
884 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
885 the "build-* directories that it finds.
887 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
888 address in a domain literal was a prefix of an interface address.
890 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
891 when verifying a sender address, unless rewrite_headers is false.
893 PH/44 Wrote a long comment about why errors_to addresses are verified as
894 recipients, not senders.
896 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
897 the ratelimit ACL was added.
899 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
901 PH/46 Added extra information about PostgreSQL errors to the error string.
903 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
904 avoiding the need to install special zones in a real server. This is
905 backwards compatible; if it can't find the fake resolver, it drops back.
906 Thus, both old and new test suites can be run.
908 TF/02 Added util/ratelimit.pl
910 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
911 clock is set back in time.
913 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
914 Candler <B.Candler@pobox.com>.
916 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
917 OK for addresses that are the result of redirections.
919 PH/48 A number of further additions for the benefit of the new test suite,
920 including a fake gethostbyname() that interfaces to the fake DNS resolver
923 TF/06 The fix for widen_domains has also been applied to qualify_single and
924 search_parents which are the other dnslookup options that can cause
927 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
930 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
932 TF/07 Exim produced the error message "an SRV record indicated no SMTP
933 service" if it encountered an MX record with an empty target hostname.
934 The message is now "an MX or SRV record indicated no SMTP service".
936 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
937 if the DNS of the sending site is misconfigured. This is quite a
938 common situation. This change restores the behaviour of treating a
939 helo verification defer as a failure.
941 PH/51 If self=fail was set on a router, the bounce message did not include the
942 actual error message.
948 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
950 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
951 command itself ended in error, the underlying message about the transport
952 filter timeout was being overwritten with the pipe command error. Now the
953 underlying error message should be appended to the second error message.
955 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
957 PH/02 Reduce the amount of output that "make" produces by default. Full output
958 can still be requested.
960 PH/03 The warning log line about a condition test deferring for a "warn" verb
961 was being output only once per connection, rather than after each
962 occurrence (because it was using the same function as for successful
963 "warn" verbs). This seems wrong, so I have changed it.
965 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
966 it should not have, which might have caused a crash in the right
967 circumstances, but probably never did.
969 PH/04 Installed a modified version of Tony Finch's patch to make submission
970 mode fix the return path as well as the Sender: header line, and to
971 add a /name= option so that you can make the user's friendly name appear
974 TF/03 Added the control = fakedefer ACL modifier.
976 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
977 Mark Lowes for thorough testing.
979 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
981 TK/03 Merged latest SRS patch from Miles Wilton.
983 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
984 with the definition in sysexits.h (which is #included earlier).
985 Fortunately, Exim does not actually use EX_OK. The code used to try to
986 preserve the sysexits.h value, by assumimg that macro definitions were
987 scanned for macro replacements. I have been disabused of this notion,
988 so now the code just undefines EX_OK before #including unistd.h.
990 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
991 in the smtp transport. When a block could not be written in a single
992 write() function, the timeout was being re-applied to each part-write.
993 This seems wrong - if the receiver was accepting one byte at a time it
994 would take for ever. The timeout is now adjusted when this happens. It
995 doesn't have to be particularly precise.
997 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
998 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1000 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1001 without a domain if the check to <postmaster@domain> fails.
1003 SC/01 Eximstats: added -xls and the ability to specify output files
1004 (patch written by Frank Heydlauf).
1006 SC/02 Eximstats: use FileHandles for outputing results.
1008 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1010 SC/04 Eximstats: fixed display of large numbers with -nvr option
1012 SC/05 Eximstats: fixed merging of reports with empty tables.
1014 SC/06 Eximstats: added the -include_original_destination flag
1016 SC/07 Eximstats: removed tabs and trailing whitespace.
1018 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1020 TK/06 MBOX spool code: Add real "From " MBOX separator line
1021 so the .eml file is really in mbox format (even though
1022 most programs do not really care). Patch from Alex Miller.
1024 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1025 The latter is generated from $received_to and is only set if the
1026 message has one envelope recipient. SA can use these headers,
1027 obviously out-of-the-box. Patch from Alex Miller.
1029 PH/08 The ${def test on a variable was returning false if the variable's
1030 value was "0", contrary to what the specification has always said!
1031 The result should be true unless the variable is empty.
1033 PH/09 The syntax error of a character other than { following "${if
1034 def:variable_name" (after optional whitespace) was not being diagnosed.
1035 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1036 accidental colon was present, for example, could give incorrect results.
1038 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1039 result is used (not including appendfile, where other changes are about
1042 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1043 This involved changing a lot of size variables from int to off_t. It
1044 should work with maildirs and everything.
1046 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1047 spamd dying while we are connected to it.
1049 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1052 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1053 function for BDB 4.3.
1055 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1057 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1058 the sources for PCRE have been re-arranged and more files are now
1061 PH/15 The code I had for printing potentially long long variables in PH/11
1062 above was not the best (it lost precision). The length of off_t variables
1063 is now inspected at build time, and an appropriate printing format (%ld
1064 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1065 to be "long long int" or "long int". This is needed for the internal
1066 formatting function string_vformat().
1068 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1069 the configuration file to be ":syslog", then the script "guesses" where
1070 the logs files are, rather than using the compiled in default. In our
1071 case the guess is not the same as the compiled default, so the script
1072 suddenly stopped working when I started to use syslog. The patch checks
1073 to see if log_file_path is "". If so, it attempts to read it from exim
1074 with no configuration file to get the compiled in version, before it
1075 falls back to the previous guessing code."
1077 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1078 implementing BATV in an Exim configuration. See NewStuff for the gory
1081 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1082 Makefile that are specific to HP-UX.
1084 PH/18 If the "use_postmaster" option was set for a recipient callout together
1085 with the "random" option, the postmaster address was used as the MAIL
1086 FROM address for the random test, but not for the subsequent recipient
1087 test. It is now used for both.
1089 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1090 patch removes a few documentation additions to RFC 3028, because the
1091 latest draft now contains them. It adds the new en;ascii-case comparator
1092 and a new error check for 8bit text in MIME parts. Comparator and
1093 require names are now matched exactly. I enabled the subaddress
1094 extension, but it is not well tested yet (read: it works for me)."
1096 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1097 rework some of the code of TK/09 above to avoid the hardwired use of
1098 "%lld" and "long long". Replaced the call to snprintf() with a call to
1101 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1102 records point to non-existent hosts", "retry timeout exceeded", and
1103 "retry time not reached for any host after a long failure period".
1105 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1106 experimental DomainKeys support:
1108 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1109 (2) On an error such as an illegally used "control", the wrong name for
1110 the control was given.
1112 These problems did NOT occur unless DomainKeys support was compiled.
1114 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1116 PH/24 Added ${if match_ip condition.
1118 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1119 databases so that it will be absolutely obvious if a crash occurs in the
1120 DB library. This is a regular occurrence (often caused by mis-matched
1123 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1124 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1125 picked up on a user's system that detects such things. There doesn't seem
1126 to be a gcc warning option for this - only an attribute that has to be
1127 put on the function's prototype. It seems that in Fedora Core 4 they have
1128 set this on a number of new functions. No doubt there will be more in due
1131 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1132 specify a transport. However, if an address that was verified by such a
1133 router was the subject of a callout, Exim crashed because it tried to
1134 read the rcpt_include_affixes from the non-existent transport. Now it
1135 just assumes that the setting of that option is false. This bug was
1136 introduced by 4.51/PH/31.
1138 PH/28 Changed -d+all to exclude +memory, because that information is very
1139 rarely of interest, but it makes the output a lot bigger. People tend to
1140 do -d+all out of habit.
1142 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1143 code in os-type was giving problems when libc.so lives in lib64, like on
1146 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1147 aren't the modern standard, and it seems that some systems' include files
1148 don't always have them. Exim was already checking for some of the newer
1149 ones like T_AAAA, and defining it itself. I've added checks for all the
1150 record types that Exim uses.
1152 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1153 not automatically generating a new one, as it is supposed to. This
1154 prevented TLS from working. If the file did exist, but contained invalid
1155 data, a new version was generated, as expected. It was only the case of a
1156 non-existent file that was broken.
1158 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1159 with a change in libdomainkeys > 0.64.
1161 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1162 from DNS. If the selector record carries the flag, it now has
1163 precedence over the domain-wide flag.
1165 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1167 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1168 the use of a port name that isn't defined in /etc/services) occurred, the
1169 message was deferred as in a normal delivery, and thus remained on the
1170 spool, instead of being failed because of the mua_wrapper setting. This
1171 is now fixed, and I tidied up some of the mua_wrapper messages at the
1174 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1175 the messages in a hash of arrays rather than using individual hashes.
1176 This is a bit cleaner and results in dramatic memory savings, albeit
1177 at a slight CPU cost.
1179 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1180 as requested by Marc Sherman.
1182 SC/10 Eximstats: added histograms for user specified patterns as requested
1185 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1187 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1188 fopen() in the content-scanning modules that did not already have it.
1194 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1195 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1197 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1199 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1201 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1202 to test Sieve filters that use "vacation".
1204 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1205 that changes the way the GnuTLS parameters are stored in the cache file.
1206 The new format can be generated externally. For backward compatibility,
1207 if the data in the cache doesn't make sense, Exim assumes it has read an
1208 old-format file, and it generates new data and writes a new file. This
1209 means that you can't go back to an older release without removing the
1212 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1213 work if there are any delivery delays because "one_time" forces the
1214 parent to be marked "delivered", so its unseen clone is never tried
1215 again. For this reason, Exim now forbids the simultaneous setting of
1218 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1219 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1220 entry if you want to know the details. The fix, however, neglected to
1221 consider the case when local delivery batching is involved. The test for
1222 "previously delivered" was not happening when checking to see if an
1223 address could be batched with a previous (undelivered) one; under
1224 certain circumstances this could lead to multiple deliveries to the same
1227 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1228 in its include files, and this causes problems building Exim.
1230 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1231 header_syntax) but Exim was just ignoring anything given after a slash.
1232 In particular, this caused confusion with an attempt to use "verify =
1233 reverse_host_lookup/defer_ok". An error is now given when options are
1234 supplied for verify items that do not have them. (Maybe reverse_host_
1235 lookup should have a defer_ok option, but that's a different point.)
1237 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1238 defined by RFC 821) to 2048, because there were problems with some AUTH
1239 commands, and RFC 1869 says the size should be increased for extended
1240 SMTP commands that take arguments.
1242 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1245 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1246 "unknown" error; now it says that the functionality isn't in the binary.
1248 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1249 an address' error message when a string expansion fails (syntax or
1250 whatever). Otherwise the password may appear in the log. Following change
1251 PH/42 below, there is no longer a chance of it appearing in a bounce
1254 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1256 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1257 of its addresses was ever tried. (Bugzilla bug #2.)
1259 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1260 the result incorrectly in the debug output. (It correctly added a newline
1261 to what was transported.)
1263 TF/01 Added $received_time.
1265 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1266 commented out examples of how to interface to a virus scanner and to
1267 SpamAssassin. Also added commented examples of av_scanner and
1268 spamd_address settings.
1270 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1271 and controls are allowed in which ACLs. There were a couple of minor
1272 errors. Some of the entries in the conditions table (which is a table of
1273 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1274 negation of where the condition IS allowed.
1276 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1278 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1279 header file does not have a version number, so I've had to invent a new
1280 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1281 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1282 radiusclient), but it was contributed by a Radius user.
1284 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1285 files or directories, for OpenSSL.
1287 PH/19 When an Exim process that is running as root has to create an Exim log
1288 file, it does so in a subprocess that runs as exim:exim so as to get the
1289 ownership right at creation (otherwise, other Exim processes might see
1290 the file with the wrong ownership). There was no test for failure of this
1291 fork() call, which would lead to the process getting stuck as it waited
1292 for a non-existent subprocess. Forks do occasionally fail when resources
1293 run out. I reviewed all the other calls to fork(); they all seem to check
1296 PH/20 When checking for unexpected SMTP input at connect time (before writing
1297 the banner), Exim was not dealing correctly with a non-positive return
1298 from the read() function. If the client had disconnected by this time,
1299 the result was a log entry for a synchronization error with an empty
1300 string after "input=" when read() returned zero. If read() returned -1
1301 (an event I could not check), uninitialized data bytes were printed.
1302 There were reports of junk text (parts of files, etc) appearing after
1305 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1307 PH/22 Added support for macro redefinition, and (re)definition in between
1308 driver and ACL definitions.
1310 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1311 forgetting to use the resulting value; it was using the unexpanded value.
1313 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1314 hadn't been configured. The fix is from Juergen Kreileder, who
1315 understands it better than I do:
1317 "Here's what I see happening with three configured cyrus_sasl
1318 authenticators configured (plain, login, cram-md5):
1320 On startup auth_cyrus_sasl_init() gets called for each of these.
1321 This means three calls to sasl_listmech() without a specified mech_list.
1322 => SASL tests which mechs of all available mechs actually work
1323 => three warnings about OTP not working
1324 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1326 With the patch, sasl_listmech() also gets called three times. But now
1327 SASL's mech_list option is set to the server_mech specified in the the
1328 authenticator. Or in other words, the answer from sasl_listmech()
1329 gets limited to just the mech you're testing for (which is different
1331 => the return list contains just 'plain' or 'login', 'cram-md5' or
1332 nothing depending on the value of ob->server_mech.
1334 I've just tested the patch: Authentication still works fine,
1335 unavailable mechs specified in the exim configuration are still
1336 caught, and the auth.log warnings about OTP are gone."
1338 PH/25 When debugging is enabled, the contents of the command line are added
1339 to the debugging output, even when log_selector=+arguments is not
1342 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1343 answer is "GNU", and only if the return is "GNU/something" is the answer
1346 PH/27 $acl_verify_message is now set immediately after the failure of a
1347 verification in an ACL, and so is available in subsequent modifiers. In
1348 particular, the message can be preserved by coding like this:
1350 warn !verify = sender
1351 set acl_m0 = $acl_verify_message
1353 Previously, $acl_verify_message was set only while expanding "message"
1354 and "log_message" when a very denied access.
1356 PH/28 Modified OS/os.c-Linux with
1358 -#ifndef OS_LOAD_AVERAGE
1359 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1361 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1362 nomenclature these days.)
1364 PH/29 Installed patch from the Sieve maintainer that adds the options
1365 sieve_useraddress and sieve_subaddress to the redirect router.
1367 PH/30 In these circumstances:
1368 . Two addresses routed to the same list of hosts;
1369 . First host does not offer TLS;
1370 . First host accepts first address;
1371 . First host gives temporary error to second address;
1372 . Second host offers TLS and a TLS session is established;
1373 . Second host accepts second address.
1374 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1375 and peerdn, if requested) that were in fact used only for the second
1378 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1379 attention to any local part prefix or suffix that was matched by the
1380 router that accepted the address. It now behaves in the same way as it
1381 does for delivery: the affixes are removed from the local part unless
1382 rcpt_include_affixes is set on the transport.
1384 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1385 timeout during the DATA phase of an incoming message.
1387 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1388 applied a patch sanctioned by the Sieve maintainer.
1390 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1391 the uid or gid is negative. A case of a negative gid caused this to be
1392 noticed. The fix allows for either to be negative.
1394 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1395 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1398 PH/36 The change PH/12 above was broken. Fixed it.
1400 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1401 the grounds that routing the same address twice would always produce the
1402 same answer. This might have been true once, but it is certainly no
1403 longer true now. Routing a child address may depend on the previous
1404 routing that produced that child. Some complicated redirection strategies
1405 went wrong when messages had multiple recipients, and made Exim's
1406 behaviour dependent on the order in which the addresses were given.
1408 I have moved the duplicate checking until after the routing is complete.
1409 Exim scans the addresses that are assigned to local and remote
1410 transports, and removes any duplicates. This means that more work will be
1411 done, as duplicates will always all be routed, but duplicates are
1412 presumably rare, so I don't expect this is of any significance.
1414 For deliveries to pipes, files, and autoreplies, the duplicate checking
1415 still happens during the routing process, since they are not going to be
1418 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1419 It corrects a timeout issue with spamd. This is Ian's comment: "The
1420 background is that sometimes spamd either never reads data from a
1421 connection it has accepted, or it never writes response data. The exiscan
1422 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1423 blindly assumes that writes won't block so it may never time out."
1425 PH/39 Allow G after quota size as well as K and M.
1427 PH/40 The value set for $authenticated_id in an authenticator may not contain
1428 binary zeroes or newlines because the value is written to log lines and
1429 to spool files. There was no check on this. Now the value is run through
1430 the string_printing() function so that such characters are converted to
1431 printable escape sequences.
1433 PH/41 $message_linecount is a new variable that contains the total number of
1434 lines in the message. Compare $body_linecount, which is the count for the
1437 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1438 bounce and delay warning messages, except in certain special cases, which
1441 (a) An SMTP error message from a remote host;
1442 (b) A message specified in a :fail: redirection;
1443 (c) A message specified in a "fail" command in a system filter;
1444 (d) A message specified in a FAIL return from the queryprogram router;
1445 (e) A message specified by the cannot_route_message router option.
1447 In these cases only, Exim does include the error details in bounce and
1448 warning messages. There are also a few cases where bland messages such
1449 as "unrouteable address" or "local delivery error" are given.
1451 PH/43 $value is now also set for the "else" part of a ${run expansion.
1453 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1454 being worked on, but at least Exim now implements the latest version to
1457 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1458 process to complete was treated as a delivery failure, a timeout while
1459 writing the message to the pipe was logged, but erroneously treated as a
1460 successful delivery. Such timeouts include transport filter timeouts. For
1461 consistency with the overall process timeout, these timeouts are now
1462 treated as errors, giving rise to delivery failures by default. However,
1463 there is now a new Boolean option for the pipe transport called
1464 timeout_defer, which, if set TRUE, converts the failures into defers for
1465 both kinds of timeout. A transport filter timeout is now identified in
1468 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1469 systems where "make" and "gmake" are different, calling "gmake" at top
1470 level broke things. I've arranged for the value of $(MAKE) to be passed
1471 from the Makefile to this script so that it can call the same version of
1475 A note about Exim versions 4.44 and 4.50
1476 ----------------------------------------
1478 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1479 changes of various kinds. As a consequence, a big documentation update was
1480 needed. This delayed the release for rather longer than seemed good, especially
1481 in the light of a couple of (minor) security issues. Therefore, the changes
1482 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1483 release. So 4.44 and 4.50 are in effect two different branches that both start
1486 I have left the 4.50 change log unchanged; it contains all the changes since
1487 4.43. The change log for 4.44 is below; many of its items are identical to
1488 those for 4.50. This seems to be the most sensible way to preserve the
1489 historical information.
1495 1. Minor wording change to the doc/README.SIEVE file.
1497 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1498 computation of the current number of files was incorrect.
1500 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1501 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1504 4. Give more explanation in the error message when the command for a transport
1505 filter fails to execute.
1507 5. There are several places where Exim runs a non-Exim command in a
1508 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1509 was being done only for the command run by the queryprogram router. It is
1510 now done for all such subprocesses. The other cases are: ${run, transport
1511 filters, and the commands run by the lmtp and pipe transports.
1513 6. Added CONFIGURE_GROUP build-time option.
1515 7. Some older OS have a limit of 256 on the maximum number of file
1516 descriptors. Exim was using setrlimit() to set 1000 as a large value
1517 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1518 systems. I've change it so that if it can't get 1000, it tries for 256.
1520 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1521 was an oversight, and furthermore, ever since the addition of extra
1522 controls (e.g. 4.43/32), the checks on when to allow different forms of
1523 "control" were broken. There should now be diagnostics for all cases when a
1524 control that does not make sense is encountered.
1526 9. Added the /retain_sender option to "control=submission".
1528 10. $recipients is now available in the predata ACL (oversight).
1530 11. Tidy the search cache before the fork to do a delivery from a message
1531 received from the command line. Otherwise the child will trigger a lookup
1532 failure and thereby defer the delivery if it tries to use (for example) a
1533 cached ldap connection that the parent has called unbind on.
1535 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1536 of $address_data from the recipient verification was clobbered by the
1537 sender verification.
1539 13. The value of address_data from a sender verification is now available in
1540 $sender_address_data in subsequent conditions in the ACL statement.
1542 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1544 15. Added a new option "connect=<time>" to callout options, to set a different
1547 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1548 was its contents. (It was OK if the option was not defined at all.)
1550 17. A "Completed" log line is now written for messages that are removed from
1551 the spool by the -Mrm option.
1553 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1554 information about exactly what failed.
1556 19. Added -dd to debug only the daemon process.
1558 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1559 handles timeouts, both on the server side and network timeouts. Renamed the
1560 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1562 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1563 It is now set to "smtps".
1565 22. $host_address is now set to the target address during the checking of
1566 ignore_target_hosts.
1568 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1569 being passed; this would have caused $sender_host_name to have been used if
1570 matching the list had actually called for a host name (not very likely,
1571 since this list is usually IP addresses). A host name is now passed as
1574 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1575 code that specifies a non-restarting handler (typically sigaction() in
1576 modern systems) in an attempt to fix a rare and obscure crash bug.
1578 25. Narrowed the window for a race in the daemon that could cause it to ignore
1579 SIGCHLD signals. This is not a major problem, because they are used only to
1580 wake it up if nothing else does.
1582 26. A malformed maildirsize file could cause Exim to calculate negative values
1583 for the mailbox size or file count. Odd effects could occur as a result.
1584 The maildirsize information is now recalculated if the size or filecount
1587 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1588 support for a long time. Removed HAVE_SYS_VFS_H.
1590 28. Installed the latest version of exipick from John Jetmore.
1592 29. In an address list, if the pattern was not a regular expression, an empty
1593 subject address (from a bounce message) matched only if the pattern was an
1594 empty string. Non-empty patterns were not even tested. This was the wrong
1595 because it is perfectly reasonable to use an empty address as part of a
1596 database query. An empty address is now tested by patterns that are
1597 lookups. However, all the other forms of pattern expect the subject to
1598 contain a local part and a domain, and therefore, for them, an empty
1599 address still always fails if the pattern is not itself empty.
1601 30. Exim went into a mad DNS loop when attempting to do a callout where the
1602 host was specified on an smtp transport, and looking it up yielded more
1603 than one IP address.
1605 31. Re-factored the code for checking spool and log partition space into a
1606 function that finds that data and another that does the check. The former
1607 is then used to implement four new variables: $spool_space, $log_space,
1608 $spool_inodes, and $log_inodes.
1610 32. The RFC2047 encoding function was originally intended for short strings
1611 such as real names; it was not keeping to the 75-character limit for
1612 encoded words that the RFC imposes. It now respects the limit, and
1613 generates multiple encoded words if necessary. To be on the safe side, I
1614 have increased the buffer size for the ${rfc2047: expansion operator from
1617 33. It is now permitted to omit both strings after an "if" condition; if the
1618 condition is true, the result is "true". As before, when the second string
1619 is omitted, a false condition yields an empty string. This makes it less
1620 cumbersome to write custom ACL and router conditions.
1622 34. Failure to deliver a bounce message always caused it to be frozen, even if
1623 there was an errors_to setting on the router. The errors_to setting is now
1626 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1627 canonical form (fully expanded) before being placed in
1628 $sender_host_address.
1630 36. The table in the code that translates DNS record types into text (T_A to
1631 "A" for instance) was missing entries for NS and CNAME. It is just possible
1632 that this could have caused confusion if both these types were looked up
1633 for the same domain, because the text type is used as part of Exim's
1634 per-process caching. But the chance of anyone hitting this buglet seems
1637 37. The dnsdb lookup has been extended in a number of ways.
1639 (1) There is a new type, "zns", which walks up the domain tree until it
1640 finds some nameserver records. It should be used with care.
1642 (2) There is a new type, "mxh", which is like "mx" except that it returns
1643 just the host names, not the priorities.
1645 (3) It is now possible to give a list of domains (or IP addresses) to be
1646 looked up. The behaviour when one of the lookups defers can be
1647 controlled by a keyword.
1649 (4) It is now possible to specify the separator character for use when
1650 multiple records are returned.
1652 38. The dnslists ACL condition has been extended: it is now possible to supply
1653 a list of IP addresses and/or domains to be looked up in a particular DNS
1656 39. Added log_selector=+queue_time_overall.
1658 40. When running the queue in the test harness, wait just a tad after forking a
1659 delivery process, to get repeatability of debugging output.
1661 41. Include certificate and key file names in error message when GnuTLS fails
1662 to set them up, because the GnuTLS error message doesn't include the name
1663 of the failing file when there is a problem reading it.
1665 42. Allow both -bf and -bF in the same test run.
1667 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1669 44. The "Exiscan patch" is now merged into the mainline Exim source.
1671 45. Sometimes the final signoff response after QUIT could fail to get
1672 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1673 < 0 before doing a fflush(). This bug looks as though it goes back to the
1674 introduction of TLS in release 3.20, but "sometimes" must have been rare
1675 because the tests only now provoked it.
1677 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1678 (this can affect the format of dates).
1680 47. exim_tidydb, when checking for the continued existence of a message for
1681 which it has found a message-specific retry record, was not finding
1682 messages that were in split spool directories. Consequently, it was
1683 deleting retry records that should have stayed in existence.
1685 48. Steve fixed some bugs in eximstats.
1687 49. The SPA authentication driver was not abandoning authentication and moving
1688 on to the next authenticator when an expansion was forced to fail,
1689 contradicting the general specification for all authenticators. Instead it
1690 was generating a temporary error. It now behaves as specified.
1692 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1693 (the order specifies the preference for clients). The order is now AES256,
1694 AES128, 3DES, ARCFOUR128.
1696 51. Small patch to Sieve code - explicitly set From: when generating an
1699 52. Exim crashed if a remote delivery caused a very long error message to be
1700 recorded - for instance if somebody sent an entire SpamAssassin report back
1701 as a large number of 550 error lines. This bug was coincidentally fixed by
1702 increasing the size of one of Exim's internal buffers (big_buffer) that
1703 happened as part of the Exiscan merge. However, to be on the safe side, I
1704 have made the code more robust (and fixed the comments that describe what
1707 53. Now that there can be additional text after "Completed" in log lines (if
1708 the queue_time_overall log selector is set), a one-byte patch to exigrep
1709 was needed to allow it to recognize "Completed" as not the last thing in
1712 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1713 patch that reportedly fixes this has been added. I am not expert enough to
1714 create a test for it. This is what the patch creator wrote:
1716 "I found a little strange behaviour of ldap code when working with
1717 Windows 2003 AD Domain, where users was placed in more than one
1718 Organization Units. When I tried to give exim partial DN, the exit code
1719 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1720 But simultaneously result of request was absolutely normal ldap result,
1721 so I produce this patch..."
1723 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1724 REFERENCE, so I have modified the code to exclude the patch when that macro
1727 55. Some experimental protocols are using DNS PTR records for new purposes. The
1728 keys for these records are domain names, not reversed IP addresses. The
1729 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1730 leaves it alone. Component reversal etc. now happens only for IP addresses.
1733 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1735 57. Double the size of the debug message buffer (to 2048) so that more of very
1736 long debug lines gets shown.
1738 58. The exicyclog utility now does better if the number of log files to keep
1739 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1741 59. Two changes related to the smtp_active_hostname option:
1743 (1) $smtp_active_hostname is now available as a variable.
1744 (2) The default for smtp_banner uses $smtp_active_hostname instead
1745 of $primary_hostname.
1747 60. The host_aton() function is supposed to be passed a string that is known
1748 to be a valid IP address. However, in the case of IPv6 addresses, it was
1749 not checking this. This is a hostage to fortune. Exim now panics and dies
1750 if the condition is not met. A case was found where this could be provoked
1751 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1752 components; fortuitously, this particular loophole had already been fixed
1753 by change 4.50/55 above.
1755 If there are any other similar loopholes, the new check in host_aton()
1756 itself should stop them being exploited. The report I received stated that
1757 data on the command line could provoke the exploit when Exim was running as
1758 exim, but did not say which command line option was involved. All I could
1759 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1760 running as the user.
1763 61. There was a buffer overflow vulnerability in the SPA authentication code
1764 (which came originally from the Samba project). I have added a test to the
1765 spa_base64_to_bits() function which I hope fixes it.
1768 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1769 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1771 63. The daemon start-up calls getloadavg() while still root for those OS that
1772 need the first call to be done as root, but it missed one case: when
1773 deliver_queue_load_max is set with deliver_drop_privilege. This is
1774 necessary for the benefit of the queue runner, because there is no re-exec
1775 when deliver_drop_privilege is set.
1777 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1778 This has been fixed.
1780 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1781 was in use, was not putting the data itself into the right store pool;
1782 consequently, it could be overwritten for a subsequent message in the same
1783 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1786 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1788 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1789 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1790 the function were treating the return as a boolean value, which happened to
1791 work because 0=false and not-0=true, but is not correct code.
1793 68. The host_aton() function was not handling scoped IPv6 addresses (those
1794 with, for example, "%eth0" on the end) correctly.
1796 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1797 negated items (that is, ~something) in unsigned ints. Some compilers
1798 apparently mutter when there is no cast.
1800 70. If an address verification called from an ACL failed, and did not produce a
1801 user-specific message (i.e. there was only a "system" message), nothing was
1802 put in $acl_verify_message. In this situation, it now puts the system
1805 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1806 session; change 4.31/43 added the unwanted input to the log line - except
1807 that it did not do this in the start of session case. It now does.
1809 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1810 This is wrong and can cause the other end to generate a synchronization
1811 error if it is another Exim or anything else that does the synchronization
1812 check. A QUIT command is no longer sent after a timeout.
1814 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1815 during host lookups.
1817 74. The defer_ok option of callout verification was not working if it was used
1818 when verifying addresses in header lines, that is, for this case:
1820 verify = header_sender/callout=defer_ok
1822 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1823 those file descriptors could be used for SMTP connections. If anything
1824 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1825 could be sent to the SMTP client, causing chaos. The daemon now opens
1826 stdin, stdout, and stderr to /dev/null when it puts itself into the
1829 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1830 log by default. The user can override this with suitable Perl magic.
1832 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1833 the log message when discard triggers, was not working for the DATA ACL or
1834 for the non-SMTP ACL.
1836 78. Error message wording change in sieve.c.
1838 79. If smtp_accept_max_per_host was set, the number of connections could be
1839 restricted to fewer than expected, because the daemon was trying to set up
1840 a new connection before checking whether the processes handling previous
1841 connections had finished. The check for completed processes is now done
1842 earlier. On busy systems, this bug wouldn't be noticed because something
1843 else would have woken the daemon, and it would have reaped the completed
1846 80. If a message was submitted locally by a user whose login name contained one
1847 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1848 It caused a spool format error. I have fixed the spool reading code. A
1849 related problem was that the "from" clause in the Received: line became
1850 illegal because of the space(s). It is now covered by ${quote_local_part.
1852 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1855 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1857 83. Patch from Sieve maintainer to fix unterminated string problem in
1858 "vacation" handling.
1860 84. Some minor changes to the Linux configuration files to help with other
1861 OS variants using glibc.
1863 85. One more patch for Sieve to update vacation handling to latest spec.
1866 ----------------------------------------------------
1867 See the note above about the 4.44 and 4.50 releases.
1868 ----------------------------------------------------
1874 1. Change 4.43/35 introduced a bug that caused file counts to be
1875 incorrectly computed when quota_filecount was set in an appendfile
1878 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1879 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1882 3. Give more explanation in the error message when the command for a transport
1883 filter fails to execute.
1885 4. There are several places where Exim runs a non-Exim command in a
1886 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1887 was being done only for the command run by the queryprogram router. It is
1888 now done for all such subprocesses. The other cases are: ${run, transport
1889 filters, and the commands run by the lmtp and pipe transports.
1891 5. Some older OS have a limit of 256 on the maximum number of file
1892 descriptors. Exim was using setrlimit() to set 1000 as a large value
1893 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1894 systems. I've change it so that if it can't get 1000, it tries for 256.
1896 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1897 was an oversight, and furthermore, ever since the addition of extra
1898 controls (e.g. 4.43/32), the checks on when to allow different forms of
1899 "control" were broken. There should now be diagnostics for all cases when a
1900 control that does not make sense is encountered.
1902 7. $recipients is now available in the predata ACL (oversight).
1904 8. Tidy the search cache before the fork to do a delivery from a message
1905 received from the command line. Otherwise the child will trigger a lookup
1906 failure and thereby defer the delivery if it tries to use (for example) a
1907 cached ldap connection that the parent has called unbind on.
1909 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1910 of $address_data from the recipient verification was clobbered by the
1911 sender verification.
1913 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1914 was its contents. (It was OK if the option was not defined at all.)
1916 11. A "Completed" log line is now written for messages that are removed from
1917 the spool by the -Mrm option.
1919 12. $host_address is now set to the target address during the checking of
1920 ignore_target_hosts.
1922 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1923 being passed; this would have caused $sender_host_name to have been used if
1924 matching the list had actually called for a host name (not very likely,
1925 since this list is usually IP addresses). A host name is now passed as
1928 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1929 code that specifies a non-restarting handler (typically sigaction() in
1930 modern systems) in an attempt to fix a rare and obscure crash bug.
1932 15. Narrowed the window for a race in the daemon that could cause it to ignore
1933 SIGCHLD signals. This is not a major problem, because they are used only to
1934 wake it up if nothing else does.
1936 16. A malformed maildirsize file could cause Exim to calculate negative values
1937 for the mailbox size or file count. Odd effects could occur as a result.
1938 The maildirsize information is now recalculated if the size or filecount
1941 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1942 support for a long time. Removed HAVE_SYS_VFS_H.
1944 18. Updated exipick to current release from John Jetmore.
1946 19. Allow an empty sender to be matched against a lookup in an address list.
1947 Previously the only cases considered were a regular expression, or an
1950 20. Exim went into a mad DNS lookup loop when doing a callout where the
1951 host was specified on the transport, if the DNS lookup yielded more than
1954 21. The RFC2047 encoding function was originally intended for short strings
1955 such as real names; it was not keeping to the 75-character limit for
1956 encoded words that the RFC imposes. It now respects the limit, and
1957 generates multiple encoded words if necessary. To be on the safe side, I
1958 have increased the buffer size for the ${rfc2047: expansion operator from
1961 22. Failure to deliver a bounce message always caused it to be frozen, even if
1962 there was an errors_to setting on the router. The errors_to setting is now
1965 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1966 canonical form (fully expanded) before being placed in
1967 $sender_host_address.
1969 24. Updated eximstats to version 1.33
1971 25. Include certificate and key file names in error message when GnuTLS fails
1972 to set them up, because the GnuTLS error message doesn't include the name
1973 of the failing file when there is a problem reading it.
1975 26. Expand error message when OpenSSL has problems setting up cert/key files.
1978 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1979 (this can affect the format of dates).
1981 28. exim_tidydb, when checking for the continued existence of a message for
1982 which it has found a message-specific retry record, was not finding
1983 messages that were in split spool directories. Consequently, it was
1984 deleting retry records that should have stayed in existence.
1986 29. eximstats updated to version 1.35
1987 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1988 1.35 - bugfix such that pie charts by volume are generated correctly
1990 30. The SPA authentication driver was not abandoning authentication and moving
1991 on to the next authenticator when an expansion was forced to fail,
1992 contradicting the general specification for all authenticators. Instead it
1993 was generating a temporary error. It now behaves as specified.
1995 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1996 (the order specifies the preference for clients). The order is now AES256,
1997 AES128, 3DES, ARCFOUR128.
1999 31. Small patch to Sieve code - explicitly set From: when generating an
2002 32. Exim crashed if a remote delivery caused a very long error message to be
2003 recorded - for instance if somebody sent an entire SpamAssassin report back
2004 as a large number of 550 error lines. This bug was coincidentally fixed by
2005 increasing the size of one of Exim's internal buffers (big_buffer) that
2006 happened as part of the Exiscan merge. However, to be on the safe side, I
2007 have made the code more robust (and fixed the comments that describe what
2010 33. Some experimental protocols are using DNS PTR records for new purposes. The
2011 keys for these records are domain names, not reversed IP addresses. The
2012 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2013 leaves it alone. Component reversal etc. now happens only for IP addresses.
2016 34. The host_aton() function is supposed to be passed a string that is known
2017 to be a valid IP address. However, in the case of IPv6 addresses, it was
2018 not checking this. This is a hostage to fortune. Exim now panics and dies
2019 if the condition is not met. A case was found where this could be provoked
2020 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2021 components; fortuitously, this particular loophole had already been fixed
2022 by change 4.50/55 or 4.44/33 above.
2024 If there are any other similar loopholes, the new check in host_aton()
2025 itself should stop them being exploited. The report I received stated that
2026 data on the command line could provoke the exploit when Exim was running as
2027 exim, but did not say which command line option was involved. All I could
2028 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2029 running as the user.
2032 35. There was a buffer overflow vulnerability in the SPA authentication code
2033 (which came originally from the Samba project). I have added a test to the
2034 spa_base64_to_bits() function which I hope fixes it.
2037 36. The daemon start-up calls getloadavg() while still root for those OS that
2038 need the first call to be done as root, but it missed one case: when
2039 deliver_queue_load_max is set with deliver_drop_privilege. This is
2040 necessary for the benefit of the queue runner, because there is no re-exec
2041 when deliver_drop_privilege is set.
2043 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2044 was in use, was not putting the data itself into the right store pool;
2045 consequently, it could be overwritten for a subsequent message in the same
2046 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2049 38. Sometimes the final signoff response after QUIT could fail to get
2050 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2051 < 0 before doing a fflush(). This bug looks as though it goes back to the
2052 introduction of TLS in release 3.20, but "sometimes" must have been rare
2053 because the tests only now provoked it.
2059 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2060 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2061 Now it returns an integer. A number of calls were still expecting a T/F
2062 return. Fortuitously, in all cases, the tests worked in OK situations,
2063 which is the norm. However, things would have gone wrong on any write
2064 failures on the smtp file descriptor. This function is used when sending
2065 messages over SMTP and also when doing verify callouts.
2067 2. When Exim is called to do synchronous delivery of a locally submitted
2068 message (the -odf or -odi options), it no longer closes stderr before doing
2071 3. Implemented the mua_wrapper option.
2073 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2075 5. Implemented the functions header_remove(), header_testname(),
2076 header_add_at_position(), and receive_remove_recipient(), and exported them
2079 6. If an ACL "warn" statement specified the addition of headers, Exim already
2080 inserted X-ACL-Warn: at the start if there was no header name. However, it
2081 was not making this test for the second and subsequent header lines if
2082 there were newlines in the string. This meant that an invalid header could
2083 be inserted if Exim was badly configured.
2085 7. Allow an ACL "warn" statement to add header lines at the start or after all
2086 the Received: headers, as well as at the end.
2088 8. Added the rcpt_4xx retry error code.
2090 9. Added postmaster_mailfrom=xxx to callout verification option.
2092 10. Added mailfrom=xxxx to the callout verification option, for verify=
2095 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2096 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2098 12. Inserted some casts to stop certain compilers warning when using pointer
2099 differences as field lengths or precisions in printf-type calls (mostly
2100 affecting debugging statements).
2102 13. Added optional readline() support for -be (dynamically loaded).
2104 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2105 same clock tick as a message's arrival, so that its received time was the
2106 same as the "first fail" time on the retry record, and that message
2107 remained on the queue past the ultimate address timeout, every queue runner
2108 would try a delivery (because it was past the ultimate address timeout) but
2109 after another failure, the ultimate address timeout, which should have then
2110 bounced the address, did not kick in. This was a "< instead of <=" error;
2111 in most cases the first failure would have been in the next clock tick
2112 after the received time, and all would be well.
2114 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2115 being recognized when the domain list was tested by the match_domain
2116 condition in an expansion string.
2118 16. Added the ${str2b64: operator.
2120 17. Exim was always calling setrlimit() to set a large limit for the number of
2121 processes, without checking whether the existing limit was already
2122 adequate. (It did check for the limit on file descriptors.) Furthermore,
2123 errors from getrlimit() and setrlimit() were being ignored. Now they are
2124 logged to the main and panic logs, but Exim does carry on, to try to do its
2125 job under whatever limits there are.
2127 18. Imported PCRE 5.0.
2129 19. Trivial typo in log message " temporarily refused connection" (the leading
2132 20. If the log selector return_path_on_delivery was set and an address was
2133 redirected to /dev/null, the delivery process crashed because it assumed
2134 that a return path would always be set for a "successful" delivery. In this
2135 case, the whole delivery is bypassed as an optimization, and therefore no
2138 21. Internal re-arrangement: the function for sending a challenge and reading
2139 a response while authentication was assuming a zero-terminated challenge
2140 string. It's now changed to take a pointer and a length, to allow for
2141 binary data in such strings.
2143 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2145 23. Exim was not respecting finduser_retries when seeking the login of the
2146 uid under which it was called; it was always trying 10 times. (The default
2147 setting of finduser_retries is zero.) Also, it was sleeping after the final
2148 failure, which is pointless.
2150 24. Implemented tls_on_connect_ports.
2152 25. Implemented acl_smtp_predata.
2154 26. If the domain in control=submission is set empty, Exim assumes that the
2155 authenticated id is a complete email address when it generates From: or
2156 Sender: header lines.
2158 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2159 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2160 chown and chgrp in /bin and hostname in /usr/bin.
2162 28. Exim was keeping the "process log" file open after each use, just as it
2163 does for the main log. This opens the possibility of it remaining open for
2164 long periods when the USR1 signal hits a daemon. Occasional processlog
2165 errors were reported, that could have been caused by this. Anyway, it seems
2166 much more sensible not to leave this file open at all, so that is what now
2169 29. The long-running daemon process does not normally write to the log once it
2170 has entered its main loop, and it closes the log before doing so. This is
2171 so that log files can straightforwardly be renamed and moved. However,
2172 there are a couple of unusual error situations where the daemon does write
2173 log entries, and I had neglected to close the log afterwards.
2175 30. The text of an SMTP error response that was received during a remote
2176 delivery was being truncated at 512 bytes. This is too short for some of
2177 the long messages that one sometimes sees. I've increased the limit to
2180 31. It is now possible to make retry rules that apply only when a message has a
2181 specific sender, in particular, an empty sender.
2183 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2184 it possible to be selective about when SMTP synchronization is enforced.
2186 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2188 32. Implemented hosts_connection_nolog.
2190 33. Added an ACL for QUIT.
2192 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2195 35. Added mailbox_size and mailbox_filecount to appendfile.
2197 36. Added control = no_multiline_responses to ACLs.
2199 37. There was a bug in the logic of the code that waits for the clock to tick
2200 in the case where the clock went backwards by a substantial amount such
2201 that the microsecond fraction of "now" was more than the microsecond
2202 fraction of "then" (but the whole seconds number was less).
2204 38. Added support for the libradius Radius client library this is found on
2205 FreeBSD (previously only the radiusclient library was supported).
2211 1. When certain lookups returned multiple values in the form name=value, the
2212 quoting of the values was not always being done properly. Specifically:
2213 (a) If the value started with a double quote, but contained no whitespace,
2215 (b) If the value contained whitespace other than a space character (i.e.
2216 tabs or newlines or carriage returns) it was not quoted.
2217 This fix has been applied to the mysql and pgsql lookups by writing a
2218 separate quoting function and calling it from the lookup code. The fix
2219 should probably also be applied to nisplus, ibase and oracle lookups, but
2220 since I cannot test any of those, I have not disturbed their existing code.
2222 2. A hit in the callout cache for a specific address caused a log line with no
2223 reason for rejecting RCPT. Now it says "Previous (cached) callout
2224 verification failure".
2226 3. There was an off-by-one bug in the queryprogram router. An over-long
2227 return line was truncated at 256 instead of 255 characters, thereby
2228 overflowing its buffer with the terminating zero. As well as fixing this, I
2229 have increased the buffer size to 1024 (and made a note to document this).
2231 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2232 when Exim is waiting for an SMTP response from a remote server, Exim
2233 restarts its select() call on the socket, thereby resetting its timeout.
2234 This is not a problem when such interrupts are rare. Somebody set up a cron
2235 job to run exiwhat every 2 minutes, which is less than the normal select()
2236 timeout (5 or 10 minutes). This meant that the select() timeout never
2237 kicked in because it was always reset. I have fixed this by comparing the
2238 time when an interrupt arrives with the time at the start of the first call
2239 to select(). If more time than the timeout has elapsed, the interrupt is
2240 treated as a timeout.
2242 5. Some internal re-factoring in preparation for the addition of Sieve
2243 extensions (by MH). In particular, the "personal" test is moved to a
2244 separate function, and given an option for scanning Cc: and Bcc: (which is
2245 not set for Exim filters).
2247 6. When Exim created an email address using the login of the caller as the
2248 local part (e.g. when creating a From: or Sender: header line), it was not
2249 quoting the local part when it contained special characters such as @.
2251 7. Installed new OpenBSD configuration files.
2253 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2254 try to make them clearer.
2256 9. Callout options, other than the timeout value, were being ignored when
2257 verifying sender addresses in header lines. For example, when using
2259 verify = header_sender/callout=no_cache
2261 the cache was (incorrectly) being used.
2263 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2264 only the Cygwin environment.
2266 11. When return_path_on_delivery was set as a log selector, if different remote
2267 addresses in the same message used different return paths and parallel
2268 remote delivery occurred, the wrong values would sometimes be logged.
2269 (Whenever a remote delivery process finished, the return path value from
2270 the most recently started remote delivery process was logged.)
2272 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2273 lines when AUTH and/or TLS are in use. This is the "received protocol"
2274 field. Exim used to use "asmtp" for authenticated SMTP, without any
2275 indication (in the protocol name) for TLS use. Now it follows the RFC and
2276 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2277 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2278 names appear in log lines as well as in Received: header lines.
2280 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2281 extensions, and comparison tests, and to fix some bugs.
2283 14. Changes to the "personal" filter test:
2285 (1) The test was buggy in that it was just doing the equivalent of
2286 "contains" tests on header lines. For example, if a user's address was
2287 anne@some.where, the "personal" test would incorrectly be true for
2289 To: susanne@some.where
2291 This test is now done by extracting each address from the header in turn,
2292 and checking the entire address. Other tests that are part of "personal"
2293 are now done using regular expressions (for example, to check local parts
2294 of addresses in From: header lines).
2296 (2) The list of non-personal local parts in From: addresses has been
2297 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2298 taken from the Sieve specification recommendations.
2300 (3) If the message contains any header line starting with "List-" it is
2301 treated as non-personal.
2303 (4) The test for "circular" in the Subject: header line has been removed
2304 because it now seems ill-conceived.
2306 15. Minor typos in src/EDITME comments corrected.
2308 16. Installed latest exipick from John Jetmore.
2310 17. If headers_add on a router specified a text string that was too long for
2311 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2312 of string_sprintf() is now avoided.
2314 18. $message_body_size was not set (it was always zero) when running the DATA
2315 ACL and the local_scan() function.
2317 19. For the "mail" command in an Exim filter, no default was being set for
2318 the once_repeat time, causing a random time value to be used if "once" was
2319 specified. (If the value happened to be <= 0, no repeat happened.) The
2320 default is now 0s, meaning "never repeat". The "vacation" command was OK
2321 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2322 (I found it when inspecting the code).
2324 20. There is now an overall timeout for performing a callout verification. It
2325 defaults to 4 times the callout timeout, which applies to individual SMTP
2326 commands during the callout. The overall timeout applies when there is more
2327 than one host that can be tried. The timeout is checked before trying the
2328 next host. This prevents very long delays if there are a large number of
2329 hosts and all are timing out (e.g. when the network connections are timing
2330 out). The value of the overall timeout can be changed by specifying an
2331 additional sub-option for "callout", called "maxwait". For example:
2333 verify = sender/callout=5s,maxwait=20s
2335 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2336 to the end before writing, but this should make it even safer).
2338 22. Exim was forgetting that it had advertised PIPELINING for the second and
2339 subsequent messages on an SMTP connection. It was also not resetting its
2340 memory on STARTTLS and an internal HELO.
2342 23. When Exim logs an SMTP synchronization error within a session, it now
2343 records whether PIPELINING has been advertised or not.
2345 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2346 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2347 rather than long int.
2349 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2351 26. Added the never_mail option to autoreply.
2357 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2358 crash if the getsockname() call failed; this can happen if a connection is
2359 closed very soon after it is established. The problem was simply in the
2360 order in which certain operations were done, causing Exim to try to write
2361 to the SMTP stream before it had set up the file descriptor. The bug has
2362 been fixed by making things happen in the correct order.
2368 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2369 before the connection was closed, thus losing the rejection response.
2371 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2372 some early Solaris releases, but causes trouble in current releases where
2373 socklen_t is defined.
2375 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2378 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2381 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2382 that releasing the top of it at the end releases what was used for sub-
2383 expansions (unless the block got too big). However, discard this block if
2384 the first thing is a variable or header, so that we can use its block when
2385 it is dynamic (useful for very large $message_headers, for example).
2387 6. Lookups now cache *every* query, not just the most recent. A new, separate
2388 store pool is used for this. It can be recovered when all lookup caches are
2389 flushed. Lookups now release memory at the end of their result strings.
2390 This has involved some general refactoring of the lookup sources.
2392 7. Some code has been added to the store_xxx() functions to reduce the amount
2393 of flapping under certain conditions.
2395 8. log_incoming_interface used to affect only the <= reception log lines. Now
2396 it causes the local interface and port to be added to several more SMTP log
2397 lines, for example "SMTP connection from", and rejection lines.
2399 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2401 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2403 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2404 could be overwritten at the end of the current message (or the start of a
2405 new message if it was set in a HELO ACL). The value is now preserved for
2406 the duration of the SMTP connection.
2408 12. If a transport had a headers_rewrite setting, and a matching header line
2409 contained an unqualified address, that address was qualified, even if it
2410 did not match any rewriting rules. The underlying bug was that the values
2411 of the flags that permit the existence of unqualified sender and recipient
2412 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2413 non-local messages, and by -bnq for local messages) were not being
2414 preserved with the message after it was received.
2416 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2417 "next input=" as part of the text comprising the host identity instead of
2418 the correct text. The code was using the same buffer for two different
2419 strings. However, depending on which order the printing function evaluated
2420 its arguments, the bug did not always show up. Under Linux, for example, my
2421 test suite worked just fine.
2423 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2424 doesn't work with some older versions of Perl. It has been changed to "my",
2425 which in any case is probably the better facility to use.
2427 15. A really picky compiler found some instances of statements for creating
2428 error messages that either had too many or two few arguments for the format
2431 16. The size of the buffer for calls to the DNS resolver has been increased
2432 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2433 for addresses that have a lot of PTR records. This alleviates a problem; it
2434 does not fully solve it.
2436 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2437 buffer now truncates the list and logs the incident, which is the same
2438 action as happens when Exim is looking up a host name and its aliases.
2439 Previously in this situation something unpredictable would happen;
2440 sometimes it was "internal error: store_reset failed".
2442 18. If a server dropped the connection unexpectedly when an Exim client was
2443 using GnuTLS and trying to read a response, the client delivery process
2444 crashed while trying to generate an error log message.
2446 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2447 string, for example:
2449 warn message = H1: something\nH2: something
2451 the text was added as a single header line from Exim's point of view
2452 though it ended up OK in the delivered message. However, searching for the
2453 second and subsequent header lines using $h_h2: did not work. This has been
2454 fixed. Similarly, if a system filter added multiple headers in this way,
2455 the routers could not see them.
2457 20. Expanded the error message when iplsearch is called with an invalid key to
2458 suggest using net-iplsearch in a host list.
2460 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2461 ACLs are no longer actually imposed (and a message to that effect is
2464 22. If a "gecos" field in a passwd entry contained escaped characters, in
2465 particular, if it contained a \" sequence, Exim got it wrong when building
2466 a From: or a Sender: header from that name. A second bug also caused
2467 incorrect handling when an unquoted " was present following a character
2468 that needed quoting.
2470 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2471 was not being matched caselessly.
2473 24. Arranged for all hyphens in the exim.8 source to be escaped with
2476 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2477 properly. Recipient callout cache records were still being keyed to include
2478 the sender, even when use_sender was set false. This led to far more
2479 callouts that were necessary. The sender is no longer included in the key
2480 when use_sender is false.
2482 26. Added "control = submission" modifier to ACLs.
2484 27. Added the ${base62d: operator to decode base 62 numbers.
2486 28. dnsdb lookups can now access SRV records.
2488 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2489 the configuration file.
2491 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2492 (-v) mode. This makes the output for a verbose queue run more intelligible.
2494 31. Added a use_postmaster feature to recipient callouts.
2496 32. Added the $body_zerocount variable, containing the number of binary zero
2497 bytes in the message body.
2499 33. The time of last modification of the "new" subdirectory is now used as the
2500 "mailbox time last read" when there is a quota error for a maildir
2503 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2505 35. Added +ignore_unknown as a special item in host lists.
2507 36. Code for decoding IPv6 addresses in host lists is now included, even if
2508 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2509 address was recognized as an IP address, but was then not correctly decoded
2510 into binary, causing unexpected and incorrect effects when compared with
2517 1. Very minor rewording of debugging text in manualroute to say "list of
2518 hosts" instead of "hostlist".
2520 2. If verify=header_syntax was set, and a header line with an unqualified
2521 address (no domain) and a large number of spaces between the end of the
2522 name and the colon was received, the reception process suffered a buffer
2523 overflow, and (when I tested it) crashed. This was caused by some obsolete
2524 code that should have been removed. The fix is to remove it!
2526 3. When running in the test harness, delay a bit after writing a bounce
2527 message to get a bit more predictability in the log output.
2529 4. Added a call to search_tidyup() just before forking a reception process. In
2530 theory, someone could use a lookup in the expansion of smtp_accept_max_
2531 per_host which, without the tidyup, could leave open a database connection.
2533 5. Added the variables $recipient_data and $sender_data which get set from a
2534 lookup success in an ACL "recipients" or "senders" condition, or a router
2535 "senders" option, similar to $domain_data and $local_part_data.
2537 6. Moved the writing of debug_print from before to after the "senders" test
2540 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2541 problems for message scanning, either using a data ACL, or using
2542 local_scan() because the Received: header was not generated till after they
2543 were called (in order to set the time as the time of reception completion).
2544 I have revised the way this works. The header is now generated after the
2545 body is received, but before the ACL or local_scan() are called. After they
2546 are run, the timestamp in the header is updated.
2552 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2553 before starting a queue runner without re-exec. This happened only when
2554 deliver_drop_privilege was set or when the Exim user was set to root. The
2555 effect of the bug was that timeouts during subsequent deliveries caused
2556 crashes instead of being properly handled. The handler is now left at its
2557 default (and expected) setting.
2559 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2560 message, again when deliver_drop_privilege is set or Exim is run as root.
2561 The bug described in (1) was not present in this case, but the tidying up
2562 of the other signals was missing. I have made the two cases consistent.
2564 3. The ignore_target_hosts setting on a manualroute router was being ignored
2565 for hosts that were looked up using the /MX notation.
2567 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2570 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2571 operated on the sender address. After changing the $sender_address to <>
2572 for the sender address verify, Exim was re-instated it as the original
2573 (before rewriting) address, but remembering that it had rewritten it, so it
2574 wasn't rewriting it again. This bug also had the effect of breaking the
2575 sender address verification caching when the sender address was rewritten.
2577 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2578 This has been changed so that if the ip literal address matches
2579 ignore_target_hosts, the router declines.
2581 7. Added expansion conditions match_domain, match_address, and match_local_
2582 part (NOT match_host).
2584 8. The placeholder for the Received: header didn't have a length field set.
2586 9. Added code to Exim itself and to exim_lock to test for a specific race
2587 condition that could lead to file corruption when using MBX delivery. The
2588 issue is with the lockfile that is created in /tmp. If this file is removed
2589 after a process has opened it but before that process has acquired a lock,
2590 there is the potential for a second process to recreate the file and also
2591 acquire a lock. This could lead to two Exim processes writing to the file
2592 at the same time. The added code performs the same test as UW imapd; it
2593 checks after acquiring the lock that its file descriptor still refers to
2594 the same named file.
2596 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2597 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2600 11. Added the smtp_active_hostname option. If used, this will typically be made
2601 to depend on the incoming interface address. Because $interface_address is
2602 not set up until the daemon has forked a reception process, error responses
2603 that can happen earlier (such as "too many connections") no longer contain
2606 12. If an expansion in a condition on a "warn" statement fails because a lookup
2607 defers, the "warn" statement is abandoned, and the next ACL statement is
2608 processed. Previously this caused the whole ACL to be aborted.
2610 13. Added the iplsearch lookup type.
2612 14. Added ident_timeout as a log selector.
2614 15. Added tls_certificate_verified as a log selector.
2616 16. Added a global option tls_require_ciphers (compare the smtp transport
2617 option of the same name). This controls incoming TLS connections.
2619 17. I finally figured out how to make tls_require_ciphers do a similar thing
2620 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2621 before starting the TLS session.
2623 18. Tabs are now shown as \t in -bP output.
2625 19. If the log selector return_path_on_delivery was set, Exim crashed when
2626 bouncing a message because it had too many Received: header lines.
2628 20. If two routers both had headers_remove settings, and the first one included
2629 a superfluous trailing colon, the final name in the first list and the
2630 first name in the second list were incorrectly joined into one item (with a
2631 colon in the middle).
2637 1. Added -C and -D options to the exinext utility, mainly to make it easier
2638 to include in the automated testing, but these could be helpful when
2639 multiple configurations are in use.
2641 2. The exinext utility was not formatting the output nicely when there was
2642 an alternate port involved in the retry record key, nor when there was a
2643 message id as well (for retries that were specific to a specific message
2644 and a specific host). It was also confused by IPv6 addresses, because of
2645 the additional colons they contain. I have fixed the IPv4 problem, and
2646 patched it up to do a reasonable job for IPv6.
2648 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2649 delivery, the log line now contains "pipelined" if PIPELINING was used.
2651 4. An SMTP transport process used to panic and die if the bind() call to set
2652 an explicit outgoing interface failed. This has been changed; it is now
2653 treated in the same way as a connect() failure.
2655 5. A reference to $sender_host_name in the part of a conditional expansion
2656 that was being skipped was still causing a DNS lookup. This no longer
2659 6. The def: expansion condition was not recognizing references to header lines
2660 that used bh_ and bheader_.
2662 7. Added the _cache feature to named lists.
2664 8. The code for checking quota_filecount in the appendfile transport was
2665 allowing one more file than it should have been.
2667 9. For compatibility with Sendmail, the command line option
2675 and sets the incoming protocol and host name (for trusted callers). The
2676 host name and its colon can be omitted when only the protocol is to be set.
2677 Note the Exim already has two private options, -pd and -ps, that refer to
2678 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2679 "s", but I don't think that's a major issue.
2681 10. A number of refactoring changes to the code, none of which should affect
2684 (a) The number of logging options was getting close to filling up the
2685 32-bit word that was used as a bit map. I have split them into two classes:
2686 those that are passed in the argument to log_write(), and those that are
2687 only ever tested independently outside of that function. These are now in
2688 separate 32-bit words, so there is plenty of room for expansion again.
2689 There is no change in the user interface or the logging behaviour.
2691 (b) When building, for example, log lines, the code previously used a
2692 macro that called string_cat() twice, in order to add two strings. This is
2693 not really sufficiently general. Furthermore, there was one instance where
2694 it was actually wrong because one of the argument was used twice, and in
2695 one call a function was used. (As it happened, calling the function twice
2696 did not affect the overall behaviour.) The macro has been replaced by a
2697 function that can join an arbitrary number of extra strings onto a growing
2700 (c) The code for expansion conditions now uses a table and a binary chop
2701 instead of a serial search (which was left over from when there were very
2702 few conditions). Also, it now recognizes conditions like "pam" even when
2703 the relevant support is not compiled in: a suitably worded error message is
2704 given if an attempt is made to use such a condition.
2706 11. Added ${time_interval:xxxxx}.
2708 12. A bug was causing one of the ddress fields not to be passed back correctly
2709 from remote delivery subprocesses. The field in question was not being
2710 subsequently used, so this caused to problems in practice.
2712 13. Added new log selectors queue_time and deliver_time.
2714 14. Might have fixed a bug in maildirsizefile handling that threw up
2715 "unexpected character" debug warnings, and recalculated the data
2716 unnecessarily. In any case, I expanded the warning message to give more
2719 15. Added the message "Restricted characters in address" to the statements in
2720 the default ACL that block characters like @ and % in local parts.
2722 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2723 Three changes have been made:
2725 (a) There was a serious bug; a negative response to MAIL caused the whole
2726 recipient domain to be cached as invalid, thereby blocking all messages
2727 to all local parts at the same domain, from all senders. This bug has
2728 been fixed. The domain is no longer cached after a negative response to
2729 MAIL if the sender used is not empty.
2731 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2734 (c) A new callout option, "use_sender" has been added for people who want
2735 the modified behaviour.
2741 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2744 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2745 indeed breaks things for older releases.
2747 3. Added additional logging to the case where there is a problem reading data
2748 from a filter that is running in a subprocess using a pipe, in order to
2749 try to track down a specific problem.
2751 4. Testing facility fudge: when running in the test harness and attempting
2752 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2753 getting "No route to host". Convert this to a timeout.
2755 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2758 6. Some OS don't have socklen_t but use size_t instead. This affects the
2759 fifth argument of getsockopt() amongst other things. This is now
2760 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2761 can be set for individual OS. I have set it for SunOS5, OSF1, and
2762 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2763 some earlier ones do not.
2765 7. Change 4.30/15 was not doing the test caselessly.
2767 8. The standard form for an IPv6 address literal was being rejected by address
2768 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2769 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2770 this, as well as the form without the "IPv6" on the front (but only when
2771 address literals are enabled, of course).
2773 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2775 10. Exim crashed if a message with an empty sender address specified by -f
2776 encountered a router with an errors_to setting. This could be provoked only
2777 by a command such as
2781 where an empty string was supplied; "<>" did not hit this bug.
2783 11. Installed PCRE release 4.5.
2785 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2786 remained set. It is now erased.
2788 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2789 times from message ids (which are base 36 rather than the normal 62).
2791 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2792 were being counted as actual protocol errors, and logged if the log
2793 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2794 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2795 and DATA following a set of rejected RCPTs do not count as protocol errors.
2796 In other words, Exim assumes they were pipelined, though this may not
2797 actually be the case. Of course, in all cases the client gets an
2798 appropriate error code.
2800 15. If a lookup fails in an ACL condition, a message about the failure may
2801 be available; it is used if testing the ACL cannot continue, because most
2802 such messages specify what the cause of the deferral is. However, some
2803 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2804 that caused an old message to be retained and used if a later statement
2805 caused a defer, replacing the real cause of the deferral.
2807 16. If an IP address had so many PTR records that the DNS lookup buffer
2808 was not large enough to hold them, Exim could crash while trying to process
2809 the truncated data. It now detects and logs this case.
2811 17. Further to 4.21/58, another change has been made: if (and only if) the
2812 first line of a message (the first header line) ends with CRLF, a bare LF
2813 in a subsequent header line has a space inserted after it, so as not to
2814 terminate the header.
2816 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2817 unnecessarily, used atoi() instead of strtol(), and didn't check the
2818 termination when getting file sizes from file names by regex.
2820 19. Completely re-implemented the support for maildirsize files, in the light
2821 of a number of problems with the previous contributed implementation
2822 (4.30/29). In particular:
2824 . If the quota is zero, the maildirsize file is maintained, but no quota is
2827 . If the maildir directory does not exist, it is created before any attempt
2828 to write a maildirsize file.
2830 . The quota value in the file is just a cache; if the quota is changed in
2831 the transport, the new value overrides.
2833 . A regular expression is available for excluding directories from the
2836 20. The autoreply transport checks the characters in options that define the
2837 message's headers; it allows continued headers, but it was checking with
2838 isspace() after an embedded newline instead of explicitly looking for a
2841 21. If all the "regular" hosts to which an address was routed had passed their
2842 expiry times, and had not reached their retry times, the address was
2843 bounced, even if fallback hosts were defined. Now Exim should go on to try
2846 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2847 equivalent code in the SMTP transport. Some hosts send humungous responses
2848 to HELO/EHLO, more than 1024 it seems.
2850 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2851 gives compiler warnings in some environments. I've now done it "properly",
2854 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2855 (because of problems with the built-in one) was declared to return uschar *
2856 instead of char *, causing compiler failure.
2858 25. Fixed a file descriptor leak when processing alias/forward files.
2860 26. Fixed a minor format string issue in dbfn.c.
2862 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2864 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2865 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2868 29. When a sender address is verified, it is cached, to save repeating the test
2869 when there is more than one recipient in a message. However, when the
2870 verification involves a callout, it is possible for different callout
2871 options to be set for different recipients. It is too complicated to keep
2872 track of this in the cache, so now Exim always runs a verification when a
2873 callout is required, relying on the callout cache for the optimization.
2874 The overhead is duplication of the address routing, but this should not be
2877 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2878 to be verified with callout=postmaster, and the main callout worked but the
2879 postmaster check failed, the verification correctly failed. However, if a
2880 subsequent RCPT command asked for sender verification *without* the
2881 postmaster check, incorrect caching caused this verification also to fail,
2884 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2885 it was not caching the DNS options (qualify_single, search_parents) that
2886 were used when the lookup failed. A subsequent lookup with different
2887 options therefore always gave the same answer, though there were cases
2888 where it should not have. (Example: a "domains = !$mx_any" option on a
2889 dnslookup router: the "domains" option is always processed without any
2890 widening, but the router might have qualify_single set.) Now Exim uses the
2891 cached value only when the same options are set.
2893 32. Added John Jetmore's "exipick" utility to the distribution.
2895 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2896 than a timeout (e.g. a certificate is required, and is not provided), an
2897 Exim server now closes the connection immediately. Previously it waited for
2898 the client to close - but if the client is SSL, it seems that they each
2899 wait for each other, leading to a delay before one of them times out.
2901 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2902 maintained 0.8.x compatibility because I don't think many are using it, and
2903 it is clearly obsolete.
2905 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2908 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2909 client certificate was expired. A simple patch fixes this, though I don't
2910 understand the full logic of why the verify callback is called multiple
2913 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2914 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2915 which causes problems with some clients (such as the Certicom SSL Plus
2916 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2917 disables the coutermeasure allowing Eudora to connect."
2919 38. Exim was not checking that a write() to a log file succeeded. This could
2920 lead to Bad Things if a log got too big, in particular if it hit a file
2921 size limit. Exim now panics and dies if it cannot write to a log file, just
2922 as it does if it cannot open a log file.
2924 39. Modified OS/Makefile-Linux so that it now contains
2926 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2928 The two -D definitions ensure that Exim is compiled with large file
2929 support, which makes it possible to handle log files that are bigger than
2932 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2933 instance) a domain was checked against a named list that involved a lookup,
2934 causing $domain_data to be set, then another domain was checked against the
2935 same list, then the first domain was re-checked, the value of $domain_data
2936 after the final check could be wrong. In particular, if the second check
2937 failed, it could be set empty. This bug probably also applied to
2940 41. The strip_trailing_dot option was not being applied to the address given
2941 with the -f command-line option.
2943 42. The code for reading a message's header from the spool was incrementing
2944 $received_count, but never initializing it. This meant that the value was
2945 incorrect (doubled) while delivering a message in the same process in which
2946 it was received. In the most common configuration of Exim, this never
2947 happens - a fresh exec is done - but it can happen when
2948 deliver_drop_privilege is set.
2950 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2951 it now includes up to 150 characters of the unexpected data in the log
2954 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2955 and building data strings. The size of both of these buffers was 10 000
2956 bytes - far larger than anybody would *ever* want, thought I. Needless to
2957 say, somebody hit the limit. I have increased the maximum line length to
2958 20 000 and the maximum data length of concatenated lines to 100 000. I have
2959 also fixed two bugs, because there was no checking on these buffers. Tsk,
2960 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2961 buffer is too small.
2963 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2964 lsearch lookups. Now it does.
2966 46. When parsing a route_list item in a manualroute router, a fixed-length
2967 buffer was used for the list of hosts. I made this 1024 bytes long,
2968 thinking that nobody would ever have a list of hosts that long. Wrong.
2969 Somebody had a whole pile of complicated expansion conditions, and the
2970 string was silently truncated, leading to an expansion error. It turns out
2971 that it is easier to change to an unlimited length (owing to other changes
2972 that have happened since this code was originally written) than to build
2973 structure for giving a limitation error. The length of the item that
2974 expands into the list of hosts is now unlimited.
2976 47. The lsearch lookup could not handle data where the length of text line was
2977 more than 4095 characters. Such lines were truncated, leading to shortened
2978 data being returned. It should now handle lines of any length.
2980 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2981 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2984 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2985 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2986 no longer happens in comment lines. A list of replacements is now placed
2987 at the head of all of the source files, except those whose only change is
2988 to replace PERL_COMMAND in the very first #! line.
2990 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2991 messages on the queue, with a bottom-up merge sort, using code contributed
2992 by Michael Haardt. This should make operations like -bp somewhat faster on
2993 large queues. It won't affect queue runners, except when queue_run_in_order
2996 51. Installed eximstats 1.31 in the distribution.
2998 52. Added support for SRV lookups to the dnslookup router.
3000 53. If an ACL referred to $message_body or $message_body_end, the value was not
3001 reset for any messages that followed in the same SMTP session.
3003 54. The store-handling optimization for building very long strings was not
3004 differentiating between the different store pools. I don't think this
3005 actually made any difference in practice, but I've tidied it.
3007 55. While running the routers to verify a sender address, $sender_address
3008 was still set to the sender address. This is wrong, because when routing to
3009 send a bounce to the sender, it would be empty. Therefore, I have changed
3010 it so that, while verifying a sender address, $sender_address is set to <>.
3011 (There is no change to what happens when verifying a recipient address.)
3013 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3014 target A or AAAA records (if not already returned) without resetting the
3015 qualify_single or search_parents options of the DNS resolver. These are
3016 inappropriate in this case because the targets of MX and SRV records must
3017 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3018 target that, when qualified, matched something in the local domain. These
3019 two options are now turned off when doing these lookups.
3021 57. It seems that at least some releases of Reiserfs (which does not have the
3022 concept of a fixed number of inodes) returns zero and not -1 for the
3023 number of available inodes. This interacted badly with check_spool_inodes,
3024 which assumed that -1 was the "no such thing" setting. What I have done is
3025 to check that the total number of inodes is greater than zero before doing
3026 the test of how many are available.
3028 58. When a "warn" ACL statement has a log_message modifier, the message is
3029 remembered, and not repeated. This is to avoid a lot of repetition when a
3030 message has many recipients that cause the same warning to be written.
3031 Howewer, Exim was preserving the list of already written lines for an
3032 entire SMTP session, which doesn't seem right. The memory is now reset if a
3033 new message is started.
3035 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3036 debugging output unless log_rewrite was also set.
3038 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3039 of (int)(handle) when we know that handle contains (void *)(-1).
3041 61. The Exim daemon panic-logs an error return when it closes the incoming
3042 connection. However "connection reset by peer" seems to be common, and
3043 isn't really an error worthy of noting specially, so that particular error
3046 62. When Exim is trying to find all the local interfaces, it used to panic and
3047 die if the ioctl to get the interface flags failed. However, it seems that
3048 on at least one OS (Solaris 9) it is possible to have an interface that is
3049 included in the list of interfaces, but for which you get a failure error
3050 for this call. This happens when the interface is not "plumbed" into a
3051 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3052 failure of the "get flags" call assumes that the interface is down.
3054 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3055 makes life easier for people who are doing arithmetic on fields extracted
3056 from dates, where you often get leading zeros that should not be
3057 interpreted as octal.
3059 64. Added qualify_domain to the redirect router, to override the global
3062 65. If a pathologically long header line contained very many addresses (the
3063 report of this problem mentioned 10 000) and each of them was rewritten,
3064 Exim could use up a very large amount of memory. (It kept on making new
3065 copies of the header line as it rewrote, and never released the old ones.)
3066 At the expense of a bit more processing, the header rewriting function has
3067 been changed so that it no longer eats memory in this way.
3069 66. The generation of the Received: header has been moved from the time that a
3070 message starts to be received, to the time that it finishes. The timestamp
3071 in the Received: header should now be very close to that of the <= log
3072 line. There are two side-effects of this change:
3074 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3075 logged header lines no longer include the local Received: line, because
3076 it has not yet been created. The same applies to a copy of the message
3077 that is returned to a non-SMTP sender when a message is rejected.
3079 (b) When a filter file is tested using -bf, no additional Received: header
3080 is added to the test message. After some thought, I decided that this
3083 This change does not affect the value of $received_for. It is still set
3084 after address rewriting, but before local_scan() is called.
3086 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3088 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3089 gave an unhelpful panic error message, and a defer error. I have managed to
3090 change this behaviour so that it now rejects any supplied certificate,
3091 which seems right, as the list of acceptable certificates is empty.
3093 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3094 gave an unhelpful defer error. I have not managed to make this reject any
3095 supplied certificates, but the error message it gives is "no certificate
3096 supplied", which is not helpful.
3098 70. exigrep's output now also includes lines that are not associated with any
3099 message, but which match the given pattern. Implemented by a patch from
3100 Martin Sluka, which also tidied up the Perl a bit.
3102 71. Recipient callout verification, like sender verification, was using <> in
3103 the MAIL FROM command. This isn't really the right thing, since the actual
3104 sender may affect whether the remote host accepts the recipient or not. I
3105 have changed it to use the actual sender in the callout; this means that
3106 the cache record is now keyed on a recipient/sender pair, not just the
3107 recipient address. There doesn't seem to be a real danger of callout loops,
3108 since a callout by the remote host to check the sender would use <>.
3109 [SEE ABOVE: changed after hitting problems.]
3111 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3112 temporary errors. However, in the case of such a code being given after
3113 the end of a data transmission (i.e. after ".") Exim was failing to write
3114 a retry record for the message. (Yes, there was some broken host that was
3115 actually sending 8xx at this point.)
3117 73. An unknown lookup type in a host list could cause Exim to panic-die when
3118 the list was checked. (An example that provoked this was putting <; in the
3119 middle of a list instead of at the start.) If this happened during a DATA
3120 ACL check, a -D file could be left lying around. This kind of configuration
3121 error no longer causes Exim to die; instead it causes a defer errror. The
3122 incident is still logged to the main and panic logs.
3124 74. Buglet left over from Exim 3 conversion. The message "too many messages
3125 in one connection" was written to the rejectlog but not the mainlog, except
3126 when address rewriting (yes!) was being logged.
3128 75. Added write_rejectlog option.
3130 76. When a system filter was run not as root (that is, when system_filter_user
3131 was set), the values of the $n variables were not being returned to the
3132 main process; thus, they were not subsequently available in the $sn
3135 77. Added +return_path_on_delivery log selector.
3137 78. A connection timeout was being treated differently from recipients deferred
3138 when testing hosts_max_try with a message that was older than the host's
3139 retry timeout. (The host should not be counted, thus allowing all hosts to
3140 be tried at least once before bouncing.) This may have been the cause of an
3141 occasionally reported bug whereby a message would remain on the queue
3142 longer than the retry timeout, but would be bounced if a delivery was
3143 forced. I say "may" because I never totally pinned down the problem;
3144 setting up timeout/retry tests is difficult. See also the next item.
3146 79. The ultimate address timeout was not being applied to errors that involved
3147 a combination of host plus message (for example, a timeout on a MAIL
3148 command). When an address resolved to a number of possible hosts, and they
3149 were not all tried for each delivery (e.g. because of hosts_max_try), a
3150 message could remain on the queue longer than the retry timeout.
3152 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3155 81. Fixed an obscure SMTP outgoing bug which required at least the following
3156 conditions: (a) there was another message waiting for the same server;
3157 (b) the server returned 5xx to all RCPT commands in the first message so
3158 that the message was not completed; (c) the server dropped the connection
3159 or gave a negative response to the RSET that Exim sends to abort the
3160 transaction. The observed case was a dropped connection after DATA that had
3161 been sent in pipelining mode. That is, the server had advertised PIPELINING
3162 but was not implementing it correctly. The effect of the bug was incorrect
3163 behaviour, such as trying another host, and this could lead to a crash.
3169 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3170 and daemon.c were passed as pointers to ints; they should have been
3171 pointers to socklen_t variables (which are typically unsigned ints).
3173 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3176 3. Fixed a really odd bug that affected only the testing scheme; patching a
3177 certain fixed string in the binary changed the value of another string that
3178 happened to be identical to the end of the original first string.
3180 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3181 name", it returns that address as the IP address. On some operating
3182 systems (e.g. Solaris), it also passes back the IP address string as the
3183 "host name". However, on others (e.g. Linux), it passes back an empty
3184 string. Exim wasn't checking for this, and was changing the host name to an
3185 empty string, assuming it had been canonicized.
3187 5. Although rare, it is permitted to have more than one PTR record for a given
3188 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3189 all the names associated with an address, because they do in Solaris.
3190 However, it seems that they do not in Linux for data that comes from the
3191 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3192 I found this out when I moved to a new Linux workstation and tried to run
3193 the Exim test suite.
3195 To get round this problem I have changed the code so that it now does its
3196 own call to the DNS to look up PTR records when searching for a host name.
3197 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3198 addresses that are only in /etc/hosts are still found.
3200 This behaviour is, however, controlled by an option called host_lookup_
3201 order, which defaults to "bydns:byaddr". If people want to use the other
3202 order, or indeed, just use one or the other means of lookup, they can
3203 specify it in this variable.
3205 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3206 some operating systems, this comes back from gethostbyaddr() as an empty
3207 string, and this is what Exim used to test for. However, it seems that in
3208 other systems, "." is yielded. Exim now tests for this case too.
3210 7. The values of check_spool_space and check_log_space are now held internally
3211 as a number of kilobytes instead of an absolute number of bytes. If a
3212 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3213 kilobyte. This means that much larger values can be stored.
3215 8. Exim monitor: an attempt to get the action menu when not actually pointing
3216 at a message produces an empty menu entitled "No message selected". This
3217 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3218 no entries in it ("Shell widget menu has zero width and/or height"). So I
3219 have added a single, blank menu entry in this case.
3221 9. Added ${quote_local_part.
3223 10. MIME decoding is now applied to the contents of Subject: header lines when
3226 11. Now that a reference to $sender_host_address automatically causes a reverse
3227 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3228 host lookup before query-style lookups in lists that might use this
3229 variable. This has therefore been abolished, and the "net-" prefix is no
3230 longer necessary for query-style lookups.
3232 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3233 have been a typo for LFLAGS, so it has been changed.
3235 13. The install script calls Exim with "-C /dev/null" in order to find the
3236 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3237 to be output. Howeve, since Exim outputs its version number before the
3238 error, it didn't break the script. It just looked ugly. I fixed this by
3239 always allowing "-C /dev/null" if the caller is root.
3241 14. Ignore overlarge ACL variable number when reading spool file - insurance
3242 against a later release with more variables having written the file.
3244 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3245 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3246 this, as well as the form without the "IPv6" on the front.
3248 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3249 OS/Makefile-Darwin file.
3251 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3252 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3254 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3255 variable. However, some OS use ints for inodes. I've added cast to long int
3256 to get rid of the compiler warning.
3258 19. I had forgotten to lock out "/../" in configuration file names when
3259 ALT_CONFIG_PREFIX was set.
3261 20. Routers used for verification do not need to specify transports. However,
3262 if such a router generated a host list, and callout was configured, Exim
3263 crashed, because it could not find a port number from the (non-existent)
3264 transport. It now assumes port 25 in this circumstance.
3266 21. Added the -t option to exigrep.
3268 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3269 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3270 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3271 them is set, LOOKUP_LSEARCH is forced.
3273 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3275 24. Added sender and host information to the "rejected by local_scan()" log
3276 line; previously there was no indication of these.
3278 25. Added .include_if_exists.
3280 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3281 Linux. It turns out that not all file systems support this. Apparently some
3282 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3283 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3284 not supported on the file descriptor, is now ignored when Exim is trying to
3285 sync a directory. This applies only to Linux.
3287 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3289 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3290 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3293 29. Added support for maildirsize files from supplied patch (modified a bit).
3295 30. The use of :fail: followed by an empty string could lead Exim to respond to
3296 sender verification failures with (e.g.):
3298 550 Verification failed for <xxx>
3299 550 Sender verify failed
3301 where the first response line was missing the '-' that indicates it is not
3302 the final line of the response.
3304 31. The loop for finding the name of the user that called Exim had a hardwired
3305 limit of 10; it now uses the value of finduser_retries, which is used for
3306 all other user lookups.
3308 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3311 33. Exim was neglecting to zero errno before one call of strtol() when
3312 expanding a string and expecting an integer value. On some systems this
3313 resulted in spurious "integer overflow" errors. Also, it was casting the
3314 result into an int without checking.
3316 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3317 did not work. The code looks as if it has *never* worked, though it appears
3318 to have been documented since at least releast 1.62. I have made it work.
3320 35. The "timeout_DNS" error in retry rules, also documented since at least
3321 1.62, also never worked. As it isn't clear exactly what this means, and
3322 clearly it isn't a major issue, I have abolished the feature by treating it
3323 as "timeout", and writing a warning to the main and panic logs.
3325 36. The display of retry rules for -brt wasn't always showing the error code
3328 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3329 timeout_connect_A, timeout_connect_MX.
3331 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3332 to the empty sender.
3334 39. The daemon was not analysing the content of -oX till after it had closed
3335 stderr and disconnected from the controlling terminal. This meant that any
3336 syntax errors were only noted on the panic log, and the return code from
3337 the command was 0. By re-arranging the code a little, I've made the
3338 decoding happen first, so such errors now appear on stderr, and the return
3339 code is 1. However, the actual setting up of the sockets still happens in
3340 the disconnected process, so errors there are still only recorded on the
3343 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3344 connections (as happens on some IP stacks) was logged at start up time as
3345 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3346 it from "IPv6 and IPv4", which means that two separate sockets are being
3349 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3350 says whether AF_INET or AF_INET6 was passed as an argument.
3352 42. Exiwhat output was messed up when time zones were included in log
3355 43. Exiwhat now gives more information about the daemon's listening ports,
3356 and whether -tls-on-connect was used.
3358 44. The "port" option of the smtp transport is now expanded.
3360 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3361 silently ignored. Now an error message is written to the main and panic
3364 46. There's a new ACL modifier called "logwrite" which writes to a log file
3365 as soon as it is encountered.
3367 47. Added $local_user_uid and $local_user_gid at routing time.
3369 48. Exim crashed when trying to verify a sender address that was being
3372 49. Exim was recognizing only a space character after ".include". It now also
3373 recognizes a tab character.
3375 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3376 extracting the relevant information from the specification. The man page no
3377 longer contains scrambled data for the -d option, and I've added a section
3378 at the front about calling Exim under different names.
3380 51. Added "extra_headers" argument to the "mail" command in filter files.
3382 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3385 53. Installed eximstats 1.29.
3387 54. Added transport_filter_timeout as a generic transport option.
3389 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3390 Cc: header lines. This was required by RFC 822, but it not required by RFC
3393 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3394 incoming messages that did not have them. Now it does so only if the
3395 message originates locally, that is, if there is no associated remote host
3396 address. When Resent- header lines are present, this applies to the Resent-
3397 lines rather than the non-Resent- lines.
3399 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3400 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3402 58. Messages for configuration errors now include the name of the main
3403 configuration file - useful now that there may be more than one file in a
3404 list (.included file names were always shown).
3406 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3407 for those rare installations that do not start the daemon as root or run it
3408 setuid root. I've cut out the call to initgroups() if the daemon is not
3411 60. The Exim user and group can now be bound into the binary as text strings
3412 that are looked up at the start of Exim's processing.
3414 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3416 62. Added $mailstore_basename variable.
3418 63. Installed patch to sieve.c from Michael Haardt.
3420 64. When Exim failed to open the panic log after failing to open the main log,
3421 the original message it was trying to log was written to stderr and debug
3422 output, but if they were not available (the usual case in production), it
3423 was lost. Now it is written to syslog before the two lines that record the
3424 failures to open the logs.
3426 65. Users' Exim filters run in subprocesses under the user's uid. It is
3427 possible for a "deliver" command or an alias in a "personal" command to
3428 provoke an address rewrite. If logging of address rewriting is configured,
3429 this fails because the process is not running as root or exim. There may be
3430 a better way of dealing with this, but for the moment (because 4.30 needs
3431 to be released), I have disabled address rewrite logging when running a
3432 filter in a non-root, non-exim process.
3438 1. The buildconfig auxiliary program wasn't quoting the value set for
3439 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3440 not defined. This bug was masked in 4.22 by the effect that was fixed in
3443 2. Some messages that were rejected after a message id was allocated were
3444 shown as "incomplete" by exigrep. It no longer does this for messages that
3445 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3447 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3448 have allow_domain_literals set, the ID did not get logged in the <= line.
3449 Domain literals are now always recognized in Message-ID: header lines.
3451 4. The first argument for a ${extract expansion item is the key name or field
3452 number. Leading and trailing spaces in this item were not being ignored,
3453 causing some misleading effects.
3455 5. When deliver_drop_privilege was set, single queue runner processes started
3456 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3457 same command in the process it spins off) were not dropping privilege.
3459 6. When the daemon running as "exim" started a queue runner, it always
3460 re-executed Exim in the spun-off process. This is a waste of effort when
3461 deliver_drop_privilege is set. The new process now just calls the
3462 queue-runner function directly.
3468 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3471 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3472 ignored. Though the use of -oP was forcing the writing of a pid file, it
3473 was always written to the default place.
3475 3. If the message "no IP address found for host xxxx" is generated during
3476 incoming verification, it is now followed by identification of the incoming
3477 connection (so you can more easily find what provoked it).
3479 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3481 5. Added some features to "harden" Exim a bit more against certain attacks:
3483 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3484 be put in Local/Makefile. This is like the never_users runtime option,
3485 but it cannot be overridden. The default setting is "root".
3487 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3488 prefix string with which any file named in a -C command line option
3491 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3492 is retained for -C and -D only if the caller of Exim is root. Without
3493 it, the exim user may also use -C and -D and retain privilege.
3495 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3496 command line option is disabled.
3498 6. Macro names set by the -D option must start with an upper case letter, just
3499 like macro names defined in the configuration file.
3501 7. Added "dereference=" facility to LDAP.
3503 8. Two instances of the typo "uknown" in the source files are fixed.
3505 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3506 the Configure-Makefile script screwed up while processing it.
3508 10. Incorporated PCRE 4.4.
3510 11. The SMTP synchronization check was not operating right at the start of an
3511 SMTP session. For example, it could not catch a HELO sent before the client
3512 waited for the greeting. There is now a check for outstanding input at the
3513 point when the greeting is written. Because of the duplex, asynchronous
3514 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3515 way, but not yet received, when the check is performed.
3517 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3518 on TCP/IP sockets, because this apparently causes some broken clients to
3521 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3522 unchanged) from the Cygwin maintainer.
3524 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3525 was supported instead of testing for mbx. Effectively a typo.
3527 15. The spa authenticator server code was not checking that the input it
3528 received was valid base64.
3530 16. The debug output line for the "set" modifier in ACLs was not showing the
3531 name of the variable that was being set.
3533 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3535 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3536 lookup on its own. Something else was needed to trigger the lookup. For
3537 example, a match in host_lookup or the need for a host name in a host list.
3538 Now, if $sender_host_name is referenced and the host name has not yet been
3539 looked up, a lookup is performed. If the lookup fails, the variable remains
3540 empty, and $host_lookup_failed is set to "1".
3542 19. Added "eqi" as a case-independent comparison operator.
3544 20. The saslauthd authentication condition could segfault if neither service
3545 nor realm was specified.
3547 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3548 error message that was logged was misleading, and incoming SMTP
3549 connections were dropped. The message is now more accurate, and temporary
3550 errors are given to SMTP connections.
3552 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3553 (except RSET) with a 421 error, until QUIT is received. However, it was
3554 failing to send a response to QUIT.
3556 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3557 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3558 ACL is now run after the helo_try_verify_hosts code.
3560 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3561 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3562 case-independent, so other case variants are also recognized). Apparently
3563 some systems use these upper case variants.
3565 25. If more than two messages were waiting for the same host, and a transport
3566 filter was specified for the transport, Exim sent two messages over the
3567 same TCP/IP connection, and then failed with "socket operation on non-
3568 socket" when it tried to send the third.
3570 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3572 27. The extern definition of crypt16() in expand.c was not being excluded when
3573 the OS had its own crypt16() function.
3575 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3576 as a preferred synonym for return_size_limit, both as an option and as an
3579 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3581 30. Changed the default configuration ACL to relax the local part checking rule
3582 for addresses that are not in any local domains. For these addresses,
3583 slashes and pipe symbols are allowed within local parts, but the sequence
3584 /../ is explicitly forbidden.
3586 31. SPA server authentication was not clearing the challenge buffer before
3589 32. log_message in a "warn" ACL statement was writing to the reject log as
3590 well as to the main log, which contradicts the documentation and doesn't
3591 seem right (because no rejection is happening). So I have stopped it.
3593 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3594 However, I am unable to do any testing of this.
3596 34. Fixed an infelicity in the appendfile transport. When checking directories
3597 for a mailbox, to see if any needed to be created, it was accidentally
3598 using path names with one or more superfluous leading slashes; tracing
3599 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3601 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3602 contents are added to the log line that is written for every discarded
3603 recipient. (Previously a log_message setting was ignored.)
3605 36. The ${quote: operator now quotes the string if it is empty.
3607 37. The install script runs exim in order to find its version number. If for
3608 some reason other than non-existence or emptiness, which it checks, it
3609 could not run './exim', it was installing it with an empty version number,
3610 i.e. as "exim-". This error state is now caught, and the installation is
3613 38. An argument was missing from the function that creates an error message
3614 when Exim fails to connect to the socket for saslauthd authentication.
3615 This could cause Exim to crash, or give a corrupted message.
3617 39. Added isip, isip4, and isip6 to ${if conditions.
3619 40. The ACL variables $acl_xx are now saved with the message, and can be
3620 accessed later in routers, transports, and filters.
3622 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3623 strings in the file are not string-expanded.
3625 42. If a MAIL command specified a SIZE value that was too large to fit into an
3626 int variable, the check against message_size_limit failed. Such values are
3627 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3628 day this will have to be increased, but I don't think I want to be around
3629 when emails are that large.
3636 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3637 iconv() is not standard in FreeBSD.
3639 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3640 IPv6 enabled. The observed symptom was a segmentation fault on return from
3641 the function os_common_find_running_interfaces() in src/os.c.
3643 3. In the check_special_case() function in daemon.c I had used "errno" as an
3644 argument name, which causes warnings on some systems. This was basically a
3645 typo, since it was named "eno" in the comments!
3647 4. The code that waits for the clock to tick (at a resolution of some fraction
3648 of a second) so as to ensure message-id uniqueness was always waiting for
3649 at least one whole tick, when it could have waited for less. [This is
3650 almost certainly not relevant at current processor speeds, where it is
3651 unlikely to ever wait at all. But we try to future-proof.]
3653 5. The function that sleeps for a time interval that includes fractions of a
3654 second contained a race. It did not block SIGALRM between setting the
3655 timer, and suspending (a couple of lines later). If the interval was short
3656 and the sigsuspend() was delayed until after it had expired, the suspension
3657 never ended. On busy systems this could lead to processes getting stuck for
3660 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3661 process, before it forks any delivery processes. The open lookup caching
3662 mechanism meant that the open file or database connection was passed into
3663 the delivery process. The problem was that delivery processes always tidy
3664 up cached lookup data. This could cause a problem for the next delivery
3665 process started by the queue runner, because the external queue runner
3666 process does not know about the closure. So the next delivery process
3667 still has data in the lookup cache. In the case of a file lookup, there was
3668 no problem because closing a file descriptor in a subprocess doesn't affect
3669 the parent. However, if the lookup was caching a connection to a database,
3670 the connection was closed, and the second delivery process was likely to
3671 see errors such as "PGSQL: query failed: server closed the connection
3672 unexpectedly". The problem has been fixed by closing all cached lookups
3673 in a queue runner before running a delivery process.
3675 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3676 seem to have the "const" qualifier which it has on other OS. I've
3679 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3680 *of the same type* (client or server) with the same public name that an
3681 error should be diagnosed.
3683 9. When Exim looked up a host name for an IP address, but failed to find the
3684 original IP address when looking up the host name (a safety check), it
3685 output the message "<ip address> does not match any IP for NULL", which was
3686 confusing, to say the least. The bug was that the host name should have
3687 appeared instead of "NULL".
3689 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3690 user that is built into the binary, and the -C or -D options is used, root
3691 privilege is dropped before the configuration file is read. In addition,
3692 logging is switched to stderr instead of the normal log files. If the
3693 configuration then re-defines the Exim user, the unprivileged environment
3694 is probably not what is expected, so Exim logs a panic warning message (but
3697 However, if deliver_drop_privilege is set, the unprivileged state may well
3698 be exactly what is intended, so the warning has been cut out in that case,
3699 and Exim is allowed to try to write to its normal log files.
3705 1. smtp_return_error_details was not giving details for temporary sender
3706 or receiver verification errors.
3708 2. Diagnose a configuration error if two authenticators have the same public
3711 3. Exim used not to create the message log file for a message until the first
3712 delivery attempt. This could be confusing when incoming messages were held
3713 for policy or load reasons. The message log file is now created at the time
3714 the message is received, and an initial "Received" line is written to it.
3716 4. The automatically generated man page for command line options had a minor
3717 bug that caused no ill effects; however, a more serious problem was that
3718 the procedure for building the man page automatically didn't always
3719 operate. Consequently, release 4.20 contains an out-of-date version. This
3720 shouldn't happen again.
3722 5. When building Exim with embedded Perl support, the script that builds the
3723 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3724 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3726 6. The freeze_tell option was not being used for messages that were frozen on
3727 arrival, either by an ACL or by local_scan().
3729 7. Added the smtp_incomplete_transaction log selector.
3731 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3732 was accepting AUTH without a new EHLO.
3734 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3735 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3738 10. Logging of TCP/IP connections (when configured) now happens in the main
3739 daemon process instead of the child process, so that the TCP/IP connection
3740 count is more accurate (but it can never be perfect).
3742 11. The use of "drop" in a nested ACL was not being handled correctly in the
3743 outer ACL. Now, if condition failure induced by the nested "drop" causes
3744 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3745 or "require"), the connection is dropped.
3747 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3748 that yield "discard" can now be used with an "accept" or a "discard" verb,
3749 but an error is generated for any others (because I can't see a useful way
3750 to define what should happen).
3752 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3753 are now processed in the same way as lines in the Exim configuration file.
3754 In particular, continuation lines are supported.
3756 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3758 15. Added -ti meaning -t -i.
3760 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3761 domains, and warn by logging if others are found.
3763 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3764 returned from GIFCONF when Exim is trying to find the list of interfaces on
3765 a host. The code in os.c has been modified to copy each ifreq to an aligned
3766 structure in all cases.
3768 Also, in some cases, the returned ifreq's were being copied to a 'struct
3769 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3770 means the last couple of bytes of an IPv6 address could be chopped if the
3771 ifreq contained only a normal sockaddr (14 bytes storage).
3773 18. Named domain lists were not supported in the hosts_treat_as_local option.
3774 An entry such as +xxxx was not recognized, and was treated as a literal
3777 19. Ensure that header lines added by a DATA ACL are included in the reject log
3778 if the ACL subsequently rejects the message.
3780 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3781 MD5 (which is deprecated).
3783 21. When testing a filter file using -bf, Exim was writing a message when it
3784 took the sender from a "From " line in the message, but it was not doing so
3785 when it took $return_path from a Return-Path: header line. It now does.
3787 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3788 with a valid header line field name (a series of printing characters
3789 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3791 23. Changed "disc" in the source to "disk" to conform to the documentation and
3792 the book and for uniformity.
3794 24. Ignore Sendmail's -Ooption=value command line item.
3796 25. When execve() failed while trying to run a command in a pipe transport,
3797 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3798 could be confused with a return value of 69 from the command itself. This
3799 has been changed to 127, the value the shell returns if it is asked to run
3800 a non-existent command. The wording for the related log line suggests a
3801 non-existent command as the problem.
3803 26. If received_header_text expands to an empty string, do not add a Received:
3804 header line to the message. (Well, it adds a token one on the spool, but
3805 marks it "old" so that it doesn't get used or transmitted.)
3807 27. Installed eximstats 1.28 (addition of -nt option).
3809 28. There was no check for failure on the call to getsockname() in the daemon
3810 code. This can fail if there is a shortage of resources on the system, with
3811 ENOMEM, for example. A temporary error is now given on failure.
3813 29. Contrary to the C standard, it seems that in some environments, the
3814 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3815 program. Exim now does this explicitly; it affects the formatting of
3816 timestamps using strftime().
3818 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3819 complaints. I've now initialized all the variables, to avoid this.
3821 32. Header lines added by a system filter were not being "seen" during
3822 transport-time rewrites.
3824 33. The info_callback() function passed to OpenSSL is set up with type void
3825 (*)(SSL *, int, int), as described somewhere. However, when calling the
3826 function (actually a macro) that sets it up, the type void(*)() is
3827 expected. I've put in a cast to prevent warnings from picky compilers.
3829 34. If a DNS black list lookup found a CNAME record, but there were no A
3830 records associated with the domain it pointed at, Exim crashed.
3832 35. If a DNS black list lookup returned more than one A record, Exim ignored
3833 all but the first. It now scans all returned addresses if a particular IP
3834 value is being sought. In this situation, the contents of the
3835 $dnslist_value variable are a list of all the addresses, separated by a
3838 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3839 to accept a host name and all its aliases if the forward lookup for any of
3840 them yielded the IP address of the incoming connection. Now it accepts only
3841 those names whose forward lookup yields the correct IP address. Any other
3842 names are discarded. This closes a loophole whereby a rogue DNS
3843 administrator could create reverse DNS records to break through a
3844 wildcarded host restriction in an ACL.
3846 37. If a user filter or a system filter that ran in a subprocess used any of
3847 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3848 the wrong values were passed to the pipe command ($thisaddress had the
3849 value of $0, $0 had the value of $1, etc). This bug was introduced by
3850 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3852 38. Improved the line breaking for long SMTP error messages from ACLs.
3853 Previously, if there was no break point between 40 and 75 characters, Exim
3854 left the rest of the message alone. Two changes have been made: (a) I've
3855 reduced the minimum length to 35 characters; (b) if it can't find a break
3856 point between 35 and 75 characters, it looks ahead and uses the first one
3857 that it finds. This may give the occasional overlong line, but at least the
3858 remaining text gets split now.
3860 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3861 file descriptors might be low, and that setting 1000 would always raise it.
3862 It turns out that in some environments, the limit is already over 1000 and
3863 that lowering it causes trouble. So now Exim takes care not to decrease it.
3865 40. When delivering a message, the value of $return_path is set to $sender_
3866 address at the start of routing (routers may change the value). By an
3867 oversight, this default was not being set up when an address was tested by
3868 -bt or -bv, which affected the outcome if any router or filter referred to
3871 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3872 log only when "message" or "log_message" are set. However, if one of the
3873 conditions was an address verification, or a call to a nested ACL, the
3874 messages generated by the underlying test were being passed through. This
3875 no longer happens. The underlying message is available in $acl_verify_
3876 message for both "message" and "log_message" expansions, so it can be
3877 passed through if needed.
3879 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3880 new expansion $bh_ to give the encoded byte string without charset
3881 translation. Translation happens only if iconv() is available; HAVE_ICONV
3882 indicates this at build time. HEADERS_CHARSET gives the charset to
3883 translate to; headers_charset can change it in the configuration, and
3884 "headers charset" can change it in an individual filter file.
3886 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3887 that creates RFC 2047 encoded "words" labels them as that charset instead
3888 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3889 expansion operator; (ii) when Exim creates a From: line for a local
3890 message; (iii) when a header line is rewritten to include a "phrase" part.
3892 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3893 buggy, causing it to skip the first lines of messages whose message ID
3894 ended in 'D'. This would not have bitten before Exim release 4.14, because
3895 message IDs were unlikely to end in 'D' before then. The effect was to have
3896 incorrect size information for certain domains.
3898 45. #include "config.h" was missing at the start of the crypt16.c module. This
3899 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3902 46. If there was a timeout during a "random" callout check, Exim treated it as
3903 a failure of the random address, and carried on sending RSET and the real
3904 address. If the delay was just some slowness somewhere, the response to the
3905 original RCPT would be taken as a response to RSET and so on, causing
3906 mayhem of various kinds.
3908 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3909 when I implemented it. It didn't allow for the fact that some option values
3910 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3911 the right test for positive values.
3913 48. Domain names in DNS records are case-independent. Exim always looks them up
3914 in lower case. Some resolvers return domain names in exactly the case they
3915 appear in the zone file, that is, they may contain uppercase letters. Not
3916 all resolvers do this - some return always lower case. Exim was treating a
3917 change of case by a resolver as a change of domain, similar to a widening
3918 of a domain abbreviation. This triggered its re-routing code and so it was
3919 trying to route what was effectively the same domain again. This normally
3920 caused routing to fail (because the router wouldn't handle the domain
3921 twice). Now Exim checks for this case specially, and just changes the
3922 casing of the domain that it ultimately uses when it transmits the message
3925 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3928 50. If a filter generated a file delivery with a non-absolute name (possible if
3929 no home directory exists for the router), the forbid_file option was not
3932 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3933 the existing equality matching.
3935 52. Exim was using ints instead of ino_t variables in some places where it was
3936 dealing with inode numbers.
3938 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3939 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3940 called TMPDIR, and if it finds it is different, it changes its value.
3942 54. The smtp_printf() function is now made available to local_scan() so
3943 additional output lines can be written before returning. There is also an
3944 smtp_fflush() function to enable the detection of a dropped connection.
3945 The variables smtp_input and smtp_batched_input are exported to
3948 55. Changed the default runtime configuration: the message "Unknown user"
3949 has been removed from the ACL, and instead placed on the localuser router,
3950 using the cannot_route_message feature. This means that any verification
3951 failures that generate their own messages won't get overridden. Similarly,
3952 the "Unrouteable address" message that was in the ACL for unverifiable
3953 relay addresses has also been removed.
3955 56. Added hosts_avoid_esmtp to the smtp transport.
3957 57. The exicyclog script was not checking for the esoteric option
3958 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3959 will work only if exicyclog is run under the appropriate euid.
3961 58. Following a discussion on the list, the rules by which Exim recognises line
3962 endings on incoming messages have been changed. The -dropcr and drop_cr
3963 options are now no-ops, retained only for backwards compatibility. The
3964 following line terminators are recognized: LF CRLF CR. However, special
3965 processing applies to CR:
3967 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3968 nor a local message in the state where . is a terminator.
3970 (ii) If a bare CR is encountered in a header line, an extra space is added
3971 after the line terminator so as not to end the header. The reasoning
3972 behind this is that bare CRs in header lines are most likely either
3973 to be mistakes, or people trying to play silly games.
3975 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3976 was being incorrectly given as 18 bytes larger than it should have been.
3977 This is a VOB (very old bug).
3979 60. This may never have affected anything current, but just in case it has:
3980 When the local host is found other than at the start of a list of hosts,
3981 the local host, those with the same MX, and any that follow, are discarded.
3982 When the list in question was part of a longer list of hosts, the following
3983 hosts (not currently being processed) were also being discarded. This no
3984 longer happens. I'm not sure if this situation could ever has previously
3987 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3990 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3991 header. This is something that is recommended in a new Internet Draft, and
3992 is something that is documented as being done by Sendmail. There are two
3993 possible values. For messages generated by the autoreply transport, Exim
3996 Auto-Submitted: auto-replied
3998 whereas for all other generated messages (e.g. bounces) it adds
4000 Auto-Submitted: auto-generated
4002 63. The "personal" condition in filters now includes a test for the
4003 Auto-Submitted: header. If it contains the string "auto-" the message it
4004 not considered personal.
4006 64. Added rcpt_include_affixes as a generic transport option.
4008 65. Added queue_only_override (default true).
4010 66. Added the syslog_duplication option.
4012 67. If what should have been the first header line of a message consisted of
4013 a space followed by a colon, Exim was mis-interpreting it as a header line.
4014 It isn't of course - it is syntactically invalid and should therefore be
4015 treated as the start of the message body. The misbehaviour could have
4016 caused a number of strange effects, including loss of data in subsequent
4017 header lines, and spool format errors.
4019 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4020 client host had authenticated. This control can now be exercised by an ACL
4021 for more flexibility.
4023 69. By default, callouts do not happen when testing with -bh. There is now a
4024 variant, -bhc, which does actually run the callout code, including
4025 consulting and updating the callout cache.
4027 70. Added support for saslauthd authentication, courtesy of Alexander
4030 71. If statvfs() failed on the spool or log directories while checking their
4031 size for availability, Exim confusingly gave the error "space shortage".
4032 Furthermore, in debugging mode it crashed with a floating point exception.
4033 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4034 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4035 serious problem, Exim now writes to the main and panic logs when this
4036 happens, with details of the failure. It then refuses to accept the
4037 incoming message, giving the message "spool directory problem" or "log
4038 directory problem" with a 421 code for SMTP messages.
4040 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4041 0, 1, and 2 exist, because some OS complain for execs without them (see
4042 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4043 descriptors. However, the code omitted to check that the open succeeded,
4044 causing mysterious errors if for some reason the permissions on /dev/null
4045 got screwed. Now Exim writes a message to the main and panic logs, and
4046 bombs out if it can't open /dev/null.
4048 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4049 interact so that it is all more flexible. It is supposed to remain
4050 backwards compatible. Also added extra_local_interfaces.
4052 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4053 to bomb out with an assertion failure - to the client this appears as a
4054 connection drop. This problem occurs in the part of the code that was taken
4055 from the Samba project. Fortunately, the assertion is in a very simple
4056 function, so I have fixed this by reproducing the function inline in the
4057 one place where it is called, and arranging for authentication to fail
4058 instead of killing the process with assert().
4060 75. The SPA client code was not working when the server requested OEM rather
4061 than Unicode encoding.
4063 76. Added code to make require_files with a specific uid setting more usable in
4064 the case where statting the file as root fails - usually a non-root-mounted
4065 NFS file system. When this happens and the failure is EACCES, Exim now
4066 forks a subprocess and does the per-uid checking as the relevant uid.
4068 77. Added process_log_path.
4070 78. If log_file_path was not explicitly set, a setting of check_log_space or
4071 check_log_inodes was ignored.
4073 79. If a space check for the spool or log partitions fails, the incident is now
4074 logged. Of course, in the latter case the data may get lost...
4076 80. Added the %p formatting code to string_format() so that it can be used to
4077 print addresses in debug_print(). Adjusted all the address printing in the
4078 debugging in store.c to use %p rather than %d.
4080 81. There was a concern that a line of code in smtp_in.c could overflow a
4081 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4082 initially expressed, the concern was not well-founded, because trailing
4083 spaces are removed early. However, if the trailing spaces were followed by
4084 a NULL, they did not get removed, so the overflow was possible. Two fixes
4087 (a) I re-wrote the offending code in a cleaner fashion.
4088 (b) If an incoming SMTP command contains a NULL character, it is rejected
4091 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4092 runs initgroups(), so that if the Exim user is in any additional groups,
4093 they will be used during message reception.
4099 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob