1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.383 2006/07/27 11:29:32 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
124 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
127 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
129 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
130 authenticated or an ident call has been made. Suppress the default
131 values for $authenticated_id and $authenticated_sender (but permit -oMai
132 and -oMas) when testing with -bh.
134 PH/24 Re-jigged the order of the tests in the default configuration so that the
135 tests for valid domains and recipients precede the DNS black list and CSA
136 tests, on the grounds that those ones are more expensive.
138 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
139 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
140 This bug exists in every version of Exim that I still have, right back to
147 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
148 other effects) broke the use of negated acl sub-conditions.
150 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
153 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
154 "Deny" causes Exim to reject the incoming connection with a 554 error.
155 Unfortunately, if there is a major crisis, such as a disk failure,
156 tcp-wrappers gives "deny", whereas what one would like would be some
157 kind of temporary error. A kludge has been added to help with this.
158 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
159 554 error is used if errno is still zero or contains ENOENT (which occurs
160 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
163 PH/03 Add -lutil to the default FreeBSD LIBS setting.
165 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
166 errors. Otherwise a message that provokes a temporary error (when other
167 messages do not) can cause a whole host to time out.
169 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
170 addresses were routed directly to files or pipes from a redirect router.
171 File deliveries just didn't batch; pipe deliveries might have suffered
174 PH/06 A failure to get a lock for a hints database would erroneously always say
175 "Failed to get write lock", even when it was really a read lock.
177 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
178 of 0600. This has been changed to use the value of the lockfile_mode
179 option (which defaults to 0600).
181 PH/08 Applied small patch from the Sieve maintainer.
183 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
184 folder from quota calculations, a direct delivery into this folder messed
185 up the contents of the maildirsize file. This was because the regex was
186 used only to exclude .Trash (or whatever) when the size of the mailbox
187 was calculated. There was no check that a delivery was happening into an
188 excluded directory. This bug has been fixed by ignoring all quota
189 processing for deliveries into excluded directories.
191 PH/10 Added the maildirfolder_create_regex option to appendfile.
197 PH/01 The code for finding all the local interface addresses on a FreeBSD
198 system running IPv6 was broken. This may well have applied to all BSD
199 systems, as well as to others that have similar system calls. The broken
200 code found IPv4 interfaces correctly, but gave incorrect values for the
201 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
202 that it would not match correctly against @[] and not recognize the IPv6
205 PH/02 The ipliteral router was not recognizing addresses of the form user@
206 [ipv6:....] because it didn't know about the "ipv6:" prefix.
208 PH/03 Added disable_ipv6.
210 PH/04 Changed $reply_address to use the raw form of the headers instead of the
211 decoded form, because it is most often used to construct To: headers
212 lines in autoreplies, and the decoded form may well be syntactically
213 invalid. However, $reply_address has leading white space removed, and all
214 newlines turned into spaces so that the autoreply transport does not
217 PH/05 If group was specified without a user on a router, and no group or user
218 was specified on a transport, the group from the router was ignored.
220 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
221 for visible compile-time settings that can be used to change these
222 numbers, for those that want even more. Backwards compatibility with old
223 spool files has been maintained. However, going back to a previous Exim
224 release will lost any variables that are in spool files.
226 PH/07 Two small changes when running in the test harness: increase delay when
227 passing a TCP/IP connection to a new process, in case the original
228 process has to generate a bounce, and remove special handling of
229 127.0.0.2 (sic), which is no longer necessary.
231 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
232 be the same on different OS.
234 PH/09 Moved a debug statement in filter processing to avoid a race problem when
237 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
238 whether --show-vars was specified or not
240 JJ/02 exipick: Added support for new ACL variable spool format introduced
243 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
244 syntactically invalid From: or Reply-to: line, and a filter used this to
245 generate an autoreply, and therefore failed to obtain an address for the
246 autoreply, Exim could try to deliver to a non-existent relative file
247 name, causing unrelated and misleading errors. What now happens is that
248 it logs this as a hard delivery error, but does not attempt to create a
251 PH/11 The exinext utility has a -C option for testing purposes, but although
252 the given file was scanned by exinext itself; it wasn't being passed on
253 when Exim was called.
255 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
256 an end-of-file indication when reading a command response.
258 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
259 compiled. In many other places in Exim, IPv6 addresses are always
260 recognized, so I have changed this. It also means that IPv4 domain
261 literals of the form [IPV4:n.n.n.n] are now always recognized.
263 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
264 used if the router is not running as root, for example, when verifying at
265 ACL time, or when using -bh. The debugging output from this situation was
266 non-existent - all you got was a failure to exec. I have made two
269 (a) Failures to set uid/gid, the current directory, or a process leader
270 in a subprocess such as that created by queryprogram now generate
271 suitable debugging ouput when -d is set.
273 (b) The queryprogram router detects when it is not running as root,
274 outputs suitable debugging information if -d is set, and then runs
275 the subprocess without attempting to change uid/gid.
277 PH/15 Minor change to Makefile for building test_host (undocumented testing
280 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
281 additional section of a DNS packet that returns MX or SRV records.
282 Instead, it always explicitly searches for A/AAAA records. This avoids
283 major problems that occur when a DNS server includes only records of one
284 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
285 fixed another bug: if SRV records were looked up and the corresponding
286 address records were *not* found in the additional section, the port
287 values from the SRV records were lost.
289 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
290 using the correct key (the original address) when searching the retry
291 rules in order to find which one to use for generating the retry hint.
293 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
294 adding the default one. Similarly, if it contains a Reply-To: header, the
295 errors_reply_to option, if set, is not used.
297 PH/19 When calculating a retry time, Exim used to measure the "time since
298 failure" by looking at the "first failed" field in the retry record. Now
299 it does not use this if it is later than than the arrival time of the
300 message. Instead it uses the arrival time. This makes for better
301 behaviour in cases where some deliveries succeed, thus re-setting the
302 "first failed" field. An example is a quota failure for a huge message
303 when small messages continue to be delivered. Without this change, the
304 "time since failure" will always be short, possible causing more frequent
305 delivery attempts for the huge message than are intended.
306 [Note: This change was subsequently modified - see PH/04 for 4.62.]
308 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
309 $1, $2, $3) because the numerical variables can be reset during some
310 expansion items (e.g. "match"), thereby losing the authentication data.
312 PH/21 Make -bV show the size of off_t variables so that the test suite can
313 decide whether to run tests for quotas > 2G.
315 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
316 mailbox_size, and mailbox_filecount in the appendfile transport. If a
317 filecount value is greater than 2G or if a quota value is greater than 2G
318 on a system where the size of off_t is not greater than 4, a panic error
321 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
322 never match. The debug and -bh output now contains an explicit error
323 message indicating a malformed IPv4 address or mask.
325 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
326 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
329 PH/25 Do not write to syslog when running in the test harness. The only
330 occasion when this arises is a failure to open the main or panic logs
331 (for which there is an explicit test).
333 PH/26 Added the /no_tell option to "control=freeze".
335 PH/27 If a host name lookup failed very early in a connection, for example, if
336 the IP address matched host_lookup and the reverse lookup yielded a name
337 that did not have a forward lookup, an error message of the form "no IP
338 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
339 could be logged. Now it outputs the IP address instead of "NULL".
341 PH/28 An enabling patch from MH: add new function child_open_exim2() which
342 allows the sender and the authenticated sender to be set when
343 submitting a message from within Exim. Since child_open_exim() is
344 documented for local_scan(), the new function should be too.
346 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
347 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
348 results in an empty string is now treated as unset.
350 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
352 PH/31 Added sender_verify_fail logging option.
354 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
355 needed by RFC 822 but not by RFC 2822 was commented out. I have now
356 tidied the source and removed it altogether.
358 PH/33 When a queue run was abandoned because the load average was too high, a
359 log line was always written; now it is written only if the queue_run log
360 selector is set. In addition, the log line for abandonment now contains
361 information about the queue run such as the pid. This is always present
362 in "start" and "stop" lines but was omitted from the "abandon" line.
364 PH/34 Omit spaces between a header name and the colon in the error message that
365 is given when verify = headers_syntax fails (if there are lots of them,
366 the message gets confusing).
368 PH/35 Change the default for dns_check_names_pattern to allow slashes within
369 names, as there are now some PTR records that contain slashes. This check
370 is only to protect against broken name servers that fall over on strange
371 characters, so the fact that it applies to all lookups doesn't matter.
373 PH/36 Now that the new test suite is complete, we can remove some of the
374 special code in Exim that was needed for the old test suite. For example,
375 sorting DNS records because real resolvers return them in an arbitrary
376 order. The new test suite's fake resolver always returns records in the
379 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
380 bounces) except when queue_only is set, to avoid logging races between
381 the different processes.
383 PH/38 Panic-die if .include specifies a non-absolute path.
385 PH/39 A tweak to the "H" retry rule from its user.
387 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
388 a label. They prevented compilation on older perls.
390 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
391 a warning to be raised on newish perls.
393 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
394 on queue. Changes to match documented behaviour of showing count of
395 messages matching specified criteria.
397 PH/40 Changed the default ident timeout from 30s to 5s.
399 PH/41 Added support for the use of login_cap features, on those BSD systems
400 that have them, for controlling the resources used by pipe deliveries.
402 PH/42 The content-scanning code uses fopen() to create files in which to put
403 message data. Previously it was not paying any attention to the mode of
404 the files. Exim runs with umask(0) because the rest of the code creates
405 files with open(), and sets the required mode explicitly. Thus, these
406 files were ending up world-writeable. This was not a big issue, because,
407 being within the spool directory, they were not world-accessible. I have
408 created a function called modefopen, which takes an additional mode
409 argument. It sets umask(777), creates the file, chmods it to the required
410 mode, then resets the umask. All the relevant calls to fopen() in the
411 content scanning code have been changed to use this function.
413 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
414 to 24 hours. This avoids potential overflow problems when processing G
415 and H retry rules. I suspect nobody ever tinkers with this value.
417 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
419 PH/45 When the plaintext authenticator is running as a client, the server's
420 challenges are checked to ensure they are valid base64 strings. By
421 default, the authentication attempt is cancelled if an invalid string is
422 received. Setting client_ignore_invalid_base64 true ignores these errors.
423 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
424 they are received. Thus, the responses can be made to depend on the
425 challenges. If an invalid string is ignored, an empty string is placed in
428 PH/46 Messages that are created by the autoreply transport now contains a
429 References: header, in accordance with RFCs 2822 and 3834.
431 PH/47 Added authenticated_sender_force to the smtp transport.
433 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
435 PH/49 Installed latest patch from the Sieve maintainer.
437 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
438 was also set, the appendfile transport was unnecessarily scanning a
439 directory of message files (e.g. for maildir delivery) to find the count
440 of files (along with the size), even though it did not need this
441 information. It now does the scan only if it needs to find either the
442 size of the count of files.
444 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
446 PH/52 Two bugs concerned with error handling when the smtp transport is
449 (i) Exim was not creating retry information for temporary errors given
450 for individual recipients after the DATA command when the smtp transport
451 was used in LMTP mode. This meant that they could be retried too
452 frequently, and not timed out correctly.
454 (ii) Exim was setting the flag that allows error details to be returned
455 for LMTP errors on RCPT commands, but not for LMTP errors for individual
456 recipients that were returned after the DATA command.
458 PH/53 This is related to PH/52, but is more general: for any failing address,
459 when detailed error information was permitted to be returned to the
460 sender, but the error was temporary, then after the final timeout, only
461 "retry timeout exceeded" was returned. Now it returns the full error as
462 well as "retry timeout exceeded".
464 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
465 do this, and (what is worse) MTAs that accept it.
467 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
468 will now be deprecated.
470 PH/56 New os.c-cygwin from the Cygwin maintainer.
472 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
473 formats (previously only available in exim formats via -bpr, -bpru,
474 and -bpra. Now also available in native and exiqgrep formats)
476 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
477 with very large, slow to parse queues
479 JJ/08 exipick: added ! as generic prefix to negate any criteria format
481 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
483 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
484 responses to authentication challenges, though it was showing the
485 challenges; (ii) I've removed the CR characters from the debug output for
488 PH/58 Allow for the insertion of a newline as well as a space when a string
489 is turned into more than one encoded-word during RFC 2047 encoding. The
490 Sieve code now uses this.
492 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
493 data_4xx, lost_connection, tls_required.
495 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
496 message was being sent as an SMTP response.
498 PH/61 Add -l and -k options to exicyclog.
500 PH/62 When verifying, if an address was redirected to one new address, so that
501 verification continued, and the new address failed or deferred after
502 having set something in $address_data, the value of $address_data was not
503 passed back to the ACL. This was different to the case when no
504 redirection occurred. The value is now passed back in both cases.
506 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
507 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
508 use login_cap.h, so on its own it isn't the distinguishing feature. The
509 new name refers directly to the setclassresources() function.
511 PH/65 Added configuration files for NetBSD3.
513 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
515 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
516 is preferred over IPv4.
518 PH/68 The bounce_return_message and bounce_return_body options were not being
519 honoured for bounces generated during the reception of non-SMTP messages.
520 In particular, this applied to messages rejected by the ACL. This bug has
521 been fixed. However, if bounce_return_message is true and bounce_return_
522 body is false, the headers that are returned for a non-SMTP message
523 include only those that have been read before the error was detected.
524 (In the case of an ACL rejection, they have all been read.)
526 PH/69 The HTML version of the specification is now built in a directory called
527 spec_html instead of spec.html, because the latter looks like a path with
528 a MIME-type, and this confuses some software.
530 PH/70 Catch two compiler warnings in sieve.c.
532 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
533 function verify_get_ident() calls ip_connect() to connect a socket, but
534 if the "connect()" function timed out, ip_connect() used to close the
535 socket. However, verify_get_ident() also closes the socket later, and in
536 between Exim writes to the log, which may get opened at this point. When
537 the socket was closed in ip_connect(), the log could get the same file
538 descriptor number as the socket. This naturally causes chaos. The fix is
539 not to close the socket in ip_connect(); the socket should be closed by
540 the function that creates it. There was only one place in the code where
541 this was missing, in the iplookup router, which I don't think anybody now
542 uses, but I've fixed it anyway.
544 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
545 well as to direct DNS lookups. Otherwise the handling of names in host
546 lists is inconsistent and therefore confusing.
552 PH/01 Two changes to the default runtime configuration:
554 (1) Move the checks for relay_from_hosts and authenticated clients from
555 after to before the (commented out) DNS black list checks.
557 (2) Add control=submission to the relay_from_hosts and authenticated
558 clients checks, on the grounds that messages accepted by these
559 statements are most likely to be submissions.
561 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
563 (1) Generate an error if the third argument for the ${prvs expansion is
566 (2) Treat a missing third argument of ${prvscheck as if it were an empty
569 (3) Reset the variables that are obtained from the first argument of
570 ${prvscheck and used in the second argument before leaving the code,
571 because their memory is reclaimed, so using them afterwards may do
574 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
575 one (it's much easier than Tom thought :-).
577 (5) Because of (4), we can now allow for the use of $prvscheck_result
578 inside the third argument.
580 PH/03 For some reason, the default setting of PATH when running a command from
581 a pipe transport was just "/usr/bin". I have changed it to
584 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
585 anything to be listed in the output from -bV.
587 PH/05 When a filter generated an autoreply, the entire To: header line was
588 quoted in the delivery log line, like this:
590 => >A.N.Other <ano@some.domain> <original@ddress> ...
592 This has been changed so that it extracts the operative address. There
593 may be more than one such address. If so, they are comma-separated, like
596 => >ano@some.domain,ona@other.domain <original@ddress> ...
598 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
599 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
600 not being looked up in the rDNS to get a host name, Exim was showing the
601 IP address twice in Received: lines, even though the IP addresses were
602 identical. For example:
604 Received: from [1.2.3.4] (helo=[1.2.3.4])
606 However, if the real host name was known, it was omitting the HELO data
607 if it matched the actual IP address. This has been tidied up so that it
608 doesn't show the same IP address twice.
610 PH/07 When both +timestamp and +memory debugging was on, the value given by
611 $tod_xxx expansions could be wrong, because the tod_stamp() function was
612 called by the debug printing, thereby overwriting the timestamp buffer.
613 Debugging no longer uses the tod_stamp() function when +timestamp is set.
615 PH/08 When the original message was included in an autoreply transport, it
616 always said "this is a copy of the message, including all the headers",
617 even if body_only or headers_only was set. It now gives an appropriate
620 PH/09 Applied a patch from the Sieve maintainer which:
622 o fixes some comments
623 o adds the (disabled) notify extension core
624 o adds some debug output for the result of if/elsif tests
625 o points to the current vacation draft in the documentation
626 and documents the missing references header update
630 o fixes a bug in processing the envelope test (when testing
631 multiple envelope elements, the last element determinted the
634 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
635 Electronic Mail") by including:
637 Auto-submitted: auto-generated
639 in the messages that it generates (bounce messages and others, such as
640 warnings). In the case of bounce messages for non-SMTP mesages, there was
641 also a typo: it was using "Auto_submitted" (underscore instead of
642 hyphen). Since every message generated by Exim is necessarily in response
643 to another message, thes have all been changed to:
645 Auto-Submitted: auto-replied
647 in accordance with these statements in the RFC:
649 The auto-replied keyword:
651 - SHOULD be used on messages sent in direct response to another
652 message by an automatic process,
654 - MUST NOT be used on manually-generated messages,
656 - MAY be used on Delivery Status Notifications (DSNs) and Message
657 Disposition Notifications (MDNs),
659 - MUST NOT be used on messages generated by automatic or periodic
660 processes, except for messages which are automatic responses to
663 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
664 to the default Received: header definition.
666 PH/12 Added log selector acl_warn_skipped (default on).
668 PH/13 After a successful wildlsearch lookup, discard the values of numeric
669 variables because (a) they are in the wrong storage pool and (b) even if
670 they were copied, it wouldn't work properly because of the caching.
672 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
673 checking when decoding. Apparently there are clients that generate
674 overlong encoded strings. Why am I not surprised?
676 PH/15 If the first argument of "${if match_address" was not empty, but did not
677 contain an "@" character, Exim crashed. Now it writes a panic log message
678 and treats the condition as false.
680 PH/16 In autoreply, treat an empty string for "once" the same as unset.
682 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
683 extension "envelope-auth". The code is finished and in agreement with
684 other implementations, but there is no documentation so far and in fact,
685 nobody wrote the draft yet. This extension is currently #undef'ed, thus
686 not changing the active code.
688 Print executed "if" and "elsif" statements when debugging is used. This
689 helps a great deal to understand what a filter does.
691 Document more things not specified clearly in RFC3028. I had all this
692 sorted out, when out of a sudden new issues came to my mind. Oops."
694 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
697 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
698 canonical form (as documented). However, after a host name lookup from
699 the IP address, check_host() was doing a simple string comparison with
700 addresses acquired from the DNS when checking that the found name did
701 have the original IP as one of its addresses. Since any found IPv6
702 addresses are likely to be in abbreviated form, the comparison could
703 fail. Luckily, there already exists a function for doing the comparison
704 by converting both addresses to binary, so now that is used instead of
707 PH/20 There was another similar case to PH/19, when a complete host name was
708 given in a host list; looking up its IP address could give an abbreviated
709 form, whereas the current host's name might or might not be abbreviated.
710 The same fix has been applied.
716 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
717 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
720 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
721 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
723 PH/03 Typo: missing ".o" in src/pcre/Makefile.
725 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
726 header line, restrict the check to what is listed in RFCs 2369 and 2929.
727 Also, for "Auto-Submitted", treat anything other than "no" as
728 non-personal, in accordance with RFC 3834. (Previously it treated
729 anything starting "auto-" as non-personal.)
731 TF/01 The control=submission/name=... option had a problem with syntax
732 errors if the name included a slash character. The /name= option
733 now slurps the rest of the string, so it can include any characters
734 but it must come last in the list of options (after /sender_retain
737 PH/05 Some modifications to the interface to the fake nameserver for the new
745 TK/01 Added the "success_on_redirect" address verification option. See
746 NewStuff for rationale and an example.
748 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
750 PH/02 Patch to exigrep to allow it to work on syslog lines.
752 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
753 fread() to skip over the body file's header line, because in Cygwin the
754 header line is locked and is inaccessible.
756 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
757 co-exist for some time) to make it clear that it is the Exim ID that is
758 referenced, not the Message-ID: header line.
760 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
761 string_format() function, because snprintf() does not exist on all
764 PH/06 The use of forbid_filter_existstest now also locks out the use of the
765 ${stat: expansion item.
767 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
768 protocol synchronization error", to keep the pedants happy.
770 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
771 well as for IRIX systems, when gcc is being used. See the host.c source
774 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
776 PH/10 Named domain lists were not working if used in a queue_smtp_domains
779 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
780 transport and to the smtp transport in LMTP mode.
782 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
784 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
785 run a filter in a subprocess. This could lead to confusion in subsequent
786 lookups in the parent process. There should also be a search_tidyup() at
787 the end of the subprocess.
789 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
790 only if the host matched helo_try_verify_hosts, which caused the
791 verification to occur when the EHLO/HELO command was issued. The ACL just
792 tested the remembered result. Now, if a previous verification attempt has
793 not happened, "verify = helo" does it there and then.
795 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
797 TK/03 Fix log output including CR from clamd.
799 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
800 exist provoked a memory error which could cause a segfault.
802 PH/15 Installed PCRE 6.2
804 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
806 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
807 of the problem. Specifically, suggested +O2 rather than +O1 for the
810 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
812 PH/20 If a delivery was routed to a non-standard port by means of an SRV
813 record, the port was not correctly logged when the outgoing_port log
814 selector was set (it logged the transort's default port).
816 PH/21 Added support for host-specific ports to manualroute, queryprogram,
817 fallback_hosts, and "hosts" in the smtp transport.
819 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
820 host errors such as "Connection refused".
822 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
823 authentication with radiusclient 0.4.9:
825 - Error returned from rc_read_config was caught wrongly
826 - Username/password not passed on to radius server due to wrong length.
828 The presumption is that some radiusclient API changes for 4.51/PH/17
829 were not taken care of correctly. The code is still untested by me (my
830 Linux distribution still has 0.3.2 of radiusclient), but it was
831 contributed by a Radius user.
833 PH/24 When doing a callout, the value of $domain wasn't set correctly when
834 expanding the "port" option of the smtp transport.
836 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
837 while reading a MIME header. Thanks to Tom Hughes for a patch.
839 PH/24 Include config.h inside local_scan.h so that configuration settings are
842 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
843 that in an ACL for RCPT (for example), you can examine exactly what was
846 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
847 commands, but it was not correctly comparing the address with the actual
848 client host address. Thus, it would show the EHLO address in Received:
849 header lines when this was not necessary.
851 PH/27 Added the % operator to ${eval:}.
853 PH/28 Exim tries to create and chdir to its spool directory when it starts;
854 it should be ignoring failures (because with -C, for example, it has lost
855 privilege). It wasn't ignoring creation failures other than "already
858 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
861 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
862 because an input line was too long, either on its own, or by virtue of
863 too many continuations, the temporary file was not being removed, and the
864 return code was incorrect.
866 PH/31 Missing "BOOL" in function definition in filtertest.c.
868 PH/32 Applied Sieve patches from the maintainer.
870 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
872 PH/33 Added "verify = not_blind".
874 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
875 Local/Makefile (with some defaults set). These are used in built scripts
876 such as exicyclog, but they have never been used in the exim_install
877 script (though there are many overriding facilities there). I have
878 arranged that the exim_install script now takes note of these two
881 PH/35 Installed configuration files for Dragonfly.
883 PH/36 When a locally submitted message by a trusted user did not contain a
884 From: header, and the sender address was obtained from -f or from an SMTP
885 MAIL command, and the trusted user did not use -F to supply a sender
886 name, $originator_name was incorrectly used when constructing a From:
887 header. Furthermore, $originator_name was used for submission mode
888 messages from external hosts without From: headers in a similar way,
889 which is clearly wrong.
891 PH/37 Added control=suppress_local_fixups.
893 PH/38 When log_selector = +received_sender was set, and the addition of the
894 sender made the log line's construction buffer exactly full, or one byte
895 less than full, an overflow happened when the terminating "\n" was
898 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
899 when the result of a list match is failure because a DNS lookup failed.
901 PH/40 RM_COMMAND is now used in the building process.
903 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
904 the "build-* directories that it finds.
906 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
907 address in a domain literal was a prefix of an interface address.
909 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
910 when verifying a sender address, unless rewrite_headers is false.
912 PH/44 Wrote a long comment about why errors_to addresses are verified as
913 recipients, not senders.
915 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
916 the ratelimit ACL was added.
918 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
920 PH/46 Added extra information about PostgreSQL errors to the error string.
922 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
923 avoiding the need to install special zones in a real server. This is
924 backwards compatible; if it can't find the fake resolver, it drops back.
925 Thus, both old and new test suites can be run.
927 TF/02 Added util/ratelimit.pl
929 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
930 clock is set back in time.
932 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
933 Candler <B.Candler@pobox.com>.
935 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
936 OK for addresses that are the result of redirections.
938 PH/48 A number of further additions for the benefit of the new test suite,
939 including a fake gethostbyname() that interfaces to the fake DNS resolver
942 TF/06 The fix for widen_domains has also been applied to qualify_single and
943 search_parents which are the other dnslookup options that can cause
946 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
949 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
951 TF/07 Exim produced the error message "an SRV record indicated no SMTP
952 service" if it encountered an MX record with an empty target hostname.
953 The message is now "an MX or SRV record indicated no SMTP service".
955 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
956 if the DNS of the sending site is misconfigured. This is quite a
957 common situation. This change restores the behaviour of treating a
958 helo verification defer as a failure.
960 PH/51 If self=fail was set on a router, the bounce message did not include the
961 actual error message.
967 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
969 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
970 command itself ended in error, the underlying message about the transport
971 filter timeout was being overwritten with the pipe command error. Now the
972 underlying error message should be appended to the second error message.
974 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
976 PH/02 Reduce the amount of output that "make" produces by default. Full output
977 can still be requested.
979 PH/03 The warning log line about a condition test deferring for a "warn" verb
980 was being output only once per connection, rather than after each
981 occurrence (because it was using the same function as for successful
982 "warn" verbs). This seems wrong, so I have changed it.
984 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
985 it should not have, which might have caused a crash in the right
986 circumstances, but probably never did.
988 PH/04 Installed a modified version of Tony Finch's patch to make submission
989 mode fix the return path as well as the Sender: header line, and to
990 add a /name= option so that you can make the user's friendly name appear
993 TF/03 Added the control = fakedefer ACL modifier.
995 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
996 Mark Lowes for thorough testing.
998 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1000 TK/03 Merged latest SRS patch from Miles Wilton.
1002 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1003 with the definition in sysexits.h (which is #included earlier).
1004 Fortunately, Exim does not actually use EX_OK. The code used to try to
1005 preserve the sysexits.h value, by assumimg that macro definitions were
1006 scanned for macro replacements. I have been disabused of this notion,
1007 so now the code just undefines EX_OK before #including unistd.h.
1009 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1010 in the smtp transport. When a block could not be written in a single
1011 write() function, the timeout was being re-applied to each part-write.
1012 This seems wrong - if the receiver was accepting one byte at a time it
1013 would take for ever. The timeout is now adjusted when this happens. It
1014 doesn't have to be particularly precise.
1016 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1017 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1019 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1020 without a domain if the check to <postmaster@domain> fails.
1022 SC/01 Eximstats: added -xls and the ability to specify output files
1023 (patch written by Frank Heydlauf).
1025 SC/02 Eximstats: use FileHandles for outputing results.
1027 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1029 SC/04 Eximstats: fixed display of large numbers with -nvr option
1031 SC/05 Eximstats: fixed merging of reports with empty tables.
1033 SC/06 Eximstats: added the -include_original_destination flag
1035 SC/07 Eximstats: removed tabs and trailing whitespace.
1037 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1039 TK/06 MBOX spool code: Add real "From " MBOX separator line
1040 so the .eml file is really in mbox format (even though
1041 most programs do not really care). Patch from Alex Miller.
1043 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1044 The latter is generated from $received_to and is only set if the
1045 message has one envelope recipient. SA can use these headers,
1046 obviously out-of-the-box. Patch from Alex Miller.
1048 PH/08 The ${def test on a variable was returning false if the variable's
1049 value was "0", contrary to what the specification has always said!
1050 The result should be true unless the variable is empty.
1052 PH/09 The syntax error of a character other than { following "${if
1053 def:variable_name" (after optional whitespace) was not being diagnosed.
1054 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1055 accidental colon was present, for example, could give incorrect results.
1057 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1058 result is used (not including appendfile, where other changes are about
1061 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1062 This involved changing a lot of size variables from int to off_t. It
1063 should work with maildirs and everything.
1065 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1066 spamd dying while we are connected to it.
1068 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1071 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1072 function for BDB 4.3.
1074 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1076 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1077 the sources for PCRE have been re-arranged and more files are now
1080 PH/15 The code I had for printing potentially long long variables in PH/11
1081 above was not the best (it lost precision). The length of off_t variables
1082 is now inspected at build time, and an appropriate printing format (%ld
1083 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1084 to be "long long int" or "long int". This is needed for the internal
1085 formatting function string_vformat().
1087 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1088 the configuration file to be ":syslog", then the script "guesses" where
1089 the logs files are, rather than using the compiled in default. In our
1090 case the guess is not the same as the compiled default, so the script
1091 suddenly stopped working when I started to use syslog. The patch checks
1092 to see if log_file_path is "". If so, it attempts to read it from exim
1093 with no configuration file to get the compiled in version, before it
1094 falls back to the previous guessing code."
1096 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1097 implementing BATV in an Exim configuration. See NewStuff for the gory
1100 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1101 Makefile that are specific to HP-UX.
1103 PH/18 If the "use_postmaster" option was set for a recipient callout together
1104 with the "random" option, the postmaster address was used as the MAIL
1105 FROM address for the random test, but not for the subsequent recipient
1106 test. It is now used for both.
1108 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1109 patch removes a few documentation additions to RFC 3028, because the
1110 latest draft now contains them. It adds the new en;ascii-case comparator
1111 and a new error check for 8bit text in MIME parts. Comparator and
1112 require names are now matched exactly. I enabled the subaddress
1113 extension, but it is not well tested yet (read: it works for me)."
1115 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1116 rework some of the code of TK/09 above to avoid the hardwired use of
1117 "%lld" and "long long". Replaced the call to snprintf() with a call to
1120 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1121 records point to non-existent hosts", "retry timeout exceeded", and
1122 "retry time not reached for any host after a long failure period".
1124 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1125 experimental DomainKeys support:
1127 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1128 (2) On an error such as an illegally used "control", the wrong name for
1129 the control was given.
1131 These problems did NOT occur unless DomainKeys support was compiled.
1133 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1135 PH/24 Added ${if match_ip condition.
1137 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1138 databases so that it will be absolutely obvious if a crash occurs in the
1139 DB library. This is a regular occurrence (often caused by mis-matched
1142 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1143 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1144 picked up on a user's system that detects such things. There doesn't seem
1145 to be a gcc warning option for this - only an attribute that has to be
1146 put on the function's prototype. It seems that in Fedora Core 4 they have
1147 set this on a number of new functions. No doubt there will be more in due
1150 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1151 specify a transport. However, if an address that was verified by such a
1152 router was the subject of a callout, Exim crashed because it tried to
1153 read the rcpt_include_affixes from the non-existent transport. Now it
1154 just assumes that the setting of that option is false. This bug was
1155 introduced by 4.51/PH/31.
1157 PH/28 Changed -d+all to exclude +memory, because that information is very
1158 rarely of interest, but it makes the output a lot bigger. People tend to
1159 do -d+all out of habit.
1161 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1162 code in os-type was giving problems when libc.so lives in lib64, like on
1165 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1166 aren't the modern standard, and it seems that some systems' include files
1167 don't always have them. Exim was already checking for some of the newer
1168 ones like T_AAAA, and defining it itself. I've added checks for all the
1169 record types that Exim uses.
1171 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1172 not automatically generating a new one, as it is supposed to. This
1173 prevented TLS from working. If the file did exist, but contained invalid
1174 data, a new version was generated, as expected. It was only the case of a
1175 non-existent file that was broken.
1177 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1178 with a change in libdomainkeys > 0.64.
1180 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1181 from DNS. If the selector record carries the flag, it now has
1182 precedence over the domain-wide flag.
1184 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1186 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1187 the use of a port name that isn't defined in /etc/services) occurred, the
1188 message was deferred as in a normal delivery, and thus remained on the
1189 spool, instead of being failed because of the mua_wrapper setting. This
1190 is now fixed, and I tidied up some of the mua_wrapper messages at the
1193 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1194 the messages in a hash of arrays rather than using individual hashes.
1195 This is a bit cleaner and results in dramatic memory savings, albeit
1196 at a slight CPU cost.
1198 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1199 as requested by Marc Sherman.
1201 SC/10 Eximstats: added histograms for user specified patterns as requested
1204 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1206 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1207 fopen() in the content-scanning modules that did not already have it.
1213 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1214 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1216 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1218 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1220 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1221 to test Sieve filters that use "vacation".
1223 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1224 that changes the way the GnuTLS parameters are stored in the cache file.
1225 The new format can be generated externally. For backward compatibility,
1226 if the data in the cache doesn't make sense, Exim assumes it has read an
1227 old-format file, and it generates new data and writes a new file. This
1228 means that you can't go back to an older release without removing the
1231 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1232 work if there are any delivery delays because "one_time" forces the
1233 parent to be marked "delivered", so its unseen clone is never tried
1234 again. For this reason, Exim now forbids the simultaneous setting of
1237 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1238 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1239 entry if you want to know the details. The fix, however, neglected to
1240 consider the case when local delivery batching is involved. The test for
1241 "previously delivered" was not happening when checking to see if an
1242 address could be batched with a previous (undelivered) one; under
1243 certain circumstances this could lead to multiple deliveries to the same
1246 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1247 in its include files, and this causes problems building Exim.
1249 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1250 header_syntax) but Exim was just ignoring anything given after a slash.
1251 In particular, this caused confusion with an attempt to use "verify =
1252 reverse_host_lookup/defer_ok". An error is now given when options are
1253 supplied for verify items that do not have them. (Maybe reverse_host_
1254 lookup should have a defer_ok option, but that's a different point.)
1256 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1257 defined by RFC 821) to 2048, because there were problems with some AUTH
1258 commands, and RFC 1869 says the size should be increased for extended
1259 SMTP commands that take arguments.
1261 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1264 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1265 "unknown" error; now it says that the functionality isn't in the binary.
1267 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1268 an address' error message when a string expansion fails (syntax or
1269 whatever). Otherwise the password may appear in the log. Following change
1270 PH/42 below, there is no longer a chance of it appearing in a bounce
1273 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1275 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1276 of its addresses was ever tried. (Bugzilla bug #2.)
1278 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1279 the result incorrectly in the debug output. (It correctly added a newline
1280 to what was transported.)
1282 TF/01 Added $received_time.
1284 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1285 commented out examples of how to interface to a virus scanner and to
1286 SpamAssassin. Also added commented examples of av_scanner and
1287 spamd_address settings.
1289 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1290 and controls are allowed in which ACLs. There were a couple of minor
1291 errors. Some of the entries in the conditions table (which is a table of
1292 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1293 negation of where the condition IS allowed.
1295 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1297 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1298 header file does not have a version number, so I've had to invent a new
1299 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1300 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1301 radiusclient), but it was contributed by a Radius user.
1303 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1304 files or directories, for OpenSSL.
1306 PH/19 When an Exim process that is running as root has to create an Exim log
1307 file, it does so in a subprocess that runs as exim:exim so as to get the
1308 ownership right at creation (otherwise, other Exim processes might see
1309 the file with the wrong ownership). There was no test for failure of this
1310 fork() call, which would lead to the process getting stuck as it waited
1311 for a non-existent subprocess. Forks do occasionally fail when resources
1312 run out. I reviewed all the other calls to fork(); they all seem to check
1315 PH/20 When checking for unexpected SMTP input at connect time (before writing
1316 the banner), Exim was not dealing correctly with a non-positive return
1317 from the read() function. If the client had disconnected by this time,
1318 the result was a log entry for a synchronization error with an empty
1319 string after "input=" when read() returned zero. If read() returned -1
1320 (an event I could not check), uninitialized data bytes were printed.
1321 There were reports of junk text (parts of files, etc) appearing after
1324 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1326 PH/22 Added support for macro redefinition, and (re)definition in between
1327 driver and ACL definitions.
1329 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1330 forgetting to use the resulting value; it was using the unexpanded value.
1332 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1333 hadn't been configured. The fix is from Juergen Kreileder, who
1334 understands it better than I do:
1336 "Here's what I see happening with three configured cyrus_sasl
1337 authenticators configured (plain, login, cram-md5):
1339 On startup auth_cyrus_sasl_init() gets called for each of these.
1340 This means three calls to sasl_listmech() without a specified mech_list.
1341 => SASL tests which mechs of all available mechs actually work
1342 => three warnings about OTP not working
1343 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1345 With the patch, sasl_listmech() also gets called three times. But now
1346 SASL's mech_list option is set to the server_mech specified in the the
1347 authenticator. Or in other words, the answer from sasl_listmech()
1348 gets limited to just the mech you're testing for (which is different
1350 => the return list contains just 'plain' or 'login', 'cram-md5' or
1351 nothing depending on the value of ob->server_mech.
1353 I've just tested the patch: Authentication still works fine,
1354 unavailable mechs specified in the exim configuration are still
1355 caught, and the auth.log warnings about OTP are gone."
1357 PH/25 When debugging is enabled, the contents of the command line are added
1358 to the debugging output, even when log_selector=+arguments is not
1361 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1362 answer is "GNU", and only if the return is "GNU/something" is the answer
1365 PH/27 $acl_verify_message is now set immediately after the failure of a
1366 verification in an ACL, and so is available in subsequent modifiers. In
1367 particular, the message can be preserved by coding like this:
1369 warn !verify = sender
1370 set acl_m0 = $acl_verify_message
1372 Previously, $acl_verify_message was set only while expanding "message"
1373 and "log_message" when a very denied access.
1375 PH/28 Modified OS/os.c-Linux with
1377 -#ifndef OS_LOAD_AVERAGE
1378 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1380 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1381 nomenclature these days.)
1383 PH/29 Installed patch from the Sieve maintainer that adds the options
1384 sieve_useraddress and sieve_subaddress to the redirect router.
1386 PH/30 In these circumstances:
1387 . Two addresses routed to the same list of hosts;
1388 . First host does not offer TLS;
1389 . First host accepts first address;
1390 . First host gives temporary error to second address;
1391 . Second host offers TLS and a TLS session is established;
1392 . Second host accepts second address.
1393 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1394 and peerdn, if requested) that were in fact used only for the second
1397 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1398 attention to any local part prefix or suffix that was matched by the
1399 router that accepted the address. It now behaves in the same way as it
1400 does for delivery: the affixes are removed from the local part unless
1401 rcpt_include_affixes is set on the transport.
1403 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1404 timeout during the DATA phase of an incoming message.
1406 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1407 applied a patch sanctioned by the Sieve maintainer.
1409 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1410 the uid or gid is negative. A case of a negative gid caused this to be
1411 noticed. The fix allows for either to be negative.
1413 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1414 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1417 PH/36 The change PH/12 above was broken. Fixed it.
1419 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1420 the grounds that routing the same address twice would always produce the
1421 same answer. This might have been true once, but it is certainly no
1422 longer true now. Routing a child address may depend on the previous
1423 routing that produced that child. Some complicated redirection strategies
1424 went wrong when messages had multiple recipients, and made Exim's
1425 behaviour dependent on the order in which the addresses were given.
1427 I have moved the duplicate checking until after the routing is complete.
1428 Exim scans the addresses that are assigned to local and remote
1429 transports, and removes any duplicates. This means that more work will be
1430 done, as duplicates will always all be routed, but duplicates are
1431 presumably rare, so I don't expect this is of any significance.
1433 For deliveries to pipes, files, and autoreplies, the duplicate checking
1434 still happens during the routing process, since they are not going to be
1437 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1438 It corrects a timeout issue with spamd. This is Ian's comment: "The
1439 background is that sometimes spamd either never reads data from a
1440 connection it has accepted, or it never writes response data. The exiscan
1441 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1442 blindly assumes that writes won't block so it may never time out."
1444 PH/39 Allow G after quota size as well as K and M.
1446 PH/40 The value set for $authenticated_id in an authenticator may not contain
1447 binary zeroes or newlines because the value is written to log lines and
1448 to spool files. There was no check on this. Now the value is run through
1449 the string_printing() function so that such characters are converted to
1450 printable escape sequences.
1452 PH/41 $message_linecount is a new variable that contains the total number of
1453 lines in the message. Compare $body_linecount, which is the count for the
1456 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1457 bounce and delay warning messages, except in certain special cases, which
1460 (a) An SMTP error message from a remote host;
1461 (b) A message specified in a :fail: redirection;
1462 (c) A message specified in a "fail" command in a system filter;
1463 (d) A message specified in a FAIL return from the queryprogram router;
1464 (e) A message specified by the cannot_route_message router option.
1466 In these cases only, Exim does include the error details in bounce and
1467 warning messages. There are also a few cases where bland messages such
1468 as "unrouteable address" or "local delivery error" are given.
1470 PH/43 $value is now also set for the "else" part of a ${run expansion.
1472 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1473 being worked on, but at least Exim now implements the latest version to
1476 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1477 process to complete was treated as a delivery failure, a timeout while
1478 writing the message to the pipe was logged, but erroneously treated as a
1479 successful delivery. Such timeouts include transport filter timeouts. For
1480 consistency with the overall process timeout, these timeouts are now
1481 treated as errors, giving rise to delivery failures by default. However,
1482 there is now a new Boolean option for the pipe transport called
1483 timeout_defer, which, if set TRUE, converts the failures into defers for
1484 both kinds of timeout. A transport filter timeout is now identified in
1487 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1488 systems where "make" and "gmake" are different, calling "gmake" at top
1489 level broke things. I've arranged for the value of $(MAKE) to be passed
1490 from the Makefile to this script so that it can call the same version of
1494 A note about Exim versions 4.44 and 4.50
1495 ----------------------------------------
1497 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1498 changes of various kinds. As a consequence, a big documentation update was
1499 needed. This delayed the release for rather longer than seemed good, especially
1500 in the light of a couple of (minor) security issues. Therefore, the changes
1501 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1502 release. So 4.44 and 4.50 are in effect two different branches that both start
1505 I have left the 4.50 change log unchanged; it contains all the changes since
1506 4.43. The change log for 4.44 is below; many of its items are identical to
1507 those for 4.50. This seems to be the most sensible way to preserve the
1508 historical information.
1514 1. Minor wording change to the doc/README.SIEVE file.
1516 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1517 computation of the current number of files was incorrect.
1519 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1520 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1523 4. Give more explanation in the error message when the command for a transport
1524 filter fails to execute.
1526 5. There are several places where Exim runs a non-Exim command in a
1527 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1528 was being done only for the command run by the queryprogram router. It is
1529 now done for all such subprocesses. The other cases are: ${run, transport
1530 filters, and the commands run by the lmtp and pipe transports.
1532 6. Added CONFIGURE_GROUP build-time option.
1534 7. Some older OS have a limit of 256 on the maximum number of file
1535 descriptors. Exim was using setrlimit() to set 1000 as a large value
1536 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1537 systems. I've change it so that if it can't get 1000, it tries for 256.
1539 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1540 was an oversight, and furthermore, ever since the addition of extra
1541 controls (e.g. 4.43/32), the checks on when to allow different forms of
1542 "control" were broken. There should now be diagnostics for all cases when a
1543 control that does not make sense is encountered.
1545 9. Added the /retain_sender option to "control=submission".
1547 10. $recipients is now available in the predata ACL (oversight).
1549 11. Tidy the search cache before the fork to do a delivery from a message
1550 received from the command line. Otherwise the child will trigger a lookup
1551 failure and thereby defer the delivery if it tries to use (for example) a
1552 cached ldap connection that the parent has called unbind on.
1554 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1555 of $address_data from the recipient verification was clobbered by the
1556 sender verification.
1558 13. The value of address_data from a sender verification is now available in
1559 $sender_address_data in subsequent conditions in the ACL statement.
1561 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1563 15. Added a new option "connect=<time>" to callout options, to set a different
1566 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1567 was its contents. (It was OK if the option was not defined at all.)
1569 17. A "Completed" log line is now written for messages that are removed from
1570 the spool by the -Mrm option.
1572 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1573 information about exactly what failed.
1575 19. Added -dd to debug only the daemon process.
1577 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1578 handles timeouts, both on the server side and network timeouts. Renamed the
1579 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1581 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1582 It is now set to "smtps".
1584 22. $host_address is now set to the target address during the checking of
1585 ignore_target_hosts.
1587 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1588 being passed; this would have caused $sender_host_name to have been used if
1589 matching the list had actually called for a host name (not very likely,
1590 since this list is usually IP addresses). A host name is now passed as
1593 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1594 code that specifies a non-restarting handler (typically sigaction() in
1595 modern systems) in an attempt to fix a rare and obscure crash bug.
1597 25. Narrowed the window for a race in the daemon that could cause it to ignore
1598 SIGCHLD signals. This is not a major problem, because they are used only to
1599 wake it up if nothing else does.
1601 26. A malformed maildirsize file could cause Exim to calculate negative values
1602 for the mailbox size or file count. Odd effects could occur as a result.
1603 The maildirsize information is now recalculated if the size or filecount
1606 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1607 support for a long time. Removed HAVE_SYS_VFS_H.
1609 28. Installed the latest version of exipick from John Jetmore.
1611 29. In an address list, if the pattern was not a regular expression, an empty
1612 subject address (from a bounce message) matched only if the pattern was an
1613 empty string. Non-empty patterns were not even tested. This was the wrong
1614 because it is perfectly reasonable to use an empty address as part of a
1615 database query. An empty address is now tested by patterns that are
1616 lookups. However, all the other forms of pattern expect the subject to
1617 contain a local part and a domain, and therefore, for them, an empty
1618 address still always fails if the pattern is not itself empty.
1620 30. Exim went into a mad DNS loop when attempting to do a callout where the
1621 host was specified on an smtp transport, and looking it up yielded more
1622 than one IP address.
1624 31. Re-factored the code for checking spool and log partition space into a
1625 function that finds that data and another that does the check. The former
1626 is then used to implement four new variables: $spool_space, $log_space,
1627 $spool_inodes, and $log_inodes.
1629 32. The RFC2047 encoding function was originally intended for short strings
1630 such as real names; it was not keeping to the 75-character limit for
1631 encoded words that the RFC imposes. It now respects the limit, and
1632 generates multiple encoded words if necessary. To be on the safe side, I
1633 have increased the buffer size for the ${rfc2047: expansion operator from
1636 33. It is now permitted to omit both strings after an "if" condition; if the
1637 condition is true, the result is "true". As before, when the second string
1638 is omitted, a false condition yields an empty string. This makes it less
1639 cumbersome to write custom ACL and router conditions.
1641 34. Failure to deliver a bounce message always caused it to be frozen, even if
1642 there was an errors_to setting on the router. The errors_to setting is now
1645 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1646 canonical form (fully expanded) before being placed in
1647 $sender_host_address.
1649 36. The table in the code that translates DNS record types into text (T_A to
1650 "A" for instance) was missing entries for NS and CNAME. It is just possible
1651 that this could have caused confusion if both these types were looked up
1652 for the same domain, because the text type is used as part of Exim's
1653 per-process caching. But the chance of anyone hitting this buglet seems
1656 37. The dnsdb lookup has been extended in a number of ways.
1658 (1) There is a new type, "zns", which walks up the domain tree until it
1659 finds some nameserver records. It should be used with care.
1661 (2) There is a new type, "mxh", which is like "mx" except that it returns
1662 just the host names, not the priorities.
1664 (3) It is now possible to give a list of domains (or IP addresses) to be
1665 looked up. The behaviour when one of the lookups defers can be
1666 controlled by a keyword.
1668 (4) It is now possible to specify the separator character for use when
1669 multiple records are returned.
1671 38. The dnslists ACL condition has been extended: it is now possible to supply
1672 a list of IP addresses and/or domains to be looked up in a particular DNS
1675 39. Added log_selector=+queue_time_overall.
1677 40. When running the queue in the test harness, wait just a tad after forking a
1678 delivery process, to get repeatability of debugging output.
1680 41. Include certificate and key file names in error message when GnuTLS fails
1681 to set them up, because the GnuTLS error message doesn't include the name
1682 of the failing file when there is a problem reading it.
1684 42. Allow both -bf and -bF in the same test run.
1686 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1688 44. The "Exiscan patch" is now merged into the mainline Exim source.
1690 45. Sometimes the final signoff response after QUIT could fail to get
1691 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1692 < 0 before doing a fflush(). This bug looks as though it goes back to the
1693 introduction of TLS in release 3.20, but "sometimes" must have been rare
1694 because the tests only now provoked it.
1696 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1697 (this can affect the format of dates).
1699 47. exim_tidydb, when checking for the continued existence of a message for
1700 which it has found a message-specific retry record, was not finding
1701 messages that were in split spool directories. Consequently, it was
1702 deleting retry records that should have stayed in existence.
1704 48. Steve fixed some bugs in eximstats.
1706 49. The SPA authentication driver was not abandoning authentication and moving
1707 on to the next authenticator when an expansion was forced to fail,
1708 contradicting the general specification for all authenticators. Instead it
1709 was generating a temporary error. It now behaves as specified.
1711 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1712 (the order specifies the preference for clients). The order is now AES256,
1713 AES128, 3DES, ARCFOUR128.
1715 51. Small patch to Sieve code - explicitly set From: when generating an
1718 52. Exim crashed if a remote delivery caused a very long error message to be
1719 recorded - for instance if somebody sent an entire SpamAssassin report back
1720 as a large number of 550 error lines. This bug was coincidentally fixed by
1721 increasing the size of one of Exim's internal buffers (big_buffer) that
1722 happened as part of the Exiscan merge. However, to be on the safe side, I
1723 have made the code more robust (and fixed the comments that describe what
1726 53. Now that there can be additional text after "Completed" in log lines (if
1727 the queue_time_overall log selector is set), a one-byte patch to exigrep
1728 was needed to allow it to recognize "Completed" as not the last thing in
1731 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1732 patch that reportedly fixes this has been added. I am not expert enough to
1733 create a test for it. This is what the patch creator wrote:
1735 "I found a little strange behaviour of ldap code when working with
1736 Windows 2003 AD Domain, where users was placed in more than one
1737 Organization Units. When I tried to give exim partial DN, the exit code
1738 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1739 But simultaneously result of request was absolutely normal ldap result,
1740 so I produce this patch..."
1742 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1743 REFERENCE, so I have modified the code to exclude the patch when that macro
1746 55. Some experimental protocols are using DNS PTR records for new purposes. The
1747 keys for these records are domain names, not reversed IP addresses. The
1748 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1749 leaves it alone. Component reversal etc. now happens only for IP addresses.
1752 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1754 57. Double the size of the debug message buffer (to 2048) so that more of very
1755 long debug lines gets shown.
1757 58. The exicyclog utility now does better if the number of log files to keep
1758 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1760 59. Two changes related to the smtp_active_hostname option:
1762 (1) $smtp_active_hostname is now available as a variable.
1763 (2) The default for smtp_banner uses $smtp_active_hostname instead
1764 of $primary_hostname.
1766 60. The host_aton() function is supposed to be passed a string that is known
1767 to be a valid IP address. However, in the case of IPv6 addresses, it was
1768 not checking this. This is a hostage to fortune. Exim now panics and dies
1769 if the condition is not met. A case was found where this could be provoked
1770 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1771 components; fortuitously, this particular loophole had already been fixed
1772 by change 4.50/55 above.
1774 If there are any other similar loopholes, the new check in host_aton()
1775 itself should stop them being exploited. The report I received stated that
1776 data on the command line could provoke the exploit when Exim was running as
1777 exim, but did not say which command line option was involved. All I could
1778 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1779 running as the user.
1782 61. There was a buffer overflow vulnerability in the SPA authentication code
1783 (which came originally from the Samba project). I have added a test to the
1784 spa_base64_to_bits() function which I hope fixes it.
1787 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1788 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1790 63. The daemon start-up calls getloadavg() while still root for those OS that
1791 need the first call to be done as root, but it missed one case: when
1792 deliver_queue_load_max is set with deliver_drop_privilege. This is
1793 necessary for the benefit of the queue runner, because there is no re-exec
1794 when deliver_drop_privilege is set.
1796 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1797 This has been fixed.
1799 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1800 was in use, was not putting the data itself into the right store pool;
1801 consequently, it could be overwritten for a subsequent message in the same
1802 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1805 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1807 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1808 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1809 the function were treating the return as a boolean value, which happened to
1810 work because 0=false and not-0=true, but is not correct code.
1812 68. The host_aton() function was not handling scoped IPv6 addresses (those
1813 with, for example, "%eth0" on the end) correctly.
1815 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1816 negated items (that is, ~something) in unsigned ints. Some compilers
1817 apparently mutter when there is no cast.
1819 70. If an address verification called from an ACL failed, and did not produce a
1820 user-specific message (i.e. there was only a "system" message), nothing was
1821 put in $acl_verify_message. In this situation, it now puts the system
1824 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1825 session; change 4.31/43 added the unwanted input to the log line - except
1826 that it did not do this in the start of session case. It now does.
1828 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1829 This is wrong and can cause the other end to generate a synchronization
1830 error if it is another Exim or anything else that does the synchronization
1831 check. A QUIT command is no longer sent after a timeout.
1833 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1834 during host lookups.
1836 74. The defer_ok option of callout verification was not working if it was used
1837 when verifying addresses in header lines, that is, for this case:
1839 verify = header_sender/callout=defer_ok
1841 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1842 those file descriptors could be used for SMTP connections. If anything
1843 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1844 could be sent to the SMTP client, causing chaos. The daemon now opens
1845 stdin, stdout, and stderr to /dev/null when it puts itself into the
1848 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1849 log by default. The user can override this with suitable Perl magic.
1851 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1852 the log message when discard triggers, was not working for the DATA ACL or
1853 for the non-SMTP ACL.
1855 78. Error message wording change in sieve.c.
1857 79. If smtp_accept_max_per_host was set, the number of connections could be
1858 restricted to fewer than expected, because the daemon was trying to set up
1859 a new connection before checking whether the processes handling previous
1860 connections had finished. The check for completed processes is now done
1861 earlier. On busy systems, this bug wouldn't be noticed because something
1862 else would have woken the daemon, and it would have reaped the completed
1865 80. If a message was submitted locally by a user whose login name contained one
1866 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1867 It caused a spool format error. I have fixed the spool reading code. A
1868 related problem was that the "from" clause in the Received: line became
1869 illegal because of the space(s). It is now covered by ${quote_local_part.
1871 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1874 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1876 83. Patch from Sieve maintainer to fix unterminated string problem in
1877 "vacation" handling.
1879 84. Some minor changes to the Linux configuration files to help with other
1880 OS variants using glibc.
1882 85. One more patch for Sieve to update vacation handling to latest spec.
1885 ----------------------------------------------------
1886 See the note above about the 4.44 and 4.50 releases.
1887 ----------------------------------------------------
1893 1. Change 4.43/35 introduced a bug that caused file counts to be
1894 incorrectly computed when quota_filecount was set in an appendfile
1897 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1898 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1901 3. Give more explanation in the error message when the command for a transport
1902 filter fails to execute.
1904 4. There are several places where Exim runs a non-Exim command in a
1905 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1906 was being done only for the command run by the queryprogram router. It is
1907 now done for all such subprocesses. The other cases are: ${run, transport
1908 filters, and the commands run by the lmtp and pipe transports.
1910 5. Some older OS have a limit of 256 on the maximum number of file
1911 descriptors. Exim was using setrlimit() to set 1000 as a large value
1912 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1913 systems. I've change it so that if it can't get 1000, it tries for 256.
1915 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1916 was an oversight, and furthermore, ever since the addition of extra
1917 controls (e.g. 4.43/32), the checks on when to allow different forms of
1918 "control" were broken. There should now be diagnostics for all cases when a
1919 control that does not make sense is encountered.
1921 7. $recipients is now available in the predata ACL (oversight).
1923 8. Tidy the search cache before the fork to do a delivery from a message
1924 received from the command line. Otherwise the child will trigger a lookup
1925 failure and thereby defer the delivery if it tries to use (for example) a
1926 cached ldap connection that the parent has called unbind on.
1928 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1929 of $address_data from the recipient verification was clobbered by the
1930 sender verification.
1932 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1933 was its contents. (It was OK if the option was not defined at all.)
1935 11. A "Completed" log line is now written for messages that are removed from
1936 the spool by the -Mrm option.
1938 12. $host_address is now set to the target address during the checking of
1939 ignore_target_hosts.
1941 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1942 being passed; this would have caused $sender_host_name to have been used if
1943 matching the list had actually called for a host name (not very likely,
1944 since this list is usually IP addresses). A host name is now passed as
1947 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1948 code that specifies a non-restarting handler (typically sigaction() in
1949 modern systems) in an attempt to fix a rare and obscure crash bug.
1951 15. Narrowed the window for a race in the daemon that could cause it to ignore
1952 SIGCHLD signals. This is not a major problem, because they are used only to
1953 wake it up if nothing else does.
1955 16. A malformed maildirsize file could cause Exim to calculate negative values
1956 for the mailbox size or file count. Odd effects could occur as a result.
1957 The maildirsize information is now recalculated if the size or filecount
1960 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1961 support for a long time. Removed HAVE_SYS_VFS_H.
1963 18. Updated exipick to current release from John Jetmore.
1965 19. Allow an empty sender to be matched against a lookup in an address list.
1966 Previously the only cases considered were a regular expression, or an
1969 20. Exim went into a mad DNS lookup loop when doing a callout where the
1970 host was specified on the transport, if the DNS lookup yielded more than
1973 21. The RFC2047 encoding function was originally intended for short strings
1974 such as real names; it was not keeping to the 75-character limit for
1975 encoded words that the RFC imposes. It now respects the limit, and
1976 generates multiple encoded words if necessary. To be on the safe side, I
1977 have increased the buffer size for the ${rfc2047: expansion operator from
1980 22. Failure to deliver a bounce message always caused it to be frozen, even if
1981 there was an errors_to setting on the router. The errors_to setting is now
1984 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1985 canonical form (fully expanded) before being placed in
1986 $sender_host_address.
1988 24. Updated eximstats to version 1.33
1990 25. Include certificate and key file names in error message when GnuTLS fails
1991 to set them up, because the GnuTLS error message doesn't include the name
1992 of the failing file when there is a problem reading it.
1994 26. Expand error message when OpenSSL has problems setting up cert/key files.
1997 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1998 (this can affect the format of dates).
2000 28. exim_tidydb, when checking for the continued existence of a message for
2001 which it has found a message-specific retry record, was not finding
2002 messages that were in split spool directories. Consequently, it was
2003 deleting retry records that should have stayed in existence.
2005 29. eximstats updated to version 1.35
2006 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2007 1.35 - bugfix such that pie charts by volume are generated correctly
2009 30. The SPA authentication driver was not abandoning authentication and moving
2010 on to the next authenticator when an expansion was forced to fail,
2011 contradicting the general specification for all authenticators. Instead it
2012 was generating a temporary error. It now behaves as specified.
2014 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2015 (the order specifies the preference for clients). The order is now AES256,
2016 AES128, 3DES, ARCFOUR128.
2018 31. Small patch to Sieve code - explicitly set From: when generating an
2021 32. Exim crashed if a remote delivery caused a very long error message to be
2022 recorded - for instance if somebody sent an entire SpamAssassin report back
2023 as a large number of 550 error lines. This bug was coincidentally fixed by
2024 increasing the size of one of Exim's internal buffers (big_buffer) that
2025 happened as part of the Exiscan merge. However, to be on the safe side, I
2026 have made the code more robust (and fixed the comments that describe what
2029 33. Some experimental protocols are using DNS PTR records for new purposes. The
2030 keys for these records are domain names, not reversed IP addresses. The
2031 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2032 leaves it alone. Component reversal etc. now happens only for IP addresses.
2035 34. The host_aton() function is supposed to be passed a string that is known
2036 to be a valid IP address. However, in the case of IPv6 addresses, it was
2037 not checking this. This is a hostage to fortune. Exim now panics and dies
2038 if the condition is not met. A case was found where this could be provoked
2039 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2040 components; fortuitously, this particular loophole had already been fixed
2041 by change 4.50/55 or 4.44/33 above.
2043 If there are any other similar loopholes, the new check in host_aton()
2044 itself should stop them being exploited. The report I received stated that
2045 data on the command line could provoke the exploit when Exim was running as
2046 exim, but did not say which command line option was involved. All I could
2047 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2048 running as the user.
2051 35. There was a buffer overflow vulnerability in the SPA authentication code
2052 (which came originally from the Samba project). I have added a test to the
2053 spa_base64_to_bits() function which I hope fixes it.
2056 36. The daemon start-up calls getloadavg() while still root for those OS that
2057 need the first call to be done as root, but it missed one case: when
2058 deliver_queue_load_max is set with deliver_drop_privilege. This is
2059 necessary for the benefit of the queue runner, because there is no re-exec
2060 when deliver_drop_privilege is set.
2062 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2063 was in use, was not putting the data itself into the right store pool;
2064 consequently, it could be overwritten for a subsequent message in the same
2065 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2068 38. Sometimes the final signoff response after QUIT could fail to get
2069 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2070 < 0 before doing a fflush(). This bug looks as though it goes back to the
2071 introduction of TLS in release 3.20, but "sometimes" must have been rare
2072 because the tests only now provoked it.
2078 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2079 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2080 Now it returns an integer. A number of calls were still expecting a T/F
2081 return. Fortuitously, in all cases, the tests worked in OK situations,
2082 which is the norm. However, things would have gone wrong on any write
2083 failures on the smtp file descriptor. This function is used when sending
2084 messages over SMTP and also when doing verify callouts.
2086 2. When Exim is called to do synchronous delivery of a locally submitted
2087 message (the -odf or -odi options), it no longer closes stderr before doing
2090 3. Implemented the mua_wrapper option.
2092 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2094 5. Implemented the functions header_remove(), header_testname(),
2095 header_add_at_position(), and receive_remove_recipient(), and exported them
2098 6. If an ACL "warn" statement specified the addition of headers, Exim already
2099 inserted X-ACL-Warn: at the start if there was no header name. However, it
2100 was not making this test for the second and subsequent header lines if
2101 there were newlines in the string. This meant that an invalid header could
2102 be inserted if Exim was badly configured.
2104 7. Allow an ACL "warn" statement to add header lines at the start or after all
2105 the Received: headers, as well as at the end.
2107 8. Added the rcpt_4xx retry error code.
2109 9. Added postmaster_mailfrom=xxx to callout verification option.
2111 10. Added mailfrom=xxxx to the callout verification option, for verify=
2114 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2115 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2117 12. Inserted some casts to stop certain compilers warning when using pointer
2118 differences as field lengths or precisions in printf-type calls (mostly
2119 affecting debugging statements).
2121 13. Added optional readline() support for -be (dynamically loaded).
2123 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2124 same clock tick as a message's arrival, so that its received time was the
2125 same as the "first fail" time on the retry record, and that message
2126 remained on the queue past the ultimate address timeout, every queue runner
2127 would try a delivery (because it was past the ultimate address timeout) but
2128 after another failure, the ultimate address timeout, which should have then
2129 bounced the address, did not kick in. This was a "< instead of <=" error;
2130 in most cases the first failure would have been in the next clock tick
2131 after the received time, and all would be well.
2133 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2134 being recognized when the domain list was tested by the match_domain
2135 condition in an expansion string.
2137 16. Added the ${str2b64: operator.
2139 17. Exim was always calling setrlimit() to set a large limit for the number of
2140 processes, without checking whether the existing limit was already
2141 adequate. (It did check for the limit on file descriptors.) Furthermore,
2142 errors from getrlimit() and setrlimit() were being ignored. Now they are
2143 logged to the main and panic logs, but Exim does carry on, to try to do its
2144 job under whatever limits there are.
2146 18. Imported PCRE 5.0.
2148 19. Trivial typo in log message " temporarily refused connection" (the leading
2151 20. If the log selector return_path_on_delivery was set and an address was
2152 redirected to /dev/null, the delivery process crashed because it assumed
2153 that a return path would always be set for a "successful" delivery. In this
2154 case, the whole delivery is bypassed as an optimization, and therefore no
2157 21. Internal re-arrangement: the function for sending a challenge and reading
2158 a response while authentication was assuming a zero-terminated challenge
2159 string. It's now changed to take a pointer and a length, to allow for
2160 binary data in such strings.
2162 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2164 23. Exim was not respecting finduser_retries when seeking the login of the
2165 uid under which it was called; it was always trying 10 times. (The default
2166 setting of finduser_retries is zero.) Also, it was sleeping after the final
2167 failure, which is pointless.
2169 24. Implemented tls_on_connect_ports.
2171 25. Implemented acl_smtp_predata.
2173 26. If the domain in control=submission is set empty, Exim assumes that the
2174 authenticated id is a complete email address when it generates From: or
2175 Sender: header lines.
2177 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2178 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2179 chown and chgrp in /bin and hostname in /usr/bin.
2181 28. Exim was keeping the "process log" file open after each use, just as it
2182 does for the main log. This opens the possibility of it remaining open for
2183 long periods when the USR1 signal hits a daemon. Occasional processlog
2184 errors were reported, that could have been caused by this. Anyway, it seems
2185 much more sensible not to leave this file open at all, so that is what now
2188 29. The long-running daemon process does not normally write to the log once it
2189 has entered its main loop, and it closes the log before doing so. This is
2190 so that log files can straightforwardly be renamed and moved. However,
2191 there are a couple of unusual error situations where the daemon does write
2192 log entries, and I had neglected to close the log afterwards.
2194 30. The text of an SMTP error response that was received during a remote
2195 delivery was being truncated at 512 bytes. This is too short for some of
2196 the long messages that one sometimes sees. I've increased the limit to
2199 31. It is now possible to make retry rules that apply only when a message has a
2200 specific sender, in particular, an empty sender.
2202 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2203 it possible to be selective about when SMTP synchronization is enforced.
2205 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2207 32. Implemented hosts_connection_nolog.
2209 33. Added an ACL for QUIT.
2211 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2214 35. Added mailbox_size and mailbox_filecount to appendfile.
2216 36. Added control = no_multiline_responses to ACLs.
2218 37. There was a bug in the logic of the code that waits for the clock to tick
2219 in the case where the clock went backwards by a substantial amount such
2220 that the microsecond fraction of "now" was more than the microsecond
2221 fraction of "then" (but the whole seconds number was less).
2223 38. Added support for the libradius Radius client library this is found on
2224 FreeBSD (previously only the radiusclient library was supported).
2230 1. When certain lookups returned multiple values in the form name=value, the
2231 quoting of the values was not always being done properly. Specifically:
2232 (a) If the value started with a double quote, but contained no whitespace,
2234 (b) If the value contained whitespace other than a space character (i.e.
2235 tabs or newlines or carriage returns) it was not quoted.
2236 This fix has been applied to the mysql and pgsql lookups by writing a
2237 separate quoting function and calling it from the lookup code. The fix
2238 should probably also be applied to nisplus, ibase and oracle lookups, but
2239 since I cannot test any of those, I have not disturbed their existing code.
2241 2. A hit in the callout cache for a specific address caused a log line with no
2242 reason for rejecting RCPT. Now it says "Previous (cached) callout
2243 verification failure".
2245 3. There was an off-by-one bug in the queryprogram router. An over-long
2246 return line was truncated at 256 instead of 255 characters, thereby
2247 overflowing its buffer with the terminating zero. As well as fixing this, I
2248 have increased the buffer size to 1024 (and made a note to document this).
2250 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2251 when Exim is waiting for an SMTP response from a remote server, Exim
2252 restarts its select() call on the socket, thereby resetting its timeout.
2253 This is not a problem when such interrupts are rare. Somebody set up a cron
2254 job to run exiwhat every 2 minutes, which is less than the normal select()
2255 timeout (5 or 10 minutes). This meant that the select() timeout never
2256 kicked in because it was always reset. I have fixed this by comparing the
2257 time when an interrupt arrives with the time at the start of the first call
2258 to select(). If more time than the timeout has elapsed, the interrupt is
2259 treated as a timeout.
2261 5. Some internal re-factoring in preparation for the addition of Sieve
2262 extensions (by MH). In particular, the "personal" test is moved to a
2263 separate function, and given an option for scanning Cc: and Bcc: (which is
2264 not set for Exim filters).
2266 6. When Exim created an email address using the login of the caller as the
2267 local part (e.g. when creating a From: or Sender: header line), it was not
2268 quoting the local part when it contained special characters such as @.
2270 7. Installed new OpenBSD configuration files.
2272 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2273 try to make them clearer.
2275 9. Callout options, other than the timeout value, were being ignored when
2276 verifying sender addresses in header lines. For example, when using
2278 verify = header_sender/callout=no_cache
2280 the cache was (incorrectly) being used.
2282 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2283 only the Cygwin environment.
2285 11. When return_path_on_delivery was set as a log selector, if different remote
2286 addresses in the same message used different return paths and parallel
2287 remote delivery occurred, the wrong values would sometimes be logged.
2288 (Whenever a remote delivery process finished, the return path value from
2289 the most recently started remote delivery process was logged.)
2291 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2292 lines when AUTH and/or TLS are in use. This is the "received protocol"
2293 field. Exim used to use "asmtp" for authenticated SMTP, without any
2294 indication (in the protocol name) for TLS use. Now it follows the RFC and
2295 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2296 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2297 names appear in log lines as well as in Received: header lines.
2299 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2300 extensions, and comparison tests, and to fix some bugs.
2302 14. Changes to the "personal" filter test:
2304 (1) The test was buggy in that it was just doing the equivalent of
2305 "contains" tests on header lines. For example, if a user's address was
2306 anne@some.where, the "personal" test would incorrectly be true for
2308 To: susanne@some.where
2310 This test is now done by extracting each address from the header in turn,
2311 and checking the entire address. Other tests that are part of "personal"
2312 are now done using regular expressions (for example, to check local parts
2313 of addresses in From: header lines).
2315 (2) The list of non-personal local parts in From: addresses has been
2316 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2317 taken from the Sieve specification recommendations.
2319 (3) If the message contains any header line starting with "List-" it is
2320 treated as non-personal.
2322 (4) The test for "circular" in the Subject: header line has been removed
2323 because it now seems ill-conceived.
2325 15. Minor typos in src/EDITME comments corrected.
2327 16. Installed latest exipick from John Jetmore.
2329 17. If headers_add on a router specified a text string that was too long for
2330 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2331 of string_sprintf() is now avoided.
2333 18. $message_body_size was not set (it was always zero) when running the DATA
2334 ACL and the local_scan() function.
2336 19. For the "mail" command in an Exim filter, no default was being set for
2337 the once_repeat time, causing a random time value to be used if "once" was
2338 specified. (If the value happened to be <= 0, no repeat happened.) The
2339 default is now 0s, meaning "never repeat". The "vacation" command was OK
2340 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2341 (I found it when inspecting the code).
2343 20. There is now an overall timeout for performing a callout verification. It
2344 defaults to 4 times the callout timeout, which applies to individual SMTP
2345 commands during the callout. The overall timeout applies when there is more
2346 than one host that can be tried. The timeout is checked before trying the
2347 next host. This prevents very long delays if there are a large number of
2348 hosts and all are timing out (e.g. when the network connections are timing
2349 out). The value of the overall timeout can be changed by specifying an
2350 additional sub-option for "callout", called "maxwait". For example:
2352 verify = sender/callout=5s,maxwait=20s
2354 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2355 to the end before writing, but this should make it even safer).
2357 22. Exim was forgetting that it had advertised PIPELINING for the second and
2358 subsequent messages on an SMTP connection. It was also not resetting its
2359 memory on STARTTLS and an internal HELO.
2361 23. When Exim logs an SMTP synchronization error within a session, it now
2362 records whether PIPELINING has been advertised or not.
2364 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2365 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2366 rather than long int.
2368 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2370 26. Added the never_mail option to autoreply.
2376 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2377 crash if the getsockname() call failed; this can happen if a connection is
2378 closed very soon after it is established. The problem was simply in the
2379 order in which certain operations were done, causing Exim to try to write
2380 to the SMTP stream before it had set up the file descriptor. The bug has
2381 been fixed by making things happen in the correct order.
2387 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2388 before the connection was closed, thus losing the rejection response.
2390 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2391 some early Solaris releases, but causes trouble in current releases where
2392 socklen_t is defined.
2394 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2397 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2400 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2401 that releasing the top of it at the end releases what was used for sub-
2402 expansions (unless the block got too big). However, discard this block if
2403 the first thing is a variable or header, so that we can use its block when
2404 it is dynamic (useful for very large $message_headers, for example).
2406 6. Lookups now cache *every* query, not just the most recent. A new, separate
2407 store pool is used for this. It can be recovered when all lookup caches are
2408 flushed. Lookups now release memory at the end of their result strings.
2409 This has involved some general refactoring of the lookup sources.
2411 7. Some code has been added to the store_xxx() functions to reduce the amount
2412 of flapping under certain conditions.
2414 8. log_incoming_interface used to affect only the <= reception log lines. Now
2415 it causes the local interface and port to be added to several more SMTP log
2416 lines, for example "SMTP connection from", and rejection lines.
2418 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2420 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2422 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2423 could be overwritten at the end of the current message (or the start of a
2424 new message if it was set in a HELO ACL). The value is now preserved for
2425 the duration of the SMTP connection.
2427 12. If a transport had a headers_rewrite setting, and a matching header line
2428 contained an unqualified address, that address was qualified, even if it
2429 did not match any rewriting rules. The underlying bug was that the values
2430 of the flags that permit the existence of unqualified sender and recipient
2431 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2432 non-local messages, and by -bnq for local messages) were not being
2433 preserved with the message after it was received.
2435 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2436 "next input=" as part of the text comprising the host identity instead of
2437 the correct text. The code was using the same buffer for two different
2438 strings. However, depending on which order the printing function evaluated
2439 its arguments, the bug did not always show up. Under Linux, for example, my
2440 test suite worked just fine.
2442 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2443 doesn't work with some older versions of Perl. It has been changed to "my",
2444 which in any case is probably the better facility to use.
2446 15. A really picky compiler found some instances of statements for creating
2447 error messages that either had too many or two few arguments for the format
2450 16. The size of the buffer for calls to the DNS resolver has been increased
2451 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2452 for addresses that have a lot of PTR records. This alleviates a problem; it
2453 does not fully solve it.
2455 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2456 buffer now truncates the list and logs the incident, which is the same
2457 action as happens when Exim is looking up a host name and its aliases.
2458 Previously in this situation something unpredictable would happen;
2459 sometimes it was "internal error: store_reset failed".
2461 18. If a server dropped the connection unexpectedly when an Exim client was
2462 using GnuTLS and trying to read a response, the client delivery process
2463 crashed while trying to generate an error log message.
2465 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2466 string, for example:
2468 warn message = H1: something\nH2: something
2470 the text was added as a single header line from Exim's point of view
2471 though it ended up OK in the delivered message. However, searching for the
2472 second and subsequent header lines using $h_h2: did not work. This has been
2473 fixed. Similarly, if a system filter added multiple headers in this way,
2474 the routers could not see them.
2476 20. Expanded the error message when iplsearch is called with an invalid key to
2477 suggest using net-iplsearch in a host list.
2479 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2480 ACLs are no longer actually imposed (and a message to that effect is
2483 22. If a "gecos" field in a passwd entry contained escaped characters, in
2484 particular, if it contained a \" sequence, Exim got it wrong when building
2485 a From: or a Sender: header from that name. A second bug also caused
2486 incorrect handling when an unquoted " was present following a character
2487 that needed quoting.
2489 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2490 was not being matched caselessly.
2492 24. Arranged for all hyphens in the exim.8 source to be escaped with
2495 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2496 properly. Recipient callout cache records were still being keyed to include
2497 the sender, even when use_sender was set false. This led to far more
2498 callouts that were necessary. The sender is no longer included in the key
2499 when use_sender is false.
2501 26. Added "control = submission" modifier to ACLs.
2503 27. Added the ${base62d: operator to decode base 62 numbers.
2505 28. dnsdb lookups can now access SRV records.
2507 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2508 the configuration file.
2510 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2511 (-v) mode. This makes the output for a verbose queue run more intelligible.
2513 31. Added a use_postmaster feature to recipient callouts.
2515 32. Added the $body_zerocount variable, containing the number of binary zero
2516 bytes in the message body.
2518 33. The time of last modification of the "new" subdirectory is now used as the
2519 "mailbox time last read" when there is a quota error for a maildir
2522 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2524 35. Added +ignore_unknown as a special item in host lists.
2526 36. Code for decoding IPv6 addresses in host lists is now included, even if
2527 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2528 address was recognized as an IP address, but was then not correctly decoded
2529 into binary, causing unexpected and incorrect effects when compared with
2536 1. Very minor rewording of debugging text in manualroute to say "list of
2537 hosts" instead of "hostlist".
2539 2. If verify=header_syntax was set, and a header line with an unqualified
2540 address (no domain) and a large number of spaces between the end of the
2541 name and the colon was received, the reception process suffered a buffer
2542 overflow, and (when I tested it) crashed. This was caused by some obsolete
2543 code that should have been removed. The fix is to remove it!
2545 3. When running in the test harness, delay a bit after writing a bounce
2546 message to get a bit more predictability in the log output.
2548 4. Added a call to search_tidyup() just before forking a reception process. In
2549 theory, someone could use a lookup in the expansion of smtp_accept_max_
2550 per_host which, without the tidyup, could leave open a database connection.
2552 5. Added the variables $recipient_data and $sender_data which get set from a
2553 lookup success in an ACL "recipients" or "senders" condition, or a router
2554 "senders" option, similar to $domain_data and $local_part_data.
2556 6. Moved the writing of debug_print from before to after the "senders" test
2559 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2560 problems for message scanning, either using a data ACL, or using
2561 local_scan() because the Received: header was not generated till after they
2562 were called (in order to set the time as the time of reception completion).
2563 I have revised the way this works. The header is now generated after the
2564 body is received, but before the ACL or local_scan() are called. After they
2565 are run, the timestamp in the header is updated.
2571 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2572 before starting a queue runner without re-exec. This happened only when
2573 deliver_drop_privilege was set or when the Exim user was set to root. The
2574 effect of the bug was that timeouts during subsequent deliveries caused
2575 crashes instead of being properly handled. The handler is now left at its
2576 default (and expected) setting.
2578 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2579 message, again when deliver_drop_privilege is set or Exim is run as root.
2580 The bug described in (1) was not present in this case, but the tidying up
2581 of the other signals was missing. I have made the two cases consistent.
2583 3. The ignore_target_hosts setting on a manualroute router was being ignored
2584 for hosts that were looked up using the /MX notation.
2586 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2589 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2590 operated on the sender address. After changing the $sender_address to <>
2591 for the sender address verify, Exim was re-instated it as the original
2592 (before rewriting) address, but remembering that it had rewritten it, so it
2593 wasn't rewriting it again. This bug also had the effect of breaking the
2594 sender address verification caching when the sender address was rewritten.
2596 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2597 This has been changed so that if the ip literal address matches
2598 ignore_target_hosts, the router declines.
2600 7. Added expansion conditions match_domain, match_address, and match_local_
2601 part (NOT match_host).
2603 8. The placeholder for the Received: header didn't have a length field set.
2605 9. Added code to Exim itself and to exim_lock to test for a specific race
2606 condition that could lead to file corruption when using MBX delivery. The
2607 issue is with the lockfile that is created in /tmp. If this file is removed
2608 after a process has opened it but before that process has acquired a lock,
2609 there is the potential for a second process to recreate the file and also
2610 acquire a lock. This could lead to two Exim processes writing to the file
2611 at the same time. The added code performs the same test as UW imapd; it
2612 checks after acquiring the lock that its file descriptor still refers to
2613 the same named file.
2615 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2616 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2619 11. Added the smtp_active_hostname option. If used, this will typically be made
2620 to depend on the incoming interface address. Because $interface_address is
2621 not set up until the daemon has forked a reception process, error responses
2622 that can happen earlier (such as "too many connections") no longer contain
2625 12. If an expansion in a condition on a "warn" statement fails because a lookup
2626 defers, the "warn" statement is abandoned, and the next ACL statement is
2627 processed. Previously this caused the whole ACL to be aborted.
2629 13. Added the iplsearch lookup type.
2631 14. Added ident_timeout as a log selector.
2633 15. Added tls_certificate_verified as a log selector.
2635 16. Added a global option tls_require_ciphers (compare the smtp transport
2636 option of the same name). This controls incoming TLS connections.
2638 17. I finally figured out how to make tls_require_ciphers do a similar thing
2639 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2640 before starting the TLS session.
2642 18. Tabs are now shown as \t in -bP output.
2644 19. If the log selector return_path_on_delivery was set, Exim crashed when
2645 bouncing a message because it had too many Received: header lines.
2647 20. If two routers both had headers_remove settings, and the first one included
2648 a superfluous trailing colon, the final name in the first list and the
2649 first name in the second list were incorrectly joined into one item (with a
2650 colon in the middle).
2656 1. Added -C and -D options to the exinext utility, mainly to make it easier
2657 to include in the automated testing, but these could be helpful when
2658 multiple configurations are in use.
2660 2. The exinext utility was not formatting the output nicely when there was
2661 an alternate port involved in the retry record key, nor when there was a
2662 message id as well (for retries that were specific to a specific message
2663 and a specific host). It was also confused by IPv6 addresses, because of
2664 the additional colons they contain. I have fixed the IPv4 problem, and
2665 patched it up to do a reasonable job for IPv6.
2667 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2668 delivery, the log line now contains "pipelined" if PIPELINING was used.
2670 4. An SMTP transport process used to panic and die if the bind() call to set
2671 an explicit outgoing interface failed. This has been changed; it is now
2672 treated in the same way as a connect() failure.
2674 5. A reference to $sender_host_name in the part of a conditional expansion
2675 that was being skipped was still causing a DNS lookup. This no longer
2678 6. The def: expansion condition was not recognizing references to header lines
2679 that used bh_ and bheader_.
2681 7. Added the _cache feature to named lists.
2683 8. The code for checking quota_filecount in the appendfile transport was
2684 allowing one more file than it should have been.
2686 9. For compatibility with Sendmail, the command line option
2694 and sets the incoming protocol and host name (for trusted callers). The
2695 host name and its colon can be omitted when only the protocol is to be set.
2696 Note the Exim already has two private options, -pd and -ps, that refer to
2697 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2698 "s", but I don't think that's a major issue.
2700 10. A number of refactoring changes to the code, none of which should affect
2703 (a) The number of logging options was getting close to filling up the
2704 32-bit word that was used as a bit map. I have split them into two classes:
2705 those that are passed in the argument to log_write(), and those that are
2706 only ever tested independently outside of that function. These are now in
2707 separate 32-bit words, so there is plenty of room for expansion again.
2708 There is no change in the user interface or the logging behaviour.
2710 (b) When building, for example, log lines, the code previously used a
2711 macro that called string_cat() twice, in order to add two strings. This is
2712 not really sufficiently general. Furthermore, there was one instance where
2713 it was actually wrong because one of the argument was used twice, and in
2714 one call a function was used. (As it happened, calling the function twice
2715 did not affect the overall behaviour.) The macro has been replaced by a
2716 function that can join an arbitrary number of extra strings onto a growing
2719 (c) The code for expansion conditions now uses a table and a binary chop
2720 instead of a serial search (which was left over from when there were very
2721 few conditions). Also, it now recognizes conditions like "pam" even when
2722 the relevant support is not compiled in: a suitably worded error message is
2723 given if an attempt is made to use such a condition.
2725 11. Added ${time_interval:xxxxx}.
2727 12. A bug was causing one of the ddress fields not to be passed back correctly
2728 from remote delivery subprocesses. The field in question was not being
2729 subsequently used, so this caused to problems in practice.
2731 13. Added new log selectors queue_time and deliver_time.
2733 14. Might have fixed a bug in maildirsizefile handling that threw up
2734 "unexpected character" debug warnings, and recalculated the data
2735 unnecessarily. In any case, I expanded the warning message to give more
2738 15. Added the message "Restricted characters in address" to the statements in
2739 the default ACL that block characters like @ and % in local parts.
2741 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2742 Three changes have been made:
2744 (a) There was a serious bug; a negative response to MAIL caused the whole
2745 recipient domain to be cached as invalid, thereby blocking all messages
2746 to all local parts at the same domain, from all senders. This bug has
2747 been fixed. The domain is no longer cached after a negative response to
2748 MAIL if the sender used is not empty.
2750 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2753 (c) A new callout option, "use_sender" has been added for people who want
2754 the modified behaviour.
2760 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2763 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2764 indeed breaks things for older releases.
2766 3. Added additional logging to the case where there is a problem reading data
2767 from a filter that is running in a subprocess using a pipe, in order to
2768 try to track down a specific problem.
2770 4. Testing facility fudge: when running in the test harness and attempting
2771 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2772 getting "No route to host". Convert this to a timeout.
2774 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2777 6. Some OS don't have socklen_t but use size_t instead. This affects the
2778 fifth argument of getsockopt() amongst other things. This is now
2779 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2780 can be set for individual OS. I have set it for SunOS5, OSF1, and
2781 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2782 some earlier ones do not.
2784 7. Change 4.30/15 was not doing the test caselessly.
2786 8. The standard form for an IPv6 address literal was being rejected by address
2787 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2788 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2789 this, as well as the form without the "IPv6" on the front (but only when
2790 address literals are enabled, of course).
2792 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2794 10. Exim crashed if a message with an empty sender address specified by -f
2795 encountered a router with an errors_to setting. This could be provoked only
2796 by a command such as
2800 where an empty string was supplied; "<>" did not hit this bug.
2802 11. Installed PCRE release 4.5.
2804 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2805 remained set. It is now erased.
2807 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2808 times from message ids (which are base 36 rather than the normal 62).
2810 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2811 were being counted as actual protocol errors, and logged if the log
2812 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2813 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2814 and DATA following a set of rejected RCPTs do not count as protocol errors.
2815 In other words, Exim assumes they were pipelined, though this may not
2816 actually be the case. Of course, in all cases the client gets an
2817 appropriate error code.
2819 15. If a lookup fails in an ACL condition, a message about the failure may
2820 be available; it is used if testing the ACL cannot continue, because most
2821 such messages specify what the cause of the deferral is. However, some
2822 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2823 that caused an old message to be retained and used if a later statement
2824 caused a defer, replacing the real cause of the deferral.
2826 16. If an IP address had so many PTR records that the DNS lookup buffer
2827 was not large enough to hold them, Exim could crash while trying to process
2828 the truncated data. It now detects and logs this case.
2830 17. Further to 4.21/58, another change has been made: if (and only if) the
2831 first line of a message (the first header line) ends with CRLF, a bare LF
2832 in a subsequent header line has a space inserted after it, so as not to
2833 terminate the header.
2835 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2836 unnecessarily, used atoi() instead of strtol(), and didn't check the
2837 termination when getting file sizes from file names by regex.
2839 19. Completely re-implemented the support for maildirsize files, in the light
2840 of a number of problems with the previous contributed implementation
2841 (4.30/29). In particular:
2843 . If the quota is zero, the maildirsize file is maintained, but no quota is
2846 . If the maildir directory does not exist, it is created before any attempt
2847 to write a maildirsize file.
2849 . The quota value in the file is just a cache; if the quota is changed in
2850 the transport, the new value overrides.
2852 . A regular expression is available for excluding directories from the
2855 20. The autoreply transport checks the characters in options that define the
2856 message's headers; it allows continued headers, but it was checking with
2857 isspace() after an embedded newline instead of explicitly looking for a
2860 21. If all the "regular" hosts to which an address was routed had passed their
2861 expiry times, and had not reached their retry times, the address was
2862 bounced, even if fallback hosts were defined. Now Exim should go on to try
2865 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2866 equivalent code in the SMTP transport. Some hosts send humungous responses
2867 to HELO/EHLO, more than 1024 it seems.
2869 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2870 gives compiler warnings in some environments. I've now done it "properly",
2873 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2874 (because of problems with the built-in one) was declared to return uschar *
2875 instead of char *, causing compiler failure.
2877 25. Fixed a file descriptor leak when processing alias/forward files.
2879 26. Fixed a minor format string issue in dbfn.c.
2881 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2883 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2884 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2887 29. When a sender address is verified, it is cached, to save repeating the test
2888 when there is more than one recipient in a message. However, when the
2889 verification involves a callout, it is possible for different callout
2890 options to be set for different recipients. It is too complicated to keep
2891 track of this in the cache, so now Exim always runs a verification when a
2892 callout is required, relying on the callout cache for the optimization.
2893 The overhead is duplication of the address routing, but this should not be
2896 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2897 to be verified with callout=postmaster, and the main callout worked but the
2898 postmaster check failed, the verification correctly failed. However, if a
2899 subsequent RCPT command asked for sender verification *without* the
2900 postmaster check, incorrect caching caused this verification also to fail,
2903 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2904 it was not caching the DNS options (qualify_single, search_parents) that
2905 were used when the lookup failed. A subsequent lookup with different
2906 options therefore always gave the same answer, though there were cases
2907 where it should not have. (Example: a "domains = !$mx_any" option on a
2908 dnslookup router: the "domains" option is always processed without any
2909 widening, but the router might have qualify_single set.) Now Exim uses the
2910 cached value only when the same options are set.
2912 32. Added John Jetmore's "exipick" utility to the distribution.
2914 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2915 than a timeout (e.g. a certificate is required, and is not provided), an
2916 Exim server now closes the connection immediately. Previously it waited for
2917 the client to close - but if the client is SSL, it seems that they each
2918 wait for each other, leading to a delay before one of them times out.
2920 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2921 maintained 0.8.x compatibility because I don't think many are using it, and
2922 it is clearly obsolete.
2924 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2927 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2928 client certificate was expired. A simple patch fixes this, though I don't
2929 understand the full logic of why the verify callback is called multiple
2932 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2933 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2934 which causes problems with some clients (such as the Certicom SSL Plus
2935 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2936 disables the coutermeasure allowing Eudora to connect."
2938 38. Exim was not checking that a write() to a log file succeeded. This could
2939 lead to Bad Things if a log got too big, in particular if it hit a file
2940 size limit. Exim now panics and dies if it cannot write to a log file, just
2941 as it does if it cannot open a log file.
2943 39. Modified OS/Makefile-Linux so that it now contains
2945 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2947 The two -D definitions ensure that Exim is compiled with large file
2948 support, which makes it possible to handle log files that are bigger than
2951 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2952 instance) a domain was checked against a named list that involved a lookup,
2953 causing $domain_data to be set, then another domain was checked against the
2954 same list, then the first domain was re-checked, the value of $domain_data
2955 after the final check could be wrong. In particular, if the second check
2956 failed, it could be set empty. This bug probably also applied to
2959 41. The strip_trailing_dot option was not being applied to the address given
2960 with the -f command-line option.
2962 42. The code for reading a message's header from the spool was incrementing
2963 $received_count, but never initializing it. This meant that the value was
2964 incorrect (doubled) while delivering a message in the same process in which
2965 it was received. In the most common configuration of Exim, this never
2966 happens - a fresh exec is done - but it can happen when
2967 deliver_drop_privilege is set.
2969 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2970 it now includes up to 150 characters of the unexpected data in the log
2973 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2974 and building data strings. The size of both of these buffers was 10 000
2975 bytes - far larger than anybody would *ever* want, thought I. Needless to
2976 say, somebody hit the limit. I have increased the maximum line length to
2977 20 000 and the maximum data length of concatenated lines to 100 000. I have
2978 also fixed two bugs, because there was no checking on these buffers. Tsk,
2979 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2980 buffer is too small.
2982 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2983 lsearch lookups. Now it does.
2985 46. When parsing a route_list item in a manualroute router, a fixed-length
2986 buffer was used for the list of hosts. I made this 1024 bytes long,
2987 thinking that nobody would ever have a list of hosts that long. Wrong.
2988 Somebody had a whole pile of complicated expansion conditions, and the
2989 string was silently truncated, leading to an expansion error. It turns out
2990 that it is easier to change to an unlimited length (owing to other changes
2991 that have happened since this code was originally written) than to build
2992 structure for giving a limitation error. The length of the item that
2993 expands into the list of hosts is now unlimited.
2995 47. The lsearch lookup could not handle data where the length of text line was
2996 more than 4095 characters. Such lines were truncated, leading to shortened
2997 data being returned. It should now handle lines of any length.
2999 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3000 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3003 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3004 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3005 no longer happens in comment lines. A list of replacements is now placed
3006 at the head of all of the source files, except those whose only change is
3007 to replace PERL_COMMAND in the very first #! line.
3009 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3010 messages on the queue, with a bottom-up merge sort, using code contributed
3011 by Michael Haardt. This should make operations like -bp somewhat faster on
3012 large queues. It won't affect queue runners, except when queue_run_in_order
3015 51. Installed eximstats 1.31 in the distribution.
3017 52. Added support for SRV lookups to the dnslookup router.
3019 53. If an ACL referred to $message_body or $message_body_end, the value was not
3020 reset for any messages that followed in the same SMTP session.
3022 54. The store-handling optimization for building very long strings was not
3023 differentiating between the different store pools. I don't think this
3024 actually made any difference in practice, but I've tidied it.
3026 55. While running the routers to verify a sender address, $sender_address
3027 was still set to the sender address. This is wrong, because when routing to
3028 send a bounce to the sender, it would be empty. Therefore, I have changed
3029 it so that, while verifying a sender address, $sender_address is set to <>.
3030 (There is no change to what happens when verifying a recipient address.)
3032 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3033 target A or AAAA records (if not already returned) without resetting the
3034 qualify_single or search_parents options of the DNS resolver. These are
3035 inappropriate in this case because the targets of MX and SRV records must
3036 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3037 target that, when qualified, matched something in the local domain. These
3038 two options are now turned off when doing these lookups.
3040 57. It seems that at least some releases of Reiserfs (which does not have the
3041 concept of a fixed number of inodes) returns zero and not -1 for the
3042 number of available inodes. This interacted badly with check_spool_inodes,
3043 which assumed that -1 was the "no such thing" setting. What I have done is
3044 to check that the total number of inodes is greater than zero before doing
3045 the test of how many are available.
3047 58. When a "warn" ACL statement has a log_message modifier, the message is
3048 remembered, and not repeated. This is to avoid a lot of repetition when a
3049 message has many recipients that cause the same warning to be written.
3050 Howewer, Exim was preserving the list of already written lines for an
3051 entire SMTP session, which doesn't seem right. The memory is now reset if a
3052 new message is started.
3054 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3055 debugging output unless log_rewrite was also set.
3057 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3058 of (int)(handle) when we know that handle contains (void *)(-1).
3060 61. The Exim daemon panic-logs an error return when it closes the incoming
3061 connection. However "connection reset by peer" seems to be common, and
3062 isn't really an error worthy of noting specially, so that particular error
3065 62. When Exim is trying to find all the local interfaces, it used to panic and
3066 die if the ioctl to get the interface flags failed. However, it seems that
3067 on at least one OS (Solaris 9) it is possible to have an interface that is
3068 included in the list of interfaces, but for which you get a failure error
3069 for this call. This happens when the interface is not "plumbed" into a
3070 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3071 failure of the "get flags" call assumes that the interface is down.
3073 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3074 makes life easier for people who are doing arithmetic on fields extracted
3075 from dates, where you often get leading zeros that should not be
3076 interpreted as octal.
3078 64. Added qualify_domain to the redirect router, to override the global
3081 65. If a pathologically long header line contained very many addresses (the
3082 report of this problem mentioned 10 000) and each of them was rewritten,
3083 Exim could use up a very large amount of memory. (It kept on making new
3084 copies of the header line as it rewrote, and never released the old ones.)
3085 At the expense of a bit more processing, the header rewriting function has
3086 been changed so that it no longer eats memory in this way.
3088 66. The generation of the Received: header has been moved from the time that a
3089 message starts to be received, to the time that it finishes. The timestamp
3090 in the Received: header should now be very close to that of the <= log
3091 line. There are two side-effects of this change:
3093 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3094 logged header lines no longer include the local Received: line, because
3095 it has not yet been created. The same applies to a copy of the message
3096 that is returned to a non-SMTP sender when a message is rejected.
3098 (b) When a filter file is tested using -bf, no additional Received: header
3099 is added to the test message. After some thought, I decided that this
3102 This change does not affect the value of $received_for. It is still set
3103 after address rewriting, but before local_scan() is called.
3105 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3107 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3108 gave an unhelpful panic error message, and a defer error. I have managed to
3109 change this behaviour so that it now rejects any supplied certificate,
3110 which seems right, as the list of acceptable certificates is empty.
3112 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3113 gave an unhelpful defer error. I have not managed to make this reject any
3114 supplied certificates, but the error message it gives is "no certificate
3115 supplied", which is not helpful.
3117 70. exigrep's output now also includes lines that are not associated with any
3118 message, but which match the given pattern. Implemented by a patch from
3119 Martin Sluka, which also tidied up the Perl a bit.
3121 71. Recipient callout verification, like sender verification, was using <> in
3122 the MAIL FROM command. This isn't really the right thing, since the actual
3123 sender may affect whether the remote host accepts the recipient or not. I
3124 have changed it to use the actual sender in the callout; this means that
3125 the cache record is now keyed on a recipient/sender pair, not just the
3126 recipient address. There doesn't seem to be a real danger of callout loops,
3127 since a callout by the remote host to check the sender would use <>.
3128 [SEE ABOVE: changed after hitting problems.]
3130 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3131 temporary errors. However, in the case of such a code being given after
3132 the end of a data transmission (i.e. after ".") Exim was failing to write
3133 a retry record for the message. (Yes, there was some broken host that was
3134 actually sending 8xx at this point.)
3136 73. An unknown lookup type in a host list could cause Exim to panic-die when
3137 the list was checked. (An example that provoked this was putting <; in the
3138 middle of a list instead of at the start.) If this happened during a DATA
3139 ACL check, a -D file could be left lying around. This kind of configuration
3140 error no longer causes Exim to die; instead it causes a defer errror. The
3141 incident is still logged to the main and panic logs.
3143 74. Buglet left over from Exim 3 conversion. The message "too many messages
3144 in one connection" was written to the rejectlog but not the mainlog, except
3145 when address rewriting (yes!) was being logged.
3147 75. Added write_rejectlog option.
3149 76. When a system filter was run not as root (that is, when system_filter_user
3150 was set), the values of the $n variables were not being returned to the
3151 main process; thus, they were not subsequently available in the $sn
3154 77. Added +return_path_on_delivery log selector.
3156 78. A connection timeout was being treated differently from recipients deferred
3157 when testing hosts_max_try with a message that was older than the host's
3158 retry timeout. (The host should not be counted, thus allowing all hosts to
3159 be tried at least once before bouncing.) This may have been the cause of an
3160 occasionally reported bug whereby a message would remain on the queue
3161 longer than the retry timeout, but would be bounced if a delivery was
3162 forced. I say "may" because I never totally pinned down the problem;
3163 setting up timeout/retry tests is difficult. See also the next item.
3165 79. The ultimate address timeout was not being applied to errors that involved
3166 a combination of host plus message (for example, a timeout on a MAIL
3167 command). When an address resolved to a number of possible hosts, and they
3168 were not all tried for each delivery (e.g. because of hosts_max_try), a
3169 message could remain on the queue longer than the retry timeout.
3171 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3174 81. Fixed an obscure SMTP outgoing bug which required at least the following
3175 conditions: (a) there was another message waiting for the same server;
3176 (b) the server returned 5xx to all RCPT commands in the first message so
3177 that the message was not completed; (c) the server dropped the connection
3178 or gave a negative response to the RSET that Exim sends to abort the
3179 transaction. The observed case was a dropped connection after DATA that had
3180 been sent in pipelining mode. That is, the server had advertised PIPELINING
3181 but was not implementing it correctly. The effect of the bug was incorrect
3182 behaviour, such as trying another host, and this could lead to a crash.
3188 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3189 and daemon.c were passed as pointers to ints; they should have been
3190 pointers to socklen_t variables (which are typically unsigned ints).
3192 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3195 3. Fixed a really odd bug that affected only the testing scheme; patching a
3196 certain fixed string in the binary changed the value of another string that
3197 happened to be identical to the end of the original first string.
3199 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3200 name", it returns that address as the IP address. On some operating
3201 systems (e.g. Solaris), it also passes back the IP address string as the
3202 "host name". However, on others (e.g. Linux), it passes back an empty
3203 string. Exim wasn't checking for this, and was changing the host name to an
3204 empty string, assuming it had been canonicized.
3206 5. Although rare, it is permitted to have more than one PTR record for a given
3207 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3208 all the names associated with an address, because they do in Solaris.
3209 However, it seems that they do not in Linux for data that comes from the
3210 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3211 I found this out when I moved to a new Linux workstation and tried to run
3212 the Exim test suite.
3214 To get round this problem I have changed the code so that it now does its
3215 own call to the DNS to look up PTR records when searching for a host name.
3216 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3217 addresses that are only in /etc/hosts are still found.
3219 This behaviour is, however, controlled by an option called host_lookup_
3220 order, which defaults to "bydns:byaddr". If people want to use the other
3221 order, or indeed, just use one or the other means of lookup, they can
3222 specify it in this variable.
3224 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3225 some operating systems, this comes back from gethostbyaddr() as an empty
3226 string, and this is what Exim used to test for. However, it seems that in
3227 other systems, "." is yielded. Exim now tests for this case too.
3229 7. The values of check_spool_space and check_log_space are now held internally
3230 as a number of kilobytes instead of an absolute number of bytes. If a
3231 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3232 kilobyte. This means that much larger values can be stored.
3234 8. Exim monitor: an attempt to get the action menu when not actually pointing
3235 at a message produces an empty menu entitled "No message selected". This
3236 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3237 no entries in it ("Shell widget menu has zero width and/or height"). So I
3238 have added a single, blank menu entry in this case.
3240 9. Added ${quote_local_part.
3242 10. MIME decoding is now applied to the contents of Subject: header lines when
3245 11. Now that a reference to $sender_host_address automatically causes a reverse
3246 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3247 host lookup before query-style lookups in lists that might use this
3248 variable. This has therefore been abolished, and the "net-" prefix is no
3249 longer necessary for query-style lookups.
3251 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3252 have been a typo for LFLAGS, so it has been changed.
3254 13. The install script calls Exim with "-C /dev/null" in order to find the
3255 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3256 to be output. Howeve, since Exim outputs its version number before the
3257 error, it didn't break the script. It just looked ugly. I fixed this by
3258 always allowing "-C /dev/null" if the caller is root.
3260 14. Ignore overlarge ACL variable number when reading spool file - insurance
3261 against a later release with more variables having written the file.
3263 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3264 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3265 this, as well as the form without the "IPv6" on the front.
3267 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3268 OS/Makefile-Darwin file.
3270 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3271 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3273 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3274 variable. However, some OS use ints for inodes. I've added cast to long int
3275 to get rid of the compiler warning.
3277 19. I had forgotten to lock out "/../" in configuration file names when
3278 ALT_CONFIG_PREFIX was set.
3280 20. Routers used for verification do not need to specify transports. However,
3281 if such a router generated a host list, and callout was configured, Exim
3282 crashed, because it could not find a port number from the (non-existent)
3283 transport. It now assumes port 25 in this circumstance.
3285 21. Added the -t option to exigrep.
3287 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3288 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3289 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3290 them is set, LOOKUP_LSEARCH is forced.
3292 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3294 24. Added sender and host information to the "rejected by local_scan()" log
3295 line; previously there was no indication of these.
3297 25. Added .include_if_exists.
3299 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3300 Linux. It turns out that not all file systems support this. Apparently some
3301 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3302 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3303 not supported on the file descriptor, is now ignored when Exim is trying to
3304 sync a directory. This applies only to Linux.
3306 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3308 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3309 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3312 29. Added support for maildirsize files from supplied patch (modified a bit).
3314 30. The use of :fail: followed by an empty string could lead Exim to respond to
3315 sender verification failures with (e.g.):
3317 550 Verification failed for <xxx>
3318 550 Sender verify failed
3320 where the first response line was missing the '-' that indicates it is not
3321 the final line of the response.
3323 31. The loop for finding the name of the user that called Exim had a hardwired
3324 limit of 10; it now uses the value of finduser_retries, which is used for
3325 all other user lookups.
3327 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3330 33. Exim was neglecting to zero errno before one call of strtol() when
3331 expanding a string and expecting an integer value. On some systems this
3332 resulted in spurious "integer overflow" errors. Also, it was casting the
3333 result into an int without checking.
3335 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3336 did not work. The code looks as if it has *never* worked, though it appears
3337 to have been documented since at least releast 1.62. I have made it work.
3339 35. The "timeout_DNS" error in retry rules, also documented since at least
3340 1.62, also never worked. As it isn't clear exactly what this means, and
3341 clearly it isn't a major issue, I have abolished the feature by treating it
3342 as "timeout", and writing a warning to the main and panic logs.
3344 36. The display of retry rules for -brt wasn't always showing the error code
3347 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3348 timeout_connect_A, timeout_connect_MX.
3350 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3351 to the empty sender.
3353 39. The daemon was not analysing the content of -oX till after it had closed
3354 stderr and disconnected from the controlling terminal. This meant that any
3355 syntax errors were only noted on the panic log, and the return code from
3356 the command was 0. By re-arranging the code a little, I've made the
3357 decoding happen first, so such errors now appear on stderr, and the return
3358 code is 1. However, the actual setting up of the sockets still happens in
3359 the disconnected process, so errors there are still only recorded on the
3362 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3363 connections (as happens on some IP stacks) was logged at start up time as
3364 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3365 it from "IPv6 and IPv4", which means that two separate sockets are being
3368 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3369 says whether AF_INET or AF_INET6 was passed as an argument.
3371 42. Exiwhat output was messed up when time zones were included in log
3374 43. Exiwhat now gives more information about the daemon's listening ports,
3375 and whether -tls-on-connect was used.
3377 44. The "port" option of the smtp transport is now expanded.
3379 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3380 silently ignored. Now an error message is written to the main and panic
3383 46. There's a new ACL modifier called "logwrite" which writes to a log file
3384 as soon as it is encountered.
3386 47. Added $local_user_uid and $local_user_gid at routing time.
3388 48. Exim crashed when trying to verify a sender address that was being
3391 49. Exim was recognizing only a space character after ".include". It now also
3392 recognizes a tab character.
3394 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3395 extracting the relevant information from the specification. The man page no
3396 longer contains scrambled data for the -d option, and I've added a section
3397 at the front about calling Exim under different names.
3399 51. Added "extra_headers" argument to the "mail" command in filter files.
3401 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3404 53. Installed eximstats 1.29.
3406 54. Added transport_filter_timeout as a generic transport option.
3408 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3409 Cc: header lines. This was required by RFC 822, but it not required by RFC
3412 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3413 incoming messages that did not have them. Now it does so only if the
3414 message originates locally, that is, if there is no associated remote host
3415 address. When Resent- header lines are present, this applies to the Resent-
3416 lines rather than the non-Resent- lines.
3418 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3419 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3421 58. Messages for configuration errors now include the name of the main
3422 configuration file - useful now that there may be more than one file in a
3423 list (.included file names were always shown).
3425 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3426 for those rare installations that do not start the daemon as root or run it
3427 setuid root. I've cut out the call to initgroups() if the daemon is not
3430 60. The Exim user and group can now be bound into the binary as text strings
3431 that are looked up at the start of Exim's processing.
3433 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3435 62. Added $mailstore_basename variable.
3437 63. Installed patch to sieve.c from Michael Haardt.
3439 64. When Exim failed to open the panic log after failing to open the main log,
3440 the original message it was trying to log was written to stderr and debug
3441 output, but if they were not available (the usual case in production), it
3442 was lost. Now it is written to syslog before the two lines that record the
3443 failures to open the logs.
3445 65. Users' Exim filters run in subprocesses under the user's uid. It is
3446 possible for a "deliver" command or an alias in a "personal" command to
3447 provoke an address rewrite. If logging of address rewriting is configured,
3448 this fails because the process is not running as root or exim. There may be
3449 a better way of dealing with this, but for the moment (because 4.30 needs
3450 to be released), I have disabled address rewrite logging when running a
3451 filter in a non-root, non-exim process.
3457 1. The buildconfig auxiliary program wasn't quoting the value set for
3458 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3459 not defined. This bug was masked in 4.22 by the effect that was fixed in
3462 2. Some messages that were rejected after a message id was allocated were
3463 shown as "incomplete" by exigrep. It no longer does this for messages that
3464 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3466 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3467 have allow_domain_literals set, the ID did not get logged in the <= line.
3468 Domain literals are now always recognized in Message-ID: header lines.
3470 4. The first argument for a ${extract expansion item is the key name or field
3471 number. Leading and trailing spaces in this item were not being ignored,
3472 causing some misleading effects.
3474 5. When deliver_drop_privilege was set, single queue runner processes started
3475 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3476 same command in the process it spins off) were not dropping privilege.
3478 6. When the daemon running as "exim" started a queue runner, it always
3479 re-executed Exim in the spun-off process. This is a waste of effort when
3480 deliver_drop_privilege is set. The new process now just calls the
3481 queue-runner function directly.
3487 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3490 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3491 ignored. Though the use of -oP was forcing the writing of a pid file, it
3492 was always written to the default place.
3494 3. If the message "no IP address found for host xxxx" is generated during
3495 incoming verification, it is now followed by identification of the incoming
3496 connection (so you can more easily find what provoked it).
3498 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3500 5. Added some features to "harden" Exim a bit more against certain attacks:
3502 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3503 be put in Local/Makefile. This is like the never_users runtime option,
3504 but it cannot be overridden. The default setting is "root".
3506 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3507 prefix string with which any file named in a -C command line option
3510 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3511 is retained for -C and -D only if the caller of Exim is root. Without
3512 it, the exim user may also use -C and -D and retain privilege.
3514 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3515 command line option is disabled.
3517 6. Macro names set by the -D option must start with an upper case letter, just
3518 like macro names defined in the configuration file.
3520 7. Added "dereference=" facility to LDAP.
3522 8. Two instances of the typo "uknown" in the source files are fixed.
3524 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3525 the Configure-Makefile script screwed up while processing it.
3527 10. Incorporated PCRE 4.4.
3529 11. The SMTP synchronization check was not operating right at the start of an
3530 SMTP session. For example, it could not catch a HELO sent before the client
3531 waited for the greeting. There is now a check for outstanding input at the
3532 point when the greeting is written. Because of the duplex, asynchronous
3533 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3534 way, but not yet received, when the check is performed.
3536 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3537 on TCP/IP sockets, because this apparently causes some broken clients to
3540 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3541 unchanged) from the Cygwin maintainer.
3543 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3544 was supported instead of testing for mbx. Effectively a typo.
3546 15. The spa authenticator server code was not checking that the input it
3547 received was valid base64.
3549 16. The debug output line for the "set" modifier in ACLs was not showing the
3550 name of the variable that was being set.
3552 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3554 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3555 lookup on its own. Something else was needed to trigger the lookup. For
3556 example, a match in host_lookup or the need for a host name in a host list.
3557 Now, if $sender_host_name is referenced and the host name has not yet been
3558 looked up, a lookup is performed. If the lookup fails, the variable remains
3559 empty, and $host_lookup_failed is set to "1".
3561 19. Added "eqi" as a case-independent comparison operator.
3563 20. The saslauthd authentication condition could segfault if neither service
3564 nor realm was specified.
3566 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3567 error message that was logged was misleading, and incoming SMTP
3568 connections were dropped. The message is now more accurate, and temporary
3569 errors are given to SMTP connections.
3571 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3572 (except RSET) with a 421 error, until QUIT is received. However, it was
3573 failing to send a response to QUIT.
3575 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3576 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3577 ACL is now run after the helo_try_verify_hosts code.
3579 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3580 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3581 case-independent, so other case variants are also recognized). Apparently
3582 some systems use these upper case variants.
3584 25. If more than two messages were waiting for the same host, and a transport
3585 filter was specified for the transport, Exim sent two messages over the
3586 same TCP/IP connection, and then failed with "socket operation on non-
3587 socket" when it tried to send the third.
3589 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3591 27. The extern definition of crypt16() in expand.c was not being excluded when
3592 the OS had its own crypt16() function.
3594 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3595 as a preferred synonym for return_size_limit, both as an option and as an
3598 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3600 30. Changed the default configuration ACL to relax the local part checking rule
3601 for addresses that are not in any local domains. For these addresses,
3602 slashes and pipe symbols are allowed within local parts, but the sequence
3603 /../ is explicitly forbidden.
3605 31. SPA server authentication was not clearing the challenge buffer before
3608 32. log_message in a "warn" ACL statement was writing to the reject log as
3609 well as to the main log, which contradicts the documentation and doesn't
3610 seem right (because no rejection is happening). So I have stopped it.
3612 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3613 However, I am unable to do any testing of this.
3615 34. Fixed an infelicity in the appendfile transport. When checking directories
3616 for a mailbox, to see if any needed to be created, it was accidentally
3617 using path names with one or more superfluous leading slashes; tracing
3618 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3620 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3621 contents are added to the log line that is written for every discarded
3622 recipient. (Previously a log_message setting was ignored.)
3624 36. The ${quote: operator now quotes the string if it is empty.
3626 37. The install script runs exim in order to find its version number. If for
3627 some reason other than non-existence or emptiness, which it checks, it
3628 could not run './exim', it was installing it with an empty version number,
3629 i.e. as "exim-". This error state is now caught, and the installation is
3632 38. An argument was missing from the function that creates an error message
3633 when Exim fails to connect to the socket for saslauthd authentication.
3634 This could cause Exim to crash, or give a corrupted message.
3636 39. Added isip, isip4, and isip6 to ${if conditions.
3638 40. The ACL variables $acl_xx are now saved with the message, and can be
3639 accessed later in routers, transports, and filters.
3641 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3642 strings in the file are not string-expanded.
3644 42. If a MAIL command specified a SIZE value that was too large to fit into an
3645 int variable, the check against message_size_limit failed. Such values are
3646 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3647 day this will have to be increased, but I don't think I want to be around
3648 when emails are that large.
3655 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3656 iconv() is not standard in FreeBSD.
3658 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3659 IPv6 enabled. The observed symptom was a segmentation fault on return from
3660 the function os_common_find_running_interfaces() in src/os.c.
3662 3. In the check_special_case() function in daemon.c I had used "errno" as an
3663 argument name, which causes warnings on some systems. This was basically a
3664 typo, since it was named "eno" in the comments!
3666 4. The code that waits for the clock to tick (at a resolution of some fraction
3667 of a second) so as to ensure message-id uniqueness was always waiting for
3668 at least one whole tick, when it could have waited for less. [This is
3669 almost certainly not relevant at current processor speeds, where it is
3670 unlikely to ever wait at all. But we try to future-proof.]
3672 5. The function that sleeps for a time interval that includes fractions of a
3673 second contained a race. It did not block SIGALRM between setting the
3674 timer, and suspending (a couple of lines later). If the interval was short
3675 and the sigsuspend() was delayed until after it had expired, the suspension
3676 never ended. On busy systems this could lead to processes getting stuck for
3679 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3680 process, before it forks any delivery processes. The open lookup caching
3681 mechanism meant that the open file or database connection was passed into
3682 the delivery process. The problem was that delivery processes always tidy
3683 up cached lookup data. This could cause a problem for the next delivery
3684 process started by the queue runner, because the external queue runner
3685 process does not know about the closure. So the next delivery process
3686 still has data in the lookup cache. In the case of a file lookup, there was
3687 no problem because closing a file descriptor in a subprocess doesn't affect
3688 the parent. However, if the lookup was caching a connection to a database,
3689 the connection was closed, and the second delivery process was likely to
3690 see errors such as "PGSQL: query failed: server closed the connection
3691 unexpectedly". The problem has been fixed by closing all cached lookups
3692 in a queue runner before running a delivery process.
3694 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3695 seem to have the "const" qualifier which it has on other OS. I've
3698 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3699 *of the same type* (client or server) with the same public name that an
3700 error should be diagnosed.
3702 9. When Exim looked up a host name for an IP address, but failed to find the
3703 original IP address when looking up the host name (a safety check), it
3704 output the message "<ip address> does not match any IP for NULL", which was
3705 confusing, to say the least. The bug was that the host name should have
3706 appeared instead of "NULL".
3708 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3709 user that is built into the binary, and the -C or -D options is used, root
3710 privilege is dropped before the configuration file is read. In addition,
3711 logging is switched to stderr instead of the normal log files. If the
3712 configuration then re-defines the Exim user, the unprivileged environment
3713 is probably not what is expected, so Exim logs a panic warning message (but
3716 However, if deliver_drop_privilege is set, the unprivileged state may well
3717 be exactly what is intended, so the warning has been cut out in that case,
3718 and Exim is allowed to try to write to its normal log files.
3724 1. smtp_return_error_details was not giving details for temporary sender
3725 or receiver verification errors.
3727 2. Diagnose a configuration error if two authenticators have the same public
3730 3. Exim used not to create the message log file for a message until the first
3731 delivery attempt. This could be confusing when incoming messages were held
3732 for policy or load reasons. The message log file is now created at the time
3733 the message is received, and an initial "Received" line is written to it.
3735 4. The automatically generated man page for command line options had a minor
3736 bug that caused no ill effects; however, a more serious problem was that
3737 the procedure for building the man page automatically didn't always
3738 operate. Consequently, release 4.20 contains an out-of-date version. This
3739 shouldn't happen again.
3741 5. When building Exim with embedded Perl support, the script that builds the
3742 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3743 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3745 6. The freeze_tell option was not being used for messages that were frozen on
3746 arrival, either by an ACL or by local_scan().
3748 7. Added the smtp_incomplete_transaction log selector.
3750 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3751 was accepting AUTH without a new EHLO.
3753 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3754 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3757 10. Logging of TCP/IP connections (when configured) now happens in the main
3758 daemon process instead of the child process, so that the TCP/IP connection
3759 count is more accurate (but it can never be perfect).
3761 11. The use of "drop" in a nested ACL was not being handled correctly in the
3762 outer ACL. Now, if condition failure induced by the nested "drop" causes
3763 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3764 or "require"), the connection is dropped.
3766 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3767 that yield "discard" can now be used with an "accept" or a "discard" verb,
3768 but an error is generated for any others (because I can't see a useful way
3769 to define what should happen).
3771 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3772 are now processed in the same way as lines in the Exim configuration file.
3773 In particular, continuation lines are supported.
3775 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3777 15. Added -ti meaning -t -i.
3779 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3780 domains, and warn by logging if others are found.
3782 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3783 returned from GIFCONF when Exim is trying to find the list of interfaces on
3784 a host. The code in os.c has been modified to copy each ifreq to an aligned
3785 structure in all cases.
3787 Also, in some cases, the returned ifreq's were being copied to a 'struct
3788 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3789 means the last couple of bytes of an IPv6 address could be chopped if the
3790 ifreq contained only a normal sockaddr (14 bytes storage).
3792 18. Named domain lists were not supported in the hosts_treat_as_local option.
3793 An entry such as +xxxx was not recognized, and was treated as a literal
3796 19. Ensure that header lines added by a DATA ACL are included in the reject log
3797 if the ACL subsequently rejects the message.
3799 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3800 MD5 (which is deprecated).
3802 21. When testing a filter file using -bf, Exim was writing a message when it
3803 took the sender from a "From " line in the message, but it was not doing so
3804 when it took $return_path from a Return-Path: header line. It now does.
3806 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3807 with a valid header line field name (a series of printing characters
3808 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3810 23. Changed "disc" in the source to "disk" to conform to the documentation and
3811 the book and for uniformity.
3813 24. Ignore Sendmail's -Ooption=value command line item.
3815 25. When execve() failed while trying to run a command in a pipe transport,
3816 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3817 could be confused with a return value of 69 from the command itself. This
3818 has been changed to 127, the value the shell returns if it is asked to run
3819 a non-existent command. The wording for the related log line suggests a
3820 non-existent command as the problem.
3822 26. If received_header_text expands to an empty string, do not add a Received:
3823 header line to the message. (Well, it adds a token one on the spool, but
3824 marks it "old" so that it doesn't get used or transmitted.)
3826 27. Installed eximstats 1.28 (addition of -nt option).
3828 28. There was no check for failure on the call to getsockname() in the daemon
3829 code. This can fail if there is a shortage of resources on the system, with
3830 ENOMEM, for example. A temporary error is now given on failure.
3832 29. Contrary to the C standard, it seems that in some environments, the
3833 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3834 program. Exim now does this explicitly; it affects the formatting of
3835 timestamps using strftime().
3837 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3838 complaints. I've now initialized all the variables, to avoid this.
3840 32. Header lines added by a system filter were not being "seen" during
3841 transport-time rewrites.
3843 33. The info_callback() function passed to OpenSSL is set up with type void
3844 (*)(SSL *, int, int), as described somewhere. However, when calling the
3845 function (actually a macro) that sets it up, the type void(*)() is
3846 expected. I've put in a cast to prevent warnings from picky compilers.
3848 34. If a DNS black list lookup found a CNAME record, but there were no A
3849 records associated with the domain it pointed at, Exim crashed.
3851 35. If a DNS black list lookup returned more than one A record, Exim ignored
3852 all but the first. It now scans all returned addresses if a particular IP
3853 value is being sought. In this situation, the contents of the
3854 $dnslist_value variable are a list of all the addresses, separated by a
3857 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3858 to accept a host name and all its aliases if the forward lookup for any of
3859 them yielded the IP address of the incoming connection. Now it accepts only
3860 those names whose forward lookup yields the correct IP address. Any other
3861 names are discarded. This closes a loophole whereby a rogue DNS
3862 administrator could create reverse DNS records to break through a
3863 wildcarded host restriction in an ACL.
3865 37. If a user filter or a system filter that ran in a subprocess used any of
3866 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3867 the wrong values were passed to the pipe command ($thisaddress had the
3868 value of $0, $0 had the value of $1, etc). This bug was introduced by
3869 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3871 38. Improved the line breaking for long SMTP error messages from ACLs.
3872 Previously, if there was no break point between 40 and 75 characters, Exim
3873 left the rest of the message alone. Two changes have been made: (a) I've
3874 reduced the minimum length to 35 characters; (b) if it can't find a break
3875 point between 35 and 75 characters, it looks ahead and uses the first one
3876 that it finds. This may give the occasional overlong line, but at least the
3877 remaining text gets split now.
3879 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3880 file descriptors might be low, and that setting 1000 would always raise it.
3881 It turns out that in some environments, the limit is already over 1000 and
3882 that lowering it causes trouble. So now Exim takes care not to decrease it.
3884 40. When delivering a message, the value of $return_path is set to $sender_
3885 address at the start of routing (routers may change the value). By an
3886 oversight, this default was not being set up when an address was tested by
3887 -bt or -bv, which affected the outcome if any router or filter referred to
3890 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3891 log only when "message" or "log_message" are set. However, if one of the
3892 conditions was an address verification, or a call to a nested ACL, the
3893 messages generated by the underlying test were being passed through. This
3894 no longer happens. The underlying message is available in $acl_verify_
3895 message for both "message" and "log_message" expansions, so it can be
3896 passed through if needed.
3898 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3899 new expansion $bh_ to give the encoded byte string without charset
3900 translation. Translation happens only if iconv() is available; HAVE_ICONV
3901 indicates this at build time. HEADERS_CHARSET gives the charset to
3902 translate to; headers_charset can change it in the configuration, and
3903 "headers charset" can change it in an individual filter file.
3905 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3906 that creates RFC 2047 encoded "words" labels them as that charset instead
3907 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3908 expansion operator; (ii) when Exim creates a From: line for a local
3909 message; (iii) when a header line is rewritten to include a "phrase" part.
3911 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3912 buggy, causing it to skip the first lines of messages whose message ID
3913 ended in 'D'. This would not have bitten before Exim release 4.14, because
3914 message IDs were unlikely to end in 'D' before then. The effect was to have
3915 incorrect size information for certain domains.
3917 45. #include "config.h" was missing at the start of the crypt16.c module. This
3918 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3921 46. If there was a timeout during a "random" callout check, Exim treated it as
3922 a failure of the random address, and carried on sending RSET and the real
3923 address. If the delay was just some slowness somewhere, the response to the
3924 original RCPT would be taken as a response to RSET and so on, causing
3925 mayhem of various kinds.
3927 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3928 when I implemented it. It didn't allow for the fact that some option values
3929 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3930 the right test for positive values.
3932 48. Domain names in DNS records are case-independent. Exim always looks them up
3933 in lower case. Some resolvers return domain names in exactly the case they
3934 appear in the zone file, that is, they may contain uppercase letters. Not
3935 all resolvers do this - some return always lower case. Exim was treating a
3936 change of case by a resolver as a change of domain, similar to a widening
3937 of a domain abbreviation. This triggered its re-routing code and so it was
3938 trying to route what was effectively the same domain again. This normally
3939 caused routing to fail (because the router wouldn't handle the domain
3940 twice). Now Exim checks for this case specially, and just changes the
3941 casing of the domain that it ultimately uses when it transmits the message
3944 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3947 50. If a filter generated a file delivery with a non-absolute name (possible if
3948 no home directory exists for the router), the forbid_file option was not
3951 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3952 the existing equality matching.
3954 52. Exim was using ints instead of ino_t variables in some places where it was
3955 dealing with inode numbers.
3957 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3958 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3959 called TMPDIR, and if it finds it is different, it changes its value.
3961 54. The smtp_printf() function is now made available to local_scan() so
3962 additional output lines can be written before returning. There is also an
3963 smtp_fflush() function to enable the detection of a dropped connection.
3964 The variables smtp_input and smtp_batched_input are exported to
3967 55. Changed the default runtime configuration: the message "Unknown user"
3968 has been removed from the ACL, and instead placed on the localuser router,
3969 using the cannot_route_message feature. This means that any verification
3970 failures that generate their own messages won't get overridden. Similarly,
3971 the "Unrouteable address" message that was in the ACL for unverifiable
3972 relay addresses has also been removed.
3974 56. Added hosts_avoid_esmtp to the smtp transport.
3976 57. The exicyclog script was not checking for the esoteric option
3977 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3978 will work only if exicyclog is run under the appropriate euid.
3980 58. Following a discussion on the list, the rules by which Exim recognises line
3981 endings on incoming messages have been changed. The -dropcr and drop_cr
3982 options are now no-ops, retained only for backwards compatibility. The
3983 following line terminators are recognized: LF CRLF CR. However, special
3984 processing applies to CR:
3986 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3987 nor a local message in the state where . is a terminator.
3989 (ii) If a bare CR is encountered in a header line, an extra space is added
3990 after the line terminator so as not to end the header. The reasoning
3991 behind this is that bare CRs in header lines are most likely either
3992 to be mistakes, or people trying to play silly games.
3994 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3995 was being incorrectly given as 18 bytes larger than it should have been.
3996 This is a VOB (very old bug).
3998 60. This may never have affected anything current, but just in case it has:
3999 When the local host is found other than at the start of a list of hosts,
4000 the local host, those with the same MX, and any that follow, are discarded.
4001 When the list in question was part of a longer list of hosts, the following
4002 hosts (not currently being processed) were also being discarded. This no
4003 longer happens. I'm not sure if this situation could ever has previously
4006 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4009 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4010 header. This is something that is recommended in a new Internet Draft, and
4011 is something that is documented as being done by Sendmail. There are two
4012 possible values. For messages generated by the autoreply transport, Exim
4015 Auto-Submitted: auto-replied
4017 whereas for all other generated messages (e.g. bounces) it adds
4019 Auto-Submitted: auto-generated
4021 63. The "personal" condition in filters now includes a test for the
4022 Auto-Submitted: header. If it contains the string "auto-" the message it
4023 not considered personal.
4025 64. Added rcpt_include_affixes as a generic transport option.
4027 65. Added queue_only_override (default true).
4029 66. Added the syslog_duplication option.
4031 67. If what should have been the first header line of a message consisted of
4032 a space followed by a colon, Exim was mis-interpreting it as a header line.
4033 It isn't of course - it is syntactically invalid and should therefore be
4034 treated as the start of the message body. The misbehaviour could have
4035 caused a number of strange effects, including loss of data in subsequent
4036 header lines, and spool format errors.
4038 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4039 client host had authenticated. This control can now be exercised by an ACL
4040 for more flexibility.
4042 69. By default, callouts do not happen when testing with -bh. There is now a
4043 variant, -bhc, which does actually run the callout code, including
4044 consulting and updating the callout cache.
4046 70. Added support for saslauthd authentication, courtesy of Alexander
4049 71. If statvfs() failed on the spool or log directories while checking their
4050 size for availability, Exim confusingly gave the error "space shortage".
4051 Furthermore, in debugging mode it crashed with a floating point exception.
4052 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4053 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4054 serious problem, Exim now writes to the main and panic logs when this
4055 happens, with details of the failure. It then refuses to accept the
4056 incoming message, giving the message "spool directory problem" or "log
4057 directory problem" with a 421 code for SMTP messages.
4059 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4060 0, 1, and 2 exist, because some OS complain for execs without them (see
4061 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4062 descriptors. However, the code omitted to check that the open succeeded,
4063 causing mysterious errors if for some reason the permissions on /dev/null
4064 got screwed. Now Exim writes a message to the main and panic logs, and
4065 bombs out if it can't open /dev/null.
4067 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4068 interact so that it is all more flexible. It is supposed to remain
4069 backwards compatible. Also added extra_local_interfaces.
4071 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4072 to bomb out with an assertion failure - to the client this appears as a
4073 connection drop. This problem occurs in the part of the code that was taken
4074 from the Samba project. Fortunately, the assertion is in a very simple
4075 function, so I have fixed this by reproducing the function inline in the
4076 one place where it is called, and arranging for authentication to fail
4077 instead of killing the process with assert().
4079 75. The SPA client code was not working when the server requested OEM rather
4080 than Unicode encoding.
4082 76. Added code to make require_files with a specific uid setting more usable in
4083 the case where statting the file as root fails - usually a non-root-mounted
4084 NFS file system. When this happens and the failure is EACCES, Exim now
4085 forks a subprocess and does the per-uid checking as the relevant uid.
4087 77. Added process_log_path.
4089 78. If log_file_path was not explicitly set, a setting of check_log_space or
4090 check_log_inodes was ignored.
4092 79. If a space check for the spool or log partitions fails, the incident is now
4093 logged. Of course, in the latter case the data may get lost...
4095 80. Added the %p formatting code to string_format() so that it can be used to
4096 print addresses in debug_print(). Adjusted all the address printing in the
4097 debugging in store.c to use %p rather than %d.
4099 81. There was a concern that a line of code in smtp_in.c could overflow a
4100 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4101 initially expressed, the concern was not well-founded, because trailing
4102 spaces are removed early. However, if the trailing spaces were followed by
4103 a NULL, they did not get removed, so the overflow was possible. Two fixes
4106 (a) I re-wrote the offending code in a cleaner fashion.
4107 (b) If an incoming SMTP command contains a NULL character, it is rejected
4110 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4111 runs initgroups(), so that if the Exim user is in any additional groups,
4112 they will be used during message reception.
4118 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob