1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.401 2006/10/03 08:54:50 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TF/01 In the add_headers option to the mail command in an Exim filter,
10 there was a bug that Exim would claim a syntax error in any
11 header after the first one which had an odd number of characters
14 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
15 callout verification, Exim cached a "reject" for the entire domain. This
16 is correct for most verifications, but it is not correct for a recipient
17 verification with use_sender or use_postmaster set, because in that case
18 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
19 case of MAIL FROM:<> rejection from other early rejections (e.g.
20 rejection of HELO). When verifying a recipient using a non-null MAIL
21 address, the cache is ignored if it shows MAIL FROM:<> rejection.
22 Whatever the result of the callout, the value of the domain cache is
23 left unchanged (for any other kind of callout, getting as far as trying
24 RCPT means that the domain itself is ok).
26 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
29 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
30 manifest itself as EPIPE rather than ECONNECT. When tidying away a
31 session, the daemon ignores ECONNECT errors and logs others; it now
32 ignores EPIPE as well.
34 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
35 (quoted-printable decoding).
37 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
38 later the small subsequent patch to fix an introduced bug.
40 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
42 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
44 PH/08 An error is now given if message_size_limit is specified negative.
46 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
47 to be given (somewhat) arbitrary names.
49 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
52 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
53 miscellaneous code fixes
55 PH/10 Added the log_reject_target ACL modifier to specify where to log
58 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
59 hostname. This is wrong, because it relates to the incoming message (and
60 probably the interface on which it is arriving) and not to the outgoing
61 callout (which could be using a different interface). This has been
62 changed to use the value of the helo_data option from the smtp transport
63 instead - this is what is used when a message is actually being sent. If
64 there is no remote transport (possible with a router that sets up host
65 addresses), $smtp_active_hostname is used.
67 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
68 tweaks were necessary in order to get it to work:
69 (a) The code assumed that strncpy() returns a negative number on buffer
70 overflow, which isn't the case. Replaced with Exim's string_format()
72 (b) There were several signed/unsigned issues. I just did the minimum
73 hacking in of casts. There is scope for a larger refactoring.
74 (c) The code used strcasecmp() which is not a standard C function.
75 Replaced with Exim's strcmpic() function.
76 (d) The code set only $1; it now sets $auth1 as well.
77 (e) A simple test gave the error "authentication client didn't specify
78 service in request". It would seem that Dovecot has changed its
79 interface. Fortunately there's a specification; I followed it and
80 changed what the client sends and it appears to be working now.
82 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
89 SC/01 Use a glob alias rather than an array ref in eximstats generated
90 parser. This improves both readability and performance.
92 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
93 Don't display local sender or destination tables in eximstats unless
94 there is data to show.
95 Added average volumes into the eximstats top table text output.
97 SC/03 Collect data on the number of addresses (recipients) as well
98 as the number of messages in eximstats.
100 TF/01 Correct an error in the documentation for the redirect router. Exim
101 does (usually) call initgroups() when daemonizing.
103 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
104 with consistent privilege compared to when running as a daemon.
106 TF/03 Note in the spec that $authenticated_id is not set for local
107 submissions from trusted users.
109 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
110 Thanks to Dean Brooks <dean@iglou.com> for the patch.
112 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
113 by adding some example configuration directives to the default
114 configuration file. A little bit of work is required to uncomment the
115 directives and define how usernames and passwords are checked, but
116 there is now a framework to start from.
118 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
119 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
120 without this. I don't know how relevant this is to other LDAP libraries.
122 PH/02 Add the verb name to the "unknown ACL verb" error.
124 PH/03 Magnus Holmgren's patch for filter_prepend_home.
126 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
128 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
129 directory not expanded when it should be if an expanded home directory
130 was set for the address (which is overridden by the transport).
132 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
135 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
136 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
137 because it is too late at that time, and has no effect.
139 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
140 security issue with \' (bugzilla #107). I could not use the
141 PQescapeStringConn() function, because it needs a PGconn value as one of
144 PH/08 When testing addresses using -bt, indicate those final addresses that
145 are duplicates that would not cause an additional delivery. At least one
146 person was confused, thinking that -bt output corresponded to deliveries.
147 (Suppressing duplicates isn't a good idea as you lose the information
148 about possibly different redirections that led to the duplicates.)
150 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
151 systems where poll() doesn't work, in particular OS X.
153 PH/10 Added more information to debugging output for retry time not reached.
155 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
156 operations in malware.c.
158 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
161 PH/13 If write_rejectlog was set false when logging was sent to syslog with
162 syslog_duplication set false, log lines that would normally be written
163 both the the main log and to the reject log were not written to syslog at
166 PH/14 In the default configuration, change the use of "message" in ACL warn
167 statements to "add_header".
169 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
170 not followed by a command (e.g. "seen endif").
172 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
173 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
176 PH/17 Added extra conditions to the default value of delay_warning_condition
180 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
181 { match{$h_precedence:}{(?i)bulk|list|junk} } \
182 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
185 The Auto-Submitted: and various List- headers are standardised, whereas I
186 don't think Precedence: ever was.
188 PH/18 Refactored debugging code in route_finduser() to show more information,
189 in particular, the error code if getpwnam() issues one.
191 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
192 This is apparently needed in addition to the PH/07 change above to avoid
193 any possible encoding problems.
195 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
196 but not after initializing Perl.
198 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
199 output them only if debugging. By default they are written stderr,
200 apparently, which is not desirable.
202 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
205 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
208 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
210 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
211 authenticated or an ident call has been made. Suppress the default
212 values for $authenticated_id and $authenticated_sender (but permit -oMai
213 and -oMas) when testing with -bh.
215 PH/24 Re-jigged the order of the tests in the default configuration so that the
216 tests for valid domains and recipients precede the DNS black list and CSA
217 tests, on the grounds that those ones are more expensive.
219 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
220 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
221 This bug exists in every version of Exim that I still have, right back to
224 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
225 However, an attempt to turn on case-sensitivity in a regex key by
226 including (?-i) didn't work because the subject string was already
227 lowercased, and the effects were non-intuitive. It turns out that a
228 one-line patch can be used to allow (?-i) to work as expected.
234 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
235 other effects) broke the use of negated acl sub-conditions.
237 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
240 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
241 "Deny" causes Exim to reject the incoming connection with a 554 error.
242 Unfortunately, if there is a major crisis, such as a disk failure,
243 tcp-wrappers gives "deny", whereas what one would like would be some
244 kind of temporary error. A kludge has been added to help with this.
245 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
246 554 error is used if errno is still zero or contains ENOENT (which occurs
247 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
250 PH/03 Add -lutil to the default FreeBSD LIBS setting.
252 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
253 errors. Otherwise a message that provokes a temporary error (when other
254 messages do not) can cause a whole host to time out.
256 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
257 addresses were routed directly to files or pipes from a redirect router.
258 File deliveries just didn't batch; pipe deliveries might have suffered
261 PH/06 A failure to get a lock for a hints database would erroneously always say
262 "Failed to get write lock", even when it was really a read lock.
264 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
265 of 0600. This has been changed to use the value of the lockfile_mode
266 option (which defaults to 0600).
268 PH/08 Applied small patch from the Sieve maintainer.
270 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
271 folder from quota calculations, a direct delivery into this folder messed
272 up the contents of the maildirsize file. This was because the regex was
273 used only to exclude .Trash (or whatever) when the size of the mailbox
274 was calculated. There was no check that a delivery was happening into an
275 excluded directory. This bug has been fixed by ignoring all quota
276 processing for deliveries into excluded directories.
278 PH/10 Added the maildirfolder_create_regex option to appendfile.
284 PH/01 The code for finding all the local interface addresses on a FreeBSD
285 system running IPv6 was broken. This may well have applied to all BSD
286 systems, as well as to others that have similar system calls. The broken
287 code found IPv4 interfaces correctly, but gave incorrect values for the
288 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
289 that it would not match correctly against @[] and not recognize the IPv6
292 PH/02 The ipliteral router was not recognizing addresses of the form user@
293 [ipv6:....] because it didn't know about the "ipv6:" prefix.
295 PH/03 Added disable_ipv6.
297 PH/04 Changed $reply_address to use the raw form of the headers instead of the
298 decoded form, because it is most often used to construct To: headers
299 lines in autoreplies, and the decoded form may well be syntactically
300 invalid. However, $reply_address has leading white space removed, and all
301 newlines turned into spaces so that the autoreply transport does not
304 PH/05 If group was specified without a user on a router, and no group or user
305 was specified on a transport, the group from the router was ignored.
307 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
308 for visible compile-time settings that can be used to change these
309 numbers, for those that want even more. Backwards compatibility with old
310 spool files has been maintained. However, going back to a previous Exim
311 release will lost any variables that are in spool files.
313 PH/07 Two small changes when running in the test harness: increase delay when
314 passing a TCP/IP connection to a new process, in case the original
315 process has to generate a bounce, and remove special handling of
316 127.0.0.2 (sic), which is no longer necessary.
318 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
319 be the same on different OS.
321 PH/09 Moved a debug statement in filter processing to avoid a race problem when
324 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
325 whether --show-vars was specified or not
327 JJ/02 exipick: Added support for new ACL variable spool format introduced
330 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
331 syntactically invalid From: or Reply-to: line, and a filter used this to
332 generate an autoreply, and therefore failed to obtain an address for the
333 autoreply, Exim could try to deliver to a non-existent relative file
334 name, causing unrelated and misleading errors. What now happens is that
335 it logs this as a hard delivery error, but does not attempt to create a
338 PH/11 The exinext utility has a -C option for testing purposes, but although
339 the given file was scanned by exinext itself; it wasn't being passed on
340 when Exim was called.
342 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
343 an end-of-file indication when reading a command response.
345 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
346 compiled. In many other places in Exim, IPv6 addresses are always
347 recognized, so I have changed this. It also means that IPv4 domain
348 literals of the form [IPV4:n.n.n.n] are now always recognized.
350 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
351 used if the router is not running as root, for example, when verifying at
352 ACL time, or when using -bh. The debugging output from this situation was
353 non-existent - all you got was a failure to exec. I have made two
356 (a) Failures to set uid/gid, the current directory, or a process leader
357 in a subprocess such as that created by queryprogram now generate
358 suitable debugging ouput when -d is set.
360 (b) The queryprogram router detects when it is not running as root,
361 outputs suitable debugging information if -d is set, and then runs
362 the subprocess without attempting to change uid/gid.
364 PH/15 Minor change to Makefile for building test_host (undocumented testing
367 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
368 additional section of a DNS packet that returns MX or SRV records.
369 Instead, it always explicitly searches for A/AAAA records. This avoids
370 major problems that occur when a DNS server includes only records of one
371 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
372 fixed another bug: if SRV records were looked up and the corresponding
373 address records were *not* found in the additional section, the port
374 values from the SRV records were lost.
376 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
377 using the correct key (the original address) when searching the retry
378 rules in order to find which one to use for generating the retry hint.
380 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
381 adding the default one. Similarly, if it contains a Reply-To: header, the
382 errors_reply_to option, if set, is not used.
384 PH/19 When calculating a retry time, Exim used to measure the "time since
385 failure" by looking at the "first failed" field in the retry record. Now
386 it does not use this if it is later than than the arrival time of the
387 message. Instead it uses the arrival time. This makes for better
388 behaviour in cases where some deliveries succeed, thus re-setting the
389 "first failed" field. An example is a quota failure for a huge message
390 when small messages continue to be delivered. Without this change, the
391 "time since failure" will always be short, possible causing more frequent
392 delivery attempts for the huge message than are intended.
393 [Note: This change was subsequently modified - see PH/04 for 4.62.]
395 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
396 $1, $2, $3) because the numerical variables can be reset during some
397 expansion items (e.g. "match"), thereby losing the authentication data.
399 PH/21 Make -bV show the size of off_t variables so that the test suite can
400 decide whether to run tests for quotas > 2G.
402 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
403 mailbox_size, and mailbox_filecount in the appendfile transport. If a
404 filecount value is greater than 2G or if a quota value is greater than 2G
405 on a system where the size of off_t is not greater than 4, a panic error
408 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
409 never match. The debug and -bh output now contains an explicit error
410 message indicating a malformed IPv4 address or mask.
412 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
413 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
416 PH/25 Do not write to syslog when running in the test harness. The only
417 occasion when this arises is a failure to open the main or panic logs
418 (for which there is an explicit test).
420 PH/26 Added the /no_tell option to "control=freeze".
422 PH/27 If a host name lookup failed very early in a connection, for example, if
423 the IP address matched host_lookup and the reverse lookup yielded a name
424 that did not have a forward lookup, an error message of the form "no IP
425 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
426 could be logged. Now it outputs the IP address instead of "NULL".
428 PH/28 An enabling patch from MH: add new function child_open_exim2() which
429 allows the sender and the authenticated sender to be set when
430 submitting a message from within Exim. Since child_open_exim() is
431 documented for local_scan(), the new function should be too.
433 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
434 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
435 results in an empty string is now treated as unset.
437 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
439 PH/31 Added sender_verify_fail logging option.
441 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
442 needed by RFC 822 but not by RFC 2822 was commented out. I have now
443 tidied the source and removed it altogether.
445 PH/33 When a queue run was abandoned because the load average was too high, a
446 log line was always written; now it is written only if the queue_run log
447 selector is set. In addition, the log line for abandonment now contains
448 information about the queue run such as the pid. This is always present
449 in "start" and "stop" lines but was omitted from the "abandon" line.
451 PH/34 Omit spaces between a header name and the colon in the error message that
452 is given when verify = headers_syntax fails (if there are lots of them,
453 the message gets confusing).
455 PH/35 Change the default for dns_check_names_pattern to allow slashes within
456 names, as there are now some PTR records that contain slashes. This check
457 is only to protect against broken name servers that fall over on strange
458 characters, so the fact that it applies to all lookups doesn't matter.
460 PH/36 Now that the new test suite is complete, we can remove some of the
461 special code in Exim that was needed for the old test suite. For example,
462 sorting DNS records because real resolvers return them in an arbitrary
463 order. The new test suite's fake resolver always returns records in the
466 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
467 bounces) except when queue_only is set, to avoid logging races between
468 the different processes.
470 PH/38 Panic-die if .include specifies a non-absolute path.
472 PH/39 A tweak to the "H" retry rule from its user.
474 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
475 a label. They prevented compilation on older perls.
477 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
478 a warning to be raised on newish perls.
480 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
481 on queue. Changes to match documented behaviour of showing count of
482 messages matching specified criteria.
484 PH/40 Changed the default ident timeout from 30s to 5s.
486 PH/41 Added support for the use of login_cap features, on those BSD systems
487 that have them, for controlling the resources used by pipe deliveries.
489 PH/42 The content-scanning code uses fopen() to create files in which to put
490 message data. Previously it was not paying any attention to the mode of
491 the files. Exim runs with umask(0) because the rest of the code creates
492 files with open(), and sets the required mode explicitly. Thus, these
493 files were ending up world-writeable. This was not a big issue, because,
494 being within the spool directory, they were not world-accessible. I have
495 created a function called modefopen, which takes an additional mode
496 argument. It sets umask(777), creates the file, chmods it to the required
497 mode, then resets the umask. All the relevant calls to fopen() in the
498 content scanning code have been changed to use this function.
500 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
501 to 24 hours. This avoids potential overflow problems when processing G
502 and H retry rules. I suspect nobody ever tinkers with this value.
504 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
506 PH/45 When the plaintext authenticator is running as a client, the server's
507 challenges are checked to ensure they are valid base64 strings. By
508 default, the authentication attempt is cancelled if an invalid string is
509 received. Setting client_ignore_invalid_base64 true ignores these errors.
510 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
511 they are received. Thus, the responses can be made to depend on the
512 challenges. If an invalid string is ignored, an empty string is placed in
515 PH/46 Messages that are created by the autoreply transport now contains a
516 References: header, in accordance with RFCs 2822 and 3834.
518 PH/47 Added authenticated_sender_force to the smtp transport.
520 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
522 PH/49 Installed latest patch from the Sieve maintainer.
524 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
525 was also set, the appendfile transport was unnecessarily scanning a
526 directory of message files (e.g. for maildir delivery) to find the count
527 of files (along with the size), even though it did not need this
528 information. It now does the scan only if it needs to find either the
529 size of the count of files.
531 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
533 PH/52 Two bugs concerned with error handling when the smtp transport is
536 (i) Exim was not creating retry information for temporary errors given
537 for individual recipients after the DATA command when the smtp transport
538 was used in LMTP mode. This meant that they could be retried too
539 frequently, and not timed out correctly.
541 (ii) Exim was setting the flag that allows error details to be returned
542 for LMTP errors on RCPT commands, but not for LMTP errors for individual
543 recipients that were returned after the DATA command.
545 PH/53 This is related to PH/52, but is more general: for any failing address,
546 when detailed error information was permitted to be returned to the
547 sender, but the error was temporary, then after the final timeout, only
548 "retry timeout exceeded" was returned. Now it returns the full error as
549 well as "retry timeout exceeded".
551 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
552 do this, and (what is worse) MTAs that accept it.
554 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
555 will now be deprecated.
557 PH/56 New os.c-cygwin from the Cygwin maintainer.
559 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
560 formats (previously only available in exim formats via -bpr, -bpru,
561 and -bpra. Now also available in native and exiqgrep formats)
563 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
564 with very large, slow to parse queues
566 JJ/08 exipick: added ! as generic prefix to negate any criteria format
568 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
570 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
571 responses to authentication challenges, though it was showing the
572 challenges; (ii) I've removed the CR characters from the debug output for
575 PH/58 Allow for the insertion of a newline as well as a space when a string
576 is turned into more than one encoded-word during RFC 2047 encoding. The
577 Sieve code now uses this.
579 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
580 data_4xx, lost_connection, tls_required.
582 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
583 message was being sent as an SMTP response.
585 PH/61 Add -l and -k options to exicyclog.
587 PH/62 When verifying, if an address was redirected to one new address, so that
588 verification continued, and the new address failed or deferred after
589 having set something in $address_data, the value of $address_data was not
590 passed back to the ACL. This was different to the case when no
591 redirection occurred. The value is now passed back in both cases.
593 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
594 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
595 use login_cap.h, so on its own it isn't the distinguishing feature. The
596 new name refers directly to the setclassresources() function.
598 PH/65 Added configuration files for NetBSD3.
600 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
602 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
603 is preferred over IPv4.
605 PH/68 The bounce_return_message and bounce_return_body options were not being
606 honoured for bounces generated during the reception of non-SMTP messages.
607 In particular, this applied to messages rejected by the ACL. This bug has
608 been fixed. However, if bounce_return_message is true and bounce_return_
609 body is false, the headers that are returned for a non-SMTP message
610 include only those that have been read before the error was detected.
611 (In the case of an ACL rejection, they have all been read.)
613 PH/69 The HTML version of the specification is now built in a directory called
614 spec_html instead of spec.html, because the latter looks like a path with
615 a MIME-type, and this confuses some software.
617 PH/70 Catch two compiler warnings in sieve.c.
619 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
620 function verify_get_ident() calls ip_connect() to connect a socket, but
621 if the "connect()" function timed out, ip_connect() used to close the
622 socket. However, verify_get_ident() also closes the socket later, and in
623 between Exim writes to the log, which may get opened at this point. When
624 the socket was closed in ip_connect(), the log could get the same file
625 descriptor number as the socket. This naturally causes chaos. The fix is
626 not to close the socket in ip_connect(); the socket should be closed by
627 the function that creates it. There was only one place in the code where
628 this was missing, in the iplookup router, which I don't think anybody now
629 uses, but I've fixed it anyway.
631 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
632 well as to direct DNS lookups. Otherwise the handling of names in host
633 lists is inconsistent and therefore confusing.
639 PH/01 Two changes to the default runtime configuration:
641 (1) Move the checks for relay_from_hosts and authenticated clients from
642 after to before the (commented out) DNS black list checks.
644 (2) Add control=submission to the relay_from_hosts and authenticated
645 clients checks, on the grounds that messages accepted by these
646 statements are most likely to be submissions.
648 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
650 (1) Generate an error if the third argument for the ${prvs expansion is
653 (2) Treat a missing third argument of ${prvscheck as if it were an empty
656 (3) Reset the variables that are obtained from the first argument of
657 ${prvscheck and used in the second argument before leaving the code,
658 because their memory is reclaimed, so using them afterwards may do
661 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
662 one (it's much easier than Tom thought :-).
664 (5) Because of (4), we can now allow for the use of $prvscheck_result
665 inside the third argument.
667 PH/03 For some reason, the default setting of PATH when running a command from
668 a pipe transport was just "/usr/bin". I have changed it to
671 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
672 anything to be listed in the output from -bV.
674 PH/05 When a filter generated an autoreply, the entire To: header line was
675 quoted in the delivery log line, like this:
677 => >A.N.Other <ano@some.domain> <original@ddress> ...
679 This has been changed so that it extracts the operative address. There
680 may be more than one such address. If so, they are comma-separated, like
683 => >ano@some.domain,ona@other.domain <original@ddress> ...
685 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
686 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
687 not being looked up in the rDNS to get a host name, Exim was showing the
688 IP address twice in Received: lines, even though the IP addresses were
689 identical. For example:
691 Received: from [1.2.3.4] (helo=[1.2.3.4])
693 However, if the real host name was known, it was omitting the HELO data
694 if it matched the actual IP address. This has been tidied up so that it
695 doesn't show the same IP address twice.
697 PH/07 When both +timestamp and +memory debugging was on, the value given by
698 $tod_xxx expansions could be wrong, because the tod_stamp() function was
699 called by the debug printing, thereby overwriting the timestamp buffer.
700 Debugging no longer uses the tod_stamp() function when +timestamp is set.
702 PH/08 When the original message was included in an autoreply transport, it
703 always said "this is a copy of the message, including all the headers",
704 even if body_only or headers_only was set. It now gives an appropriate
707 PH/09 Applied a patch from the Sieve maintainer which:
709 o fixes some comments
710 o adds the (disabled) notify extension core
711 o adds some debug output for the result of if/elsif tests
712 o points to the current vacation draft in the documentation
713 and documents the missing references header update
717 o fixes a bug in processing the envelope test (when testing
718 multiple envelope elements, the last element determinted the
721 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
722 Electronic Mail") by including:
724 Auto-submitted: auto-generated
726 in the messages that it generates (bounce messages and others, such as
727 warnings). In the case of bounce messages for non-SMTP mesages, there was
728 also a typo: it was using "Auto_submitted" (underscore instead of
729 hyphen). Since every message generated by Exim is necessarily in response
730 to another message, thes have all been changed to:
732 Auto-Submitted: auto-replied
734 in accordance with these statements in the RFC:
736 The auto-replied keyword:
738 - SHOULD be used on messages sent in direct response to another
739 message by an automatic process,
741 - MUST NOT be used on manually-generated messages,
743 - MAY be used on Delivery Status Notifications (DSNs) and Message
744 Disposition Notifications (MDNs),
746 - MUST NOT be used on messages generated by automatic or periodic
747 processes, except for messages which are automatic responses to
750 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
751 to the default Received: header definition.
753 PH/12 Added log selector acl_warn_skipped (default on).
755 PH/13 After a successful wildlsearch lookup, discard the values of numeric
756 variables because (a) they are in the wrong storage pool and (b) even if
757 they were copied, it wouldn't work properly because of the caching.
759 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
760 checking when decoding. Apparently there are clients that generate
761 overlong encoded strings. Why am I not surprised?
763 PH/15 If the first argument of "${if match_address" was not empty, but did not
764 contain an "@" character, Exim crashed. Now it writes a panic log message
765 and treats the condition as false.
767 PH/16 In autoreply, treat an empty string for "once" the same as unset.
769 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
770 extension "envelope-auth". The code is finished and in agreement with
771 other implementations, but there is no documentation so far and in fact,
772 nobody wrote the draft yet. This extension is currently #undef'ed, thus
773 not changing the active code.
775 Print executed "if" and "elsif" statements when debugging is used. This
776 helps a great deal to understand what a filter does.
778 Document more things not specified clearly in RFC3028. I had all this
779 sorted out, when out of a sudden new issues came to my mind. Oops."
781 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
784 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
785 canonical form (as documented). However, after a host name lookup from
786 the IP address, check_host() was doing a simple string comparison with
787 addresses acquired from the DNS when checking that the found name did
788 have the original IP as one of its addresses. Since any found IPv6
789 addresses are likely to be in abbreviated form, the comparison could
790 fail. Luckily, there already exists a function for doing the comparison
791 by converting both addresses to binary, so now that is used instead of
794 PH/20 There was another similar case to PH/19, when a complete host name was
795 given in a host list; looking up its IP address could give an abbreviated
796 form, whereas the current host's name might or might not be abbreviated.
797 The same fix has been applied.
803 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
804 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
807 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
808 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
810 PH/03 Typo: missing ".o" in src/pcre/Makefile.
812 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
813 header line, restrict the check to what is listed in RFCs 2369 and 2929.
814 Also, for "Auto-Submitted", treat anything other than "no" as
815 non-personal, in accordance with RFC 3834. (Previously it treated
816 anything starting "auto-" as non-personal.)
818 TF/01 The control=submission/name=... option had a problem with syntax
819 errors if the name included a slash character. The /name= option
820 now slurps the rest of the string, so it can include any characters
821 but it must come last in the list of options (after /sender_retain
824 PH/05 Some modifications to the interface to the fake nameserver for the new
832 TK/01 Added the "success_on_redirect" address verification option. See
833 NewStuff for rationale and an example.
835 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
837 PH/02 Patch to exigrep to allow it to work on syslog lines.
839 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
840 fread() to skip over the body file's header line, because in Cygwin the
841 header line is locked and is inaccessible.
843 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
844 co-exist for some time) to make it clear that it is the Exim ID that is
845 referenced, not the Message-ID: header line.
847 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
848 string_format() function, because snprintf() does not exist on all
851 PH/06 The use of forbid_filter_existstest now also locks out the use of the
852 ${stat: expansion item.
854 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
855 protocol synchronization error", to keep the pedants happy.
857 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
858 well as for IRIX systems, when gcc is being used. See the host.c source
861 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
863 PH/10 Named domain lists were not working if used in a queue_smtp_domains
866 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
867 transport and to the smtp transport in LMTP mode.
869 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
871 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
872 run a filter in a subprocess. This could lead to confusion in subsequent
873 lookups in the parent process. There should also be a search_tidyup() at
874 the end of the subprocess.
876 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
877 only if the host matched helo_try_verify_hosts, which caused the
878 verification to occur when the EHLO/HELO command was issued. The ACL just
879 tested the remembered result. Now, if a previous verification attempt has
880 not happened, "verify = helo" does it there and then.
882 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
884 TK/03 Fix log output including CR from clamd.
886 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
887 exist provoked a memory error which could cause a segfault.
889 PH/15 Installed PCRE 6.2
891 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
893 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
894 of the problem. Specifically, suggested +O2 rather than +O1 for the
897 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
899 PH/20 If a delivery was routed to a non-standard port by means of an SRV
900 record, the port was not correctly logged when the outgoing_port log
901 selector was set (it logged the transort's default port).
903 PH/21 Added support for host-specific ports to manualroute, queryprogram,
904 fallback_hosts, and "hosts" in the smtp transport.
906 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
907 host errors such as "Connection refused".
909 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
910 authentication with radiusclient 0.4.9:
912 - Error returned from rc_read_config was caught wrongly
913 - Username/password not passed on to radius server due to wrong length.
915 The presumption is that some radiusclient API changes for 4.51/PH/17
916 were not taken care of correctly. The code is still untested by me (my
917 Linux distribution still has 0.3.2 of radiusclient), but it was
918 contributed by a Radius user.
920 PH/24 When doing a callout, the value of $domain wasn't set correctly when
921 expanding the "port" option of the smtp transport.
923 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
924 while reading a MIME header. Thanks to Tom Hughes for a patch.
926 PH/24 Include config.h inside local_scan.h so that configuration settings are
929 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
930 that in an ACL for RCPT (for example), you can examine exactly what was
933 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
934 commands, but it was not correctly comparing the address with the actual
935 client host address. Thus, it would show the EHLO address in Received:
936 header lines when this was not necessary.
938 PH/27 Added the % operator to ${eval:}.
940 PH/28 Exim tries to create and chdir to its spool directory when it starts;
941 it should be ignoring failures (because with -C, for example, it has lost
942 privilege). It wasn't ignoring creation failures other than "already
945 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
948 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
949 because an input line was too long, either on its own, or by virtue of
950 too many continuations, the temporary file was not being removed, and the
951 return code was incorrect.
953 PH/31 Missing "BOOL" in function definition in filtertest.c.
955 PH/32 Applied Sieve patches from the maintainer.
957 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
959 PH/33 Added "verify = not_blind".
961 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
962 Local/Makefile (with some defaults set). These are used in built scripts
963 such as exicyclog, but they have never been used in the exim_install
964 script (though there are many overriding facilities there). I have
965 arranged that the exim_install script now takes note of these two
968 PH/35 Installed configuration files for Dragonfly.
970 PH/36 When a locally submitted message by a trusted user did not contain a
971 From: header, and the sender address was obtained from -f or from an SMTP
972 MAIL command, and the trusted user did not use -F to supply a sender
973 name, $originator_name was incorrectly used when constructing a From:
974 header. Furthermore, $originator_name was used for submission mode
975 messages from external hosts without From: headers in a similar way,
976 which is clearly wrong.
978 PH/37 Added control=suppress_local_fixups.
980 PH/38 When log_selector = +received_sender was set, and the addition of the
981 sender made the log line's construction buffer exactly full, or one byte
982 less than full, an overflow happened when the terminating "\n" was
985 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
986 when the result of a list match is failure because a DNS lookup failed.
988 PH/40 RM_COMMAND is now used in the building process.
990 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
991 the "build-* directories that it finds.
993 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
994 address in a domain literal was a prefix of an interface address.
996 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
997 when verifying a sender address, unless rewrite_headers is false.
999 PH/44 Wrote a long comment about why errors_to addresses are verified as
1000 recipients, not senders.
1002 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1003 the ratelimit ACL was added.
1005 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1007 PH/46 Added extra information about PostgreSQL errors to the error string.
1009 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1010 avoiding the need to install special zones in a real server. This is
1011 backwards compatible; if it can't find the fake resolver, it drops back.
1012 Thus, both old and new test suites can be run.
1014 TF/02 Added util/ratelimit.pl
1016 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1017 clock is set back in time.
1019 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1020 Candler <B.Candler@pobox.com>.
1022 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1023 OK for addresses that are the result of redirections.
1025 PH/48 A number of further additions for the benefit of the new test suite,
1026 including a fake gethostbyname() that interfaces to the fake DNS resolver
1029 TF/06 The fix for widen_domains has also been applied to qualify_single and
1030 search_parents which are the other dnslookup options that can cause
1033 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1036 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1038 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1039 service" if it encountered an MX record with an empty target hostname.
1040 The message is now "an MX or SRV record indicated no SMTP service".
1042 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1043 if the DNS of the sending site is misconfigured. This is quite a
1044 common situation. This change restores the behaviour of treating a
1045 helo verification defer as a failure.
1047 PH/51 If self=fail was set on a router, the bounce message did not include the
1048 actual error message.
1054 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1056 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1057 command itself ended in error, the underlying message about the transport
1058 filter timeout was being overwritten with the pipe command error. Now the
1059 underlying error message should be appended to the second error message.
1061 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1063 PH/02 Reduce the amount of output that "make" produces by default. Full output
1064 can still be requested.
1066 PH/03 The warning log line about a condition test deferring for a "warn" verb
1067 was being output only once per connection, rather than after each
1068 occurrence (because it was using the same function as for successful
1069 "warn" verbs). This seems wrong, so I have changed it.
1071 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1072 it should not have, which might have caused a crash in the right
1073 circumstances, but probably never did.
1075 PH/04 Installed a modified version of Tony Finch's patch to make submission
1076 mode fix the return path as well as the Sender: header line, and to
1077 add a /name= option so that you can make the user's friendly name appear
1080 TF/03 Added the control = fakedefer ACL modifier.
1082 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1083 Mark Lowes for thorough testing.
1085 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1087 TK/03 Merged latest SRS patch from Miles Wilton.
1089 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1090 with the definition in sysexits.h (which is #included earlier).
1091 Fortunately, Exim does not actually use EX_OK. The code used to try to
1092 preserve the sysexits.h value, by assumimg that macro definitions were
1093 scanned for macro replacements. I have been disabused of this notion,
1094 so now the code just undefines EX_OK before #including unistd.h.
1096 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1097 in the smtp transport. When a block could not be written in a single
1098 write() function, the timeout was being re-applied to each part-write.
1099 This seems wrong - if the receiver was accepting one byte at a time it
1100 would take for ever. The timeout is now adjusted when this happens. It
1101 doesn't have to be particularly precise.
1103 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1104 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1106 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1107 without a domain if the check to <postmaster@domain> fails.
1109 SC/01 Eximstats: added -xls and the ability to specify output files
1110 (patch written by Frank Heydlauf).
1112 SC/02 Eximstats: use FileHandles for outputing results.
1114 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1116 SC/04 Eximstats: fixed display of large numbers with -nvr option
1118 SC/05 Eximstats: fixed merging of reports with empty tables.
1120 SC/06 Eximstats: added the -include_original_destination flag
1122 SC/07 Eximstats: removed tabs and trailing whitespace.
1124 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1126 TK/06 MBOX spool code: Add real "From " MBOX separator line
1127 so the .eml file is really in mbox format (even though
1128 most programs do not really care). Patch from Alex Miller.
1130 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1131 The latter is generated from $received_to and is only set if the
1132 message has one envelope recipient. SA can use these headers,
1133 obviously out-of-the-box. Patch from Alex Miller.
1135 PH/08 The ${def test on a variable was returning false if the variable's
1136 value was "0", contrary to what the specification has always said!
1137 The result should be true unless the variable is empty.
1139 PH/09 The syntax error of a character other than { following "${if
1140 def:variable_name" (after optional whitespace) was not being diagnosed.
1141 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1142 accidental colon was present, for example, could give incorrect results.
1144 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1145 result is used (not including appendfile, where other changes are about
1148 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1149 This involved changing a lot of size variables from int to off_t. It
1150 should work with maildirs and everything.
1152 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1153 spamd dying while we are connected to it.
1155 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1158 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1159 function for BDB 4.3.
1161 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1163 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1164 the sources for PCRE have been re-arranged and more files are now
1167 PH/15 The code I had for printing potentially long long variables in PH/11
1168 above was not the best (it lost precision). The length of off_t variables
1169 is now inspected at build time, and an appropriate printing format (%ld
1170 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1171 to be "long long int" or "long int". This is needed for the internal
1172 formatting function string_vformat().
1174 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1175 the configuration file to be ":syslog", then the script "guesses" where
1176 the logs files are, rather than using the compiled in default. In our
1177 case the guess is not the same as the compiled default, so the script
1178 suddenly stopped working when I started to use syslog. The patch checks
1179 to see if log_file_path is "". If so, it attempts to read it from exim
1180 with no configuration file to get the compiled in version, before it
1181 falls back to the previous guessing code."
1183 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1184 implementing BATV in an Exim configuration. See NewStuff for the gory
1187 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1188 Makefile that are specific to HP-UX.
1190 PH/18 If the "use_postmaster" option was set for a recipient callout together
1191 with the "random" option, the postmaster address was used as the MAIL
1192 FROM address for the random test, but not for the subsequent recipient
1193 test. It is now used for both.
1195 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1196 patch removes a few documentation additions to RFC 3028, because the
1197 latest draft now contains them. It adds the new en;ascii-case comparator
1198 and a new error check for 8bit text in MIME parts. Comparator and
1199 require names are now matched exactly. I enabled the subaddress
1200 extension, but it is not well tested yet (read: it works for me)."
1202 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1203 rework some of the code of TK/09 above to avoid the hardwired use of
1204 "%lld" and "long long". Replaced the call to snprintf() with a call to
1207 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1208 records point to non-existent hosts", "retry timeout exceeded", and
1209 "retry time not reached for any host after a long failure period".
1211 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1212 experimental DomainKeys support:
1214 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1215 (2) On an error such as an illegally used "control", the wrong name for
1216 the control was given.
1218 These problems did NOT occur unless DomainKeys support was compiled.
1220 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1222 PH/24 Added ${if match_ip condition.
1224 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1225 databases so that it will be absolutely obvious if a crash occurs in the
1226 DB library. This is a regular occurrence (often caused by mis-matched
1229 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1230 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1231 picked up on a user's system that detects such things. There doesn't seem
1232 to be a gcc warning option for this - only an attribute that has to be
1233 put on the function's prototype. It seems that in Fedora Core 4 they have
1234 set this on a number of new functions. No doubt there will be more in due
1237 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1238 specify a transport. However, if an address that was verified by such a
1239 router was the subject of a callout, Exim crashed because it tried to
1240 read the rcpt_include_affixes from the non-existent transport. Now it
1241 just assumes that the setting of that option is false. This bug was
1242 introduced by 4.51/PH/31.
1244 PH/28 Changed -d+all to exclude +memory, because that information is very
1245 rarely of interest, but it makes the output a lot bigger. People tend to
1246 do -d+all out of habit.
1248 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1249 code in os-type was giving problems when libc.so lives in lib64, like on
1252 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1253 aren't the modern standard, and it seems that some systems' include files
1254 don't always have them. Exim was already checking for some of the newer
1255 ones like T_AAAA, and defining it itself. I've added checks for all the
1256 record types that Exim uses.
1258 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1259 not automatically generating a new one, as it is supposed to. This
1260 prevented TLS from working. If the file did exist, but contained invalid
1261 data, a new version was generated, as expected. It was only the case of a
1262 non-existent file that was broken.
1264 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1265 with a change in libdomainkeys > 0.64.
1267 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1268 from DNS. If the selector record carries the flag, it now has
1269 precedence over the domain-wide flag.
1271 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1273 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1274 the use of a port name that isn't defined in /etc/services) occurred, the
1275 message was deferred as in a normal delivery, and thus remained on the
1276 spool, instead of being failed because of the mua_wrapper setting. This
1277 is now fixed, and I tidied up some of the mua_wrapper messages at the
1280 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1281 the messages in a hash of arrays rather than using individual hashes.
1282 This is a bit cleaner and results in dramatic memory savings, albeit
1283 at a slight CPU cost.
1285 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1286 as requested by Marc Sherman.
1288 SC/10 Eximstats: added histograms for user specified patterns as requested
1291 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1293 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1294 fopen() in the content-scanning modules that did not already have it.
1300 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1301 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1303 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1305 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1307 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1308 to test Sieve filters that use "vacation".
1310 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1311 that changes the way the GnuTLS parameters are stored in the cache file.
1312 The new format can be generated externally. For backward compatibility,
1313 if the data in the cache doesn't make sense, Exim assumes it has read an
1314 old-format file, and it generates new data and writes a new file. This
1315 means that you can't go back to an older release without removing the
1318 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1319 work if there are any delivery delays because "one_time" forces the
1320 parent to be marked "delivered", so its unseen clone is never tried
1321 again. For this reason, Exim now forbids the simultaneous setting of
1324 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1325 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1326 entry if you want to know the details. The fix, however, neglected to
1327 consider the case when local delivery batching is involved. The test for
1328 "previously delivered" was not happening when checking to see if an
1329 address could be batched with a previous (undelivered) one; under
1330 certain circumstances this could lead to multiple deliveries to the same
1333 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1334 in its include files, and this causes problems building Exim.
1336 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1337 header_syntax) but Exim was just ignoring anything given after a slash.
1338 In particular, this caused confusion with an attempt to use "verify =
1339 reverse_host_lookup/defer_ok". An error is now given when options are
1340 supplied for verify items that do not have them. (Maybe reverse_host_
1341 lookup should have a defer_ok option, but that's a different point.)
1343 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1344 defined by RFC 821) to 2048, because there were problems with some AUTH
1345 commands, and RFC 1869 says the size should be increased for extended
1346 SMTP commands that take arguments.
1348 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1351 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1352 "unknown" error; now it says that the functionality isn't in the binary.
1354 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1355 an address' error message when a string expansion fails (syntax or
1356 whatever). Otherwise the password may appear in the log. Following change
1357 PH/42 below, there is no longer a chance of it appearing in a bounce
1360 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1362 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1363 of its addresses was ever tried. (Bugzilla bug #2.)
1365 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1366 the result incorrectly in the debug output. (It correctly added a newline
1367 to what was transported.)
1369 TF/01 Added $received_time.
1371 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1372 commented out examples of how to interface to a virus scanner and to
1373 SpamAssassin. Also added commented examples of av_scanner and
1374 spamd_address settings.
1376 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1377 and controls are allowed in which ACLs. There were a couple of minor
1378 errors. Some of the entries in the conditions table (which is a table of
1379 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1380 negation of where the condition IS allowed.
1382 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1384 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1385 header file does not have a version number, so I've had to invent a new
1386 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1387 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1388 radiusclient), but it was contributed by a Radius user.
1390 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1391 files or directories, for OpenSSL.
1393 PH/19 When an Exim process that is running as root has to create an Exim log
1394 file, it does so in a subprocess that runs as exim:exim so as to get the
1395 ownership right at creation (otherwise, other Exim processes might see
1396 the file with the wrong ownership). There was no test for failure of this
1397 fork() call, which would lead to the process getting stuck as it waited
1398 for a non-existent subprocess. Forks do occasionally fail when resources
1399 run out. I reviewed all the other calls to fork(); they all seem to check
1402 PH/20 When checking for unexpected SMTP input at connect time (before writing
1403 the banner), Exim was not dealing correctly with a non-positive return
1404 from the read() function. If the client had disconnected by this time,
1405 the result was a log entry for a synchronization error with an empty
1406 string after "input=" when read() returned zero. If read() returned -1
1407 (an event I could not check), uninitialized data bytes were printed.
1408 There were reports of junk text (parts of files, etc) appearing after
1411 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1413 PH/22 Added support for macro redefinition, and (re)definition in between
1414 driver and ACL definitions.
1416 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1417 forgetting to use the resulting value; it was using the unexpanded value.
1419 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1420 hadn't been configured. The fix is from Juergen Kreileder, who
1421 understands it better than I do:
1423 "Here's what I see happening with three configured cyrus_sasl
1424 authenticators configured (plain, login, cram-md5):
1426 On startup auth_cyrus_sasl_init() gets called for each of these.
1427 This means three calls to sasl_listmech() without a specified mech_list.
1428 => SASL tests which mechs of all available mechs actually work
1429 => three warnings about OTP not working
1430 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1432 With the patch, sasl_listmech() also gets called three times. But now
1433 SASL's mech_list option is set to the server_mech specified in the the
1434 authenticator. Or in other words, the answer from sasl_listmech()
1435 gets limited to just the mech you're testing for (which is different
1437 => the return list contains just 'plain' or 'login', 'cram-md5' or
1438 nothing depending on the value of ob->server_mech.
1440 I've just tested the patch: Authentication still works fine,
1441 unavailable mechs specified in the exim configuration are still
1442 caught, and the auth.log warnings about OTP are gone."
1444 PH/25 When debugging is enabled, the contents of the command line are added
1445 to the debugging output, even when log_selector=+arguments is not
1448 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1449 answer is "GNU", and only if the return is "GNU/something" is the answer
1452 PH/27 $acl_verify_message is now set immediately after the failure of a
1453 verification in an ACL, and so is available in subsequent modifiers. In
1454 particular, the message can be preserved by coding like this:
1456 warn !verify = sender
1457 set acl_m0 = $acl_verify_message
1459 Previously, $acl_verify_message was set only while expanding "message"
1460 and "log_message" when a very denied access.
1462 PH/28 Modified OS/os.c-Linux with
1464 -#ifndef OS_LOAD_AVERAGE
1465 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1467 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1468 nomenclature these days.)
1470 PH/29 Installed patch from the Sieve maintainer that adds the options
1471 sieve_useraddress and sieve_subaddress to the redirect router.
1473 PH/30 In these circumstances:
1474 . Two addresses routed to the same list of hosts;
1475 . First host does not offer TLS;
1476 . First host accepts first address;
1477 . First host gives temporary error to second address;
1478 . Second host offers TLS and a TLS session is established;
1479 . Second host accepts second address.
1480 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1481 and peerdn, if requested) that were in fact used only for the second
1484 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1485 attention to any local part prefix or suffix that was matched by the
1486 router that accepted the address. It now behaves in the same way as it
1487 does for delivery: the affixes are removed from the local part unless
1488 rcpt_include_affixes is set on the transport.
1490 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1491 timeout during the DATA phase of an incoming message.
1493 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1494 applied a patch sanctioned by the Sieve maintainer.
1496 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1497 the uid or gid is negative. A case of a negative gid caused this to be
1498 noticed. The fix allows for either to be negative.
1500 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1501 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1504 PH/36 The change PH/12 above was broken. Fixed it.
1506 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1507 the grounds that routing the same address twice would always produce the
1508 same answer. This might have been true once, but it is certainly no
1509 longer true now. Routing a child address may depend on the previous
1510 routing that produced that child. Some complicated redirection strategies
1511 went wrong when messages had multiple recipients, and made Exim's
1512 behaviour dependent on the order in which the addresses were given.
1514 I have moved the duplicate checking until after the routing is complete.
1515 Exim scans the addresses that are assigned to local and remote
1516 transports, and removes any duplicates. This means that more work will be
1517 done, as duplicates will always all be routed, but duplicates are
1518 presumably rare, so I don't expect this is of any significance.
1520 For deliveries to pipes, files, and autoreplies, the duplicate checking
1521 still happens during the routing process, since they are not going to be
1524 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1525 It corrects a timeout issue with spamd. This is Ian's comment: "The
1526 background is that sometimes spamd either never reads data from a
1527 connection it has accepted, or it never writes response data. The exiscan
1528 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1529 blindly assumes that writes won't block so it may never time out."
1531 PH/39 Allow G after quota size as well as K and M.
1533 PH/40 The value set for $authenticated_id in an authenticator may not contain
1534 binary zeroes or newlines because the value is written to log lines and
1535 to spool files. There was no check on this. Now the value is run through
1536 the string_printing() function so that such characters are converted to
1537 printable escape sequences.
1539 PH/41 $message_linecount is a new variable that contains the total number of
1540 lines in the message. Compare $body_linecount, which is the count for the
1543 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1544 bounce and delay warning messages, except in certain special cases, which
1547 (a) An SMTP error message from a remote host;
1548 (b) A message specified in a :fail: redirection;
1549 (c) A message specified in a "fail" command in a system filter;
1550 (d) A message specified in a FAIL return from the queryprogram router;
1551 (e) A message specified by the cannot_route_message router option.
1553 In these cases only, Exim does include the error details in bounce and
1554 warning messages. There are also a few cases where bland messages such
1555 as "unrouteable address" or "local delivery error" are given.
1557 PH/43 $value is now also set for the "else" part of a ${run expansion.
1559 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1560 being worked on, but at least Exim now implements the latest version to
1563 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1564 process to complete was treated as a delivery failure, a timeout while
1565 writing the message to the pipe was logged, but erroneously treated as a
1566 successful delivery. Such timeouts include transport filter timeouts. For
1567 consistency with the overall process timeout, these timeouts are now
1568 treated as errors, giving rise to delivery failures by default. However,
1569 there is now a new Boolean option for the pipe transport called
1570 timeout_defer, which, if set TRUE, converts the failures into defers for
1571 both kinds of timeout. A transport filter timeout is now identified in
1574 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1575 systems where "make" and "gmake" are different, calling "gmake" at top
1576 level broke things. I've arranged for the value of $(MAKE) to be passed
1577 from the Makefile to this script so that it can call the same version of
1581 A note about Exim versions 4.44 and 4.50
1582 ----------------------------------------
1584 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1585 changes of various kinds. As a consequence, a big documentation update was
1586 needed. This delayed the release for rather longer than seemed good, especially
1587 in the light of a couple of (minor) security issues. Therefore, the changes
1588 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1589 release. So 4.44 and 4.50 are in effect two different branches that both start
1592 I have left the 4.50 change log unchanged; it contains all the changes since
1593 4.43. The change log for 4.44 is below; many of its items are identical to
1594 those for 4.50. This seems to be the most sensible way to preserve the
1595 historical information.
1601 1. Minor wording change to the doc/README.SIEVE file.
1603 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1604 computation of the current number of files was incorrect.
1606 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1607 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1610 4. Give more explanation in the error message when the command for a transport
1611 filter fails to execute.
1613 5. There are several places where Exim runs a non-Exim command in a
1614 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1615 was being done only for the command run by the queryprogram router. It is
1616 now done for all such subprocesses. The other cases are: ${run, transport
1617 filters, and the commands run by the lmtp and pipe transports.
1619 6. Added CONFIGURE_GROUP build-time option.
1621 7. Some older OS have a limit of 256 on the maximum number of file
1622 descriptors. Exim was using setrlimit() to set 1000 as a large value
1623 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1624 systems. I've change it so that if it can't get 1000, it tries for 256.
1626 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1627 was an oversight, and furthermore, ever since the addition of extra
1628 controls (e.g. 4.43/32), the checks on when to allow different forms of
1629 "control" were broken. There should now be diagnostics for all cases when a
1630 control that does not make sense is encountered.
1632 9. Added the /retain_sender option to "control=submission".
1634 10. $recipients is now available in the predata ACL (oversight).
1636 11. Tidy the search cache before the fork to do a delivery from a message
1637 received from the command line. Otherwise the child will trigger a lookup
1638 failure and thereby defer the delivery if it tries to use (for example) a
1639 cached ldap connection that the parent has called unbind on.
1641 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1642 of $address_data from the recipient verification was clobbered by the
1643 sender verification.
1645 13. The value of address_data from a sender verification is now available in
1646 $sender_address_data in subsequent conditions in the ACL statement.
1648 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1650 15. Added a new option "connect=<time>" to callout options, to set a different
1653 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1654 was its contents. (It was OK if the option was not defined at all.)
1656 17. A "Completed" log line is now written for messages that are removed from
1657 the spool by the -Mrm option.
1659 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1660 information about exactly what failed.
1662 19. Added -dd to debug only the daemon process.
1664 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1665 handles timeouts, both on the server side and network timeouts. Renamed the
1666 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1668 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1669 It is now set to "smtps".
1671 22. $host_address is now set to the target address during the checking of
1672 ignore_target_hosts.
1674 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1675 being passed; this would have caused $sender_host_name to have been used if
1676 matching the list had actually called for a host name (not very likely,
1677 since this list is usually IP addresses). A host name is now passed as
1680 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1681 code that specifies a non-restarting handler (typically sigaction() in
1682 modern systems) in an attempt to fix a rare and obscure crash bug.
1684 25. Narrowed the window for a race in the daemon that could cause it to ignore
1685 SIGCHLD signals. This is not a major problem, because they are used only to
1686 wake it up if nothing else does.
1688 26. A malformed maildirsize file could cause Exim to calculate negative values
1689 for the mailbox size or file count. Odd effects could occur as a result.
1690 The maildirsize information is now recalculated if the size or filecount
1693 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1694 support for a long time. Removed HAVE_SYS_VFS_H.
1696 28. Installed the latest version of exipick from John Jetmore.
1698 29. In an address list, if the pattern was not a regular expression, an empty
1699 subject address (from a bounce message) matched only if the pattern was an
1700 empty string. Non-empty patterns were not even tested. This was the wrong
1701 because it is perfectly reasonable to use an empty address as part of a
1702 database query. An empty address is now tested by patterns that are
1703 lookups. However, all the other forms of pattern expect the subject to
1704 contain a local part and a domain, and therefore, for them, an empty
1705 address still always fails if the pattern is not itself empty.
1707 30. Exim went into a mad DNS loop when attempting to do a callout where the
1708 host was specified on an smtp transport, and looking it up yielded more
1709 than one IP address.
1711 31. Re-factored the code for checking spool and log partition space into a
1712 function that finds that data and another that does the check. The former
1713 is then used to implement four new variables: $spool_space, $log_space,
1714 $spool_inodes, and $log_inodes.
1716 32. The RFC2047 encoding function was originally intended for short strings
1717 such as real names; it was not keeping to the 75-character limit for
1718 encoded words that the RFC imposes. It now respects the limit, and
1719 generates multiple encoded words if necessary. To be on the safe side, I
1720 have increased the buffer size for the ${rfc2047: expansion operator from
1723 33. It is now permitted to omit both strings after an "if" condition; if the
1724 condition is true, the result is "true". As before, when the second string
1725 is omitted, a false condition yields an empty string. This makes it less
1726 cumbersome to write custom ACL and router conditions.
1728 34. Failure to deliver a bounce message always caused it to be frozen, even if
1729 there was an errors_to setting on the router. The errors_to setting is now
1732 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1733 canonical form (fully expanded) before being placed in
1734 $sender_host_address.
1736 36. The table in the code that translates DNS record types into text (T_A to
1737 "A" for instance) was missing entries for NS and CNAME. It is just possible
1738 that this could have caused confusion if both these types were looked up
1739 for the same domain, because the text type is used as part of Exim's
1740 per-process caching. But the chance of anyone hitting this buglet seems
1743 37. The dnsdb lookup has been extended in a number of ways.
1745 (1) There is a new type, "zns", which walks up the domain tree until it
1746 finds some nameserver records. It should be used with care.
1748 (2) There is a new type, "mxh", which is like "mx" except that it returns
1749 just the host names, not the priorities.
1751 (3) It is now possible to give a list of domains (or IP addresses) to be
1752 looked up. The behaviour when one of the lookups defers can be
1753 controlled by a keyword.
1755 (4) It is now possible to specify the separator character for use when
1756 multiple records are returned.
1758 38. The dnslists ACL condition has been extended: it is now possible to supply
1759 a list of IP addresses and/or domains to be looked up in a particular DNS
1762 39. Added log_selector=+queue_time_overall.
1764 40. When running the queue in the test harness, wait just a tad after forking a
1765 delivery process, to get repeatability of debugging output.
1767 41. Include certificate and key file names in error message when GnuTLS fails
1768 to set them up, because the GnuTLS error message doesn't include the name
1769 of the failing file when there is a problem reading it.
1771 42. Allow both -bf and -bF in the same test run.
1773 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1775 44. The "Exiscan patch" is now merged into the mainline Exim source.
1777 45. Sometimes the final signoff response after QUIT could fail to get
1778 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1779 < 0 before doing a fflush(). This bug looks as though it goes back to the
1780 introduction of TLS in release 3.20, but "sometimes" must have been rare
1781 because the tests only now provoked it.
1783 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1784 (this can affect the format of dates).
1786 47. exim_tidydb, when checking for the continued existence of a message for
1787 which it has found a message-specific retry record, was not finding
1788 messages that were in split spool directories. Consequently, it was
1789 deleting retry records that should have stayed in existence.
1791 48. Steve fixed some bugs in eximstats.
1793 49. The SPA authentication driver was not abandoning authentication and moving
1794 on to the next authenticator when an expansion was forced to fail,
1795 contradicting the general specification for all authenticators. Instead it
1796 was generating a temporary error. It now behaves as specified.
1798 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1799 (the order specifies the preference for clients). The order is now AES256,
1800 AES128, 3DES, ARCFOUR128.
1802 51. Small patch to Sieve code - explicitly set From: when generating an
1805 52. Exim crashed if a remote delivery caused a very long error message to be
1806 recorded - for instance if somebody sent an entire SpamAssassin report back
1807 as a large number of 550 error lines. This bug was coincidentally fixed by
1808 increasing the size of one of Exim's internal buffers (big_buffer) that
1809 happened as part of the Exiscan merge. However, to be on the safe side, I
1810 have made the code more robust (and fixed the comments that describe what
1813 53. Now that there can be additional text after "Completed" in log lines (if
1814 the queue_time_overall log selector is set), a one-byte patch to exigrep
1815 was needed to allow it to recognize "Completed" as not the last thing in
1818 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1819 patch that reportedly fixes this has been added. I am not expert enough to
1820 create a test for it. This is what the patch creator wrote:
1822 "I found a little strange behaviour of ldap code when working with
1823 Windows 2003 AD Domain, where users was placed in more than one
1824 Organization Units. When I tried to give exim partial DN, the exit code
1825 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1826 But simultaneously result of request was absolutely normal ldap result,
1827 so I produce this patch..."
1829 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1830 REFERENCE, so I have modified the code to exclude the patch when that macro
1833 55. Some experimental protocols are using DNS PTR records for new purposes. The
1834 keys for these records are domain names, not reversed IP addresses. The
1835 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1836 leaves it alone. Component reversal etc. now happens only for IP addresses.
1839 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1841 57. Double the size of the debug message buffer (to 2048) so that more of very
1842 long debug lines gets shown.
1844 58. The exicyclog utility now does better if the number of log files to keep
1845 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1847 59. Two changes related to the smtp_active_hostname option:
1849 (1) $smtp_active_hostname is now available as a variable.
1850 (2) The default for smtp_banner uses $smtp_active_hostname instead
1851 of $primary_hostname.
1853 60. The host_aton() function is supposed to be passed a string that is known
1854 to be a valid IP address. However, in the case of IPv6 addresses, it was
1855 not checking this. This is a hostage to fortune. Exim now panics and dies
1856 if the condition is not met. A case was found where this could be provoked
1857 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1858 components; fortuitously, this particular loophole had already been fixed
1859 by change 4.50/55 above.
1861 If there are any other similar loopholes, the new check in host_aton()
1862 itself should stop them being exploited. The report I received stated that
1863 data on the command line could provoke the exploit when Exim was running as
1864 exim, but did not say which command line option was involved. All I could
1865 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1866 running as the user.
1869 61. There was a buffer overflow vulnerability in the SPA authentication code
1870 (which came originally from the Samba project). I have added a test to the
1871 spa_base64_to_bits() function which I hope fixes it.
1874 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1875 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1877 63. The daemon start-up calls getloadavg() while still root for those OS that
1878 need the first call to be done as root, but it missed one case: when
1879 deliver_queue_load_max is set with deliver_drop_privilege. This is
1880 necessary for the benefit of the queue runner, because there is no re-exec
1881 when deliver_drop_privilege is set.
1883 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1884 This has been fixed.
1886 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1887 was in use, was not putting the data itself into the right store pool;
1888 consequently, it could be overwritten for a subsequent message in the same
1889 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1892 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1894 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1895 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1896 the function were treating the return as a boolean value, which happened to
1897 work because 0=false and not-0=true, but is not correct code.
1899 68. The host_aton() function was not handling scoped IPv6 addresses (those
1900 with, for example, "%eth0" on the end) correctly.
1902 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1903 negated items (that is, ~something) in unsigned ints. Some compilers
1904 apparently mutter when there is no cast.
1906 70. If an address verification called from an ACL failed, and did not produce a
1907 user-specific message (i.e. there was only a "system" message), nothing was
1908 put in $acl_verify_message. In this situation, it now puts the system
1911 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1912 session; change 4.31/43 added the unwanted input to the log line - except
1913 that it did not do this in the start of session case. It now does.
1915 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1916 This is wrong and can cause the other end to generate a synchronization
1917 error if it is another Exim or anything else that does the synchronization
1918 check. A QUIT command is no longer sent after a timeout.
1920 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1921 during host lookups.
1923 74. The defer_ok option of callout verification was not working if it was used
1924 when verifying addresses in header lines, that is, for this case:
1926 verify = header_sender/callout=defer_ok
1928 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1929 those file descriptors could be used for SMTP connections. If anything
1930 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1931 could be sent to the SMTP client, causing chaos. The daemon now opens
1932 stdin, stdout, and stderr to /dev/null when it puts itself into the
1935 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1936 log by default. The user can override this with suitable Perl magic.
1938 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1939 the log message when discard triggers, was not working for the DATA ACL or
1940 for the non-SMTP ACL.
1942 78. Error message wording change in sieve.c.
1944 79. If smtp_accept_max_per_host was set, the number of connections could be
1945 restricted to fewer than expected, because the daemon was trying to set up
1946 a new connection before checking whether the processes handling previous
1947 connections had finished. The check for completed processes is now done
1948 earlier. On busy systems, this bug wouldn't be noticed because something
1949 else would have woken the daemon, and it would have reaped the completed
1952 80. If a message was submitted locally by a user whose login name contained one
1953 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1954 It caused a spool format error. I have fixed the spool reading code. A
1955 related problem was that the "from" clause in the Received: line became
1956 illegal because of the space(s). It is now covered by ${quote_local_part.
1958 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1961 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1963 83. Patch from Sieve maintainer to fix unterminated string problem in
1964 "vacation" handling.
1966 84. Some minor changes to the Linux configuration files to help with other
1967 OS variants using glibc.
1969 85. One more patch for Sieve to update vacation handling to latest spec.
1972 ----------------------------------------------------
1973 See the note above about the 4.44 and 4.50 releases.
1974 ----------------------------------------------------
1980 1. Change 4.43/35 introduced a bug that caused file counts to be
1981 incorrectly computed when quota_filecount was set in an appendfile
1984 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1985 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1988 3. Give more explanation in the error message when the command for a transport
1989 filter fails to execute.
1991 4. There are several places where Exim runs a non-Exim command in a
1992 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1993 was being done only for the command run by the queryprogram router. It is
1994 now done for all such subprocesses. The other cases are: ${run, transport
1995 filters, and the commands run by the lmtp and pipe transports.
1997 5. Some older OS have a limit of 256 on the maximum number of file
1998 descriptors. Exim was using setrlimit() to set 1000 as a large value
1999 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2000 systems. I've change it so that if it can't get 1000, it tries for 256.
2002 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2003 was an oversight, and furthermore, ever since the addition of extra
2004 controls (e.g. 4.43/32), the checks on when to allow different forms of
2005 "control" were broken. There should now be diagnostics for all cases when a
2006 control that does not make sense is encountered.
2008 7. $recipients is now available in the predata ACL (oversight).
2010 8. Tidy the search cache before the fork to do a delivery from a message
2011 received from the command line. Otherwise the child will trigger a lookup
2012 failure and thereby defer the delivery if it tries to use (for example) a
2013 cached ldap connection that the parent has called unbind on.
2015 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2016 of $address_data from the recipient verification was clobbered by the
2017 sender verification.
2019 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2020 was its contents. (It was OK if the option was not defined at all.)
2022 11. A "Completed" log line is now written for messages that are removed from
2023 the spool by the -Mrm option.
2025 12. $host_address is now set to the target address during the checking of
2026 ignore_target_hosts.
2028 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2029 being passed; this would have caused $sender_host_name to have been used if
2030 matching the list had actually called for a host name (not very likely,
2031 since this list is usually IP addresses). A host name is now passed as
2034 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2035 code that specifies a non-restarting handler (typically sigaction() in
2036 modern systems) in an attempt to fix a rare and obscure crash bug.
2038 15. Narrowed the window for a race in the daemon that could cause it to ignore
2039 SIGCHLD signals. This is not a major problem, because they are used only to
2040 wake it up if nothing else does.
2042 16. A malformed maildirsize file could cause Exim to calculate negative values
2043 for the mailbox size or file count. Odd effects could occur as a result.
2044 The maildirsize information is now recalculated if the size or filecount
2047 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2048 support for a long time. Removed HAVE_SYS_VFS_H.
2050 18. Updated exipick to current release from John Jetmore.
2052 19. Allow an empty sender to be matched against a lookup in an address list.
2053 Previously the only cases considered were a regular expression, or an
2056 20. Exim went into a mad DNS lookup loop when doing a callout where the
2057 host was specified on the transport, if the DNS lookup yielded more than
2060 21. The RFC2047 encoding function was originally intended for short strings
2061 such as real names; it was not keeping to the 75-character limit for
2062 encoded words that the RFC imposes. It now respects the limit, and
2063 generates multiple encoded words if necessary. To be on the safe side, I
2064 have increased the buffer size for the ${rfc2047: expansion operator from
2067 22. Failure to deliver a bounce message always caused it to be frozen, even if
2068 there was an errors_to setting on the router. The errors_to setting is now
2071 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2072 canonical form (fully expanded) before being placed in
2073 $sender_host_address.
2075 24. Updated eximstats to version 1.33
2077 25. Include certificate and key file names in error message when GnuTLS fails
2078 to set them up, because the GnuTLS error message doesn't include the name
2079 of the failing file when there is a problem reading it.
2081 26. Expand error message when OpenSSL has problems setting up cert/key files.
2084 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2085 (this can affect the format of dates).
2087 28. exim_tidydb, when checking for the continued existence of a message for
2088 which it has found a message-specific retry record, was not finding
2089 messages that were in split spool directories. Consequently, it was
2090 deleting retry records that should have stayed in existence.
2092 29. eximstats updated to version 1.35
2093 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2094 1.35 - bugfix such that pie charts by volume are generated correctly
2096 30. The SPA authentication driver was not abandoning authentication and moving
2097 on to the next authenticator when an expansion was forced to fail,
2098 contradicting the general specification for all authenticators. Instead it
2099 was generating a temporary error. It now behaves as specified.
2101 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2102 (the order specifies the preference for clients). The order is now AES256,
2103 AES128, 3DES, ARCFOUR128.
2105 31. Small patch to Sieve code - explicitly set From: when generating an
2108 32. Exim crashed if a remote delivery caused a very long error message to be
2109 recorded - for instance if somebody sent an entire SpamAssassin report back
2110 as a large number of 550 error lines. This bug was coincidentally fixed by
2111 increasing the size of one of Exim's internal buffers (big_buffer) that
2112 happened as part of the Exiscan merge. However, to be on the safe side, I
2113 have made the code more robust (and fixed the comments that describe what
2116 33. Some experimental protocols are using DNS PTR records for new purposes. The
2117 keys for these records are domain names, not reversed IP addresses. The
2118 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2119 leaves it alone. Component reversal etc. now happens only for IP addresses.
2122 34. The host_aton() function is supposed to be passed a string that is known
2123 to be a valid IP address. However, in the case of IPv6 addresses, it was
2124 not checking this. This is a hostage to fortune. Exim now panics and dies
2125 if the condition is not met. A case was found where this could be provoked
2126 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2127 components; fortuitously, this particular loophole had already been fixed
2128 by change 4.50/55 or 4.44/33 above.
2130 If there are any other similar loopholes, the new check in host_aton()
2131 itself should stop them being exploited. The report I received stated that
2132 data on the command line could provoke the exploit when Exim was running as
2133 exim, but did not say which command line option was involved. All I could
2134 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2135 running as the user.
2138 35. There was a buffer overflow vulnerability in the SPA authentication code
2139 (which came originally from the Samba project). I have added a test to the
2140 spa_base64_to_bits() function which I hope fixes it.
2143 36. The daemon start-up calls getloadavg() while still root for those OS that
2144 need the first call to be done as root, but it missed one case: when
2145 deliver_queue_load_max is set with deliver_drop_privilege. This is
2146 necessary for the benefit of the queue runner, because there is no re-exec
2147 when deliver_drop_privilege is set.
2149 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2150 was in use, was not putting the data itself into the right store pool;
2151 consequently, it could be overwritten for a subsequent message in the same
2152 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2155 38. Sometimes the final signoff response after QUIT could fail to get
2156 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2157 < 0 before doing a fflush(). This bug looks as though it goes back to the
2158 introduction of TLS in release 3.20, but "sometimes" must have been rare
2159 because the tests only now provoked it.
2165 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2166 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2167 Now it returns an integer. A number of calls were still expecting a T/F
2168 return. Fortuitously, in all cases, the tests worked in OK situations,
2169 which is the norm. However, things would have gone wrong on any write
2170 failures on the smtp file descriptor. This function is used when sending
2171 messages over SMTP and also when doing verify callouts.
2173 2. When Exim is called to do synchronous delivery of a locally submitted
2174 message (the -odf or -odi options), it no longer closes stderr before doing
2177 3. Implemented the mua_wrapper option.
2179 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2181 5. Implemented the functions header_remove(), header_testname(),
2182 header_add_at_position(), and receive_remove_recipient(), and exported them
2185 6. If an ACL "warn" statement specified the addition of headers, Exim already
2186 inserted X-ACL-Warn: at the start if there was no header name. However, it
2187 was not making this test for the second and subsequent header lines if
2188 there were newlines in the string. This meant that an invalid header could
2189 be inserted if Exim was badly configured.
2191 7. Allow an ACL "warn" statement to add header lines at the start or after all
2192 the Received: headers, as well as at the end.
2194 8. Added the rcpt_4xx retry error code.
2196 9. Added postmaster_mailfrom=xxx to callout verification option.
2198 10. Added mailfrom=xxxx to the callout verification option, for verify=
2201 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2202 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2204 12. Inserted some casts to stop certain compilers warning when using pointer
2205 differences as field lengths or precisions in printf-type calls (mostly
2206 affecting debugging statements).
2208 13. Added optional readline() support for -be (dynamically loaded).
2210 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2211 same clock tick as a message's arrival, so that its received time was the
2212 same as the "first fail" time on the retry record, and that message
2213 remained on the queue past the ultimate address timeout, every queue runner
2214 would try a delivery (because it was past the ultimate address timeout) but
2215 after another failure, the ultimate address timeout, which should have then
2216 bounced the address, did not kick in. This was a "< instead of <=" error;
2217 in most cases the first failure would have been in the next clock tick
2218 after the received time, and all would be well.
2220 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2221 being recognized when the domain list was tested by the match_domain
2222 condition in an expansion string.
2224 16. Added the ${str2b64: operator.
2226 17. Exim was always calling setrlimit() to set a large limit for the number of
2227 processes, without checking whether the existing limit was already
2228 adequate. (It did check for the limit on file descriptors.) Furthermore,
2229 errors from getrlimit() and setrlimit() were being ignored. Now they are
2230 logged to the main and panic logs, but Exim does carry on, to try to do its
2231 job under whatever limits there are.
2233 18. Imported PCRE 5.0.
2235 19. Trivial typo in log message " temporarily refused connection" (the leading
2238 20. If the log selector return_path_on_delivery was set and an address was
2239 redirected to /dev/null, the delivery process crashed because it assumed
2240 that a return path would always be set for a "successful" delivery. In this
2241 case, the whole delivery is bypassed as an optimization, and therefore no
2244 21. Internal re-arrangement: the function for sending a challenge and reading
2245 a response while authentication was assuming a zero-terminated challenge
2246 string. It's now changed to take a pointer and a length, to allow for
2247 binary data in such strings.
2249 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2251 23. Exim was not respecting finduser_retries when seeking the login of the
2252 uid under which it was called; it was always trying 10 times. (The default
2253 setting of finduser_retries is zero.) Also, it was sleeping after the final
2254 failure, which is pointless.
2256 24. Implemented tls_on_connect_ports.
2258 25. Implemented acl_smtp_predata.
2260 26. If the domain in control=submission is set empty, Exim assumes that the
2261 authenticated id is a complete email address when it generates From: or
2262 Sender: header lines.
2264 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2265 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2266 chown and chgrp in /bin and hostname in /usr/bin.
2268 28. Exim was keeping the "process log" file open after each use, just as it
2269 does for the main log. This opens the possibility of it remaining open for
2270 long periods when the USR1 signal hits a daemon. Occasional processlog
2271 errors were reported, that could have been caused by this. Anyway, it seems
2272 much more sensible not to leave this file open at all, so that is what now
2275 29. The long-running daemon process does not normally write to the log once it
2276 has entered its main loop, and it closes the log before doing so. This is
2277 so that log files can straightforwardly be renamed and moved. However,
2278 there are a couple of unusual error situations where the daemon does write
2279 log entries, and I had neglected to close the log afterwards.
2281 30. The text of an SMTP error response that was received during a remote
2282 delivery was being truncated at 512 bytes. This is too short for some of
2283 the long messages that one sometimes sees. I've increased the limit to
2286 31. It is now possible to make retry rules that apply only when a message has a
2287 specific sender, in particular, an empty sender.
2289 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2290 it possible to be selective about when SMTP synchronization is enforced.
2292 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2294 32. Implemented hosts_connection_nolog.
2296 33. Added an ACL for QUIT.
2298 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2301 35. Added mailbox_size and mailbox_filecount to appendfile.
2303 36. Added control = no_multiline_responses to ACLs.
2305 37. There was a bug in the logic of the code that waits for the clock to tick
2306 in the case where the clock went backwards by a substantial amount such
2307 that the microsecond fraction of "now" was more than the microsecond
2308 fraction of "then" (but the whole seconds number was less).
2310 38. Added support for the libradius Radius client library this is found on
2311 FreeBSD (previously only the radiusclient library was supported).
2317 1. When certain lookups returned multiple values in the form name=value, the
2318 quoting of the values was not always being done properly. Specifically:
2319 (a) If the value started with a double quote, but contained no whitespace,
2321 (b) If the value contained whitespace other than a space character (i.e.
2322 tabs or newlines or carriage returns) it was not quoted.
2323 This fix has been applied to the mysql and pgsql lookups by writing a
2324 separate quoting function and calling it from the lookup code. The fix
2325 should probably also be applied to nisplus, ibase and oracle lookups, but
2326 since I cannot test any of those, I have not disturbed their existing code.
2328 2. A hit in the callout cache for a specific address caused a log line with no
2329 reason for rejecting RCPT. Now it says "Previous (cached) callout
2330 verification failure".
2332 3. There was an off-by-one bug in the queryprogram router. An over-long
2333 return line was truncated at 256 instead of 255 characters, thereby
2334 overflowing its buffer with the terminating zero. As well as fixing this, I
2335 have increased the buffer size to 1024 (and made a note to document this).
2337 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2338 when Exim is waiting for an SMTP response from a remote server, Exim
2339 restarts its select() call on the socket, thereby resetting its timeout.
2340 This is not a problem when such interrupts are rare. Somebody set up a cron
2341 job to run exiwhat every 2 minutes, which is less than the normal select()
2342 timeout (5 or 10 minutes). This meant that the select() timeout never
2343 kicked in because it was always reset. I have fixed this by comparing the
2344 time when an interrupt arrives with the time at the start of the first call
2345 to select(). If more time than the timeout has elapsed, the interrupt is
2346 treated as a timeout.
2348 5. Some internal re-factoring in preparation for the addition of Sieve
2349 extensions (by MH). In particular, the "personal" test is moved to a
2350 separate function, and given an option for scanning Cc: and Bcc: (which is
2351 not set for Exim filters).
2353 6. When Exim created an email address using the login of the caller as the
2354 local part (e.g. when creating a From: or Sender: header line), it was not
2355 quoting the local part when it contained special characters such as @.
2357 7. Installed new OpenBSD configuration files.
2359 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2360 try to make them clearer.
2362 9. Callout options, other than the timeout value, were being ignored when
2363 verifying sender addresses in header lines. For example, when using
2365 verify = header_sender/callout=no_cache
2367 the cache was (incorrectly) being used.
2369 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2370 only the Cygwin environment.
2372 11. When return_path_on_delivery was set as a log selector, if different remote
2373 addresses in the same message used different return paths and parallel
2374 remote delivery occurred, the wrong values would sometimes be logged.
2375 (Whenever a remote delivery process finished, the return path value from
2376 the most recently started remote delivery process was logged.)
2378 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2379 lines when AUTH and/or TLS are in use. This is the "received protocol"
2380 field. Exim used to use "asmtp" for authenticated SMTP, without any
2381 indication (in the protocol name) for TLS use. Now it follows the RFC and
2382 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2383 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2384 names appear in log lines as well as in Received: header lines.
2386 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2387 extensions, and comparison tests, and to fix some bugs.
2389 14. Changes to the "personal" filter test:
2391 (1) The test was buggy in that it was just doing the equivalent of
2392 "contains" tests on header lines. For example, if a user's address was
2393 anne@some.where, the "personal" test would incorrectly be true for
2395 To: susanne@some.where
2397 This test is now done by extracting each address from the header in turn,
2398 and checking the entire address. Other tests that are part of "personal"
2399 are now done using regular expressions (for example, to check local parts
2400 of addresses in From: header lines).
2402 (2) The list of non-personal local parts in From: addresses has been
2403 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2404 taken from the Sieve specification recommendations.
2406 (3) If the message contains any header line starting with "List-" it is
2407 treated as non-personal.
2409 (4) The test for "circular" in the Subject: header line has been removed
2410 because it now seems ill-conceived.
2412 15. Minor typos in src/EDITME comments corrected.
2414 16. Installed latest exipick from John Jetmore.
2416 17. If headers_add on a router specified a text string that was too long for
2417 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2418 of string_sprintf() is now avoided.
2420 18. $message_body_size was not set (it was always zero) when running the DATA
2421 ACL and the local_scan() function.
2423 19. For the "mail" command in an Exim filter, no default was being set for
2424 the once_repeat time, causing a random time value to be used if "once" was
2425 specified. (If the value happened to be <= 0, no repeat happened.) The
2426 default is now 0s, meaning "never repeat". The "vacation" command was OK
2427 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2428 (I found it when inspecting the code).
2430 20. There is now an overall timeout for performing a callout verification. It
2431 defaults to 4 times the callout timeout, which applies to individual SMTP
2432 commands during the callout. The overall timeout applies when there is more
2433 than one host that can be tried. The timeout is checked before trying the
2434 next host. This prevents very long delays if there are a large number of
2435 hosts and all are timing out (e.g. when the network connections are timing
2436 out). The value of the overall timeout can be changed by specifying an
2437 additional sub-option for "callout", called "maxwait". For example:
2439 verify = sender/callout=5s,maxwait=20s
2441 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2442 to the end before writing, but this should make it even safer).
2444 22. Exim was forgetting that it had advertised PIPELINING for the second and
2445 subsequent messages on an SMTP connection. It was also not resetting its
2446 memory on STARTTLS and an internal HELO.
2448 23. When Exim logs an SMTP synchronization error within a session, it now
2449 records whether PIPELINING has been advertised or not.
2451 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2452 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2453 rather than long int.
2455 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2457 26. Added the never_mail option to autoreply.
2463 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2464 crash if the getsockname() call failed; this can happen if a connection is
2465 closed very soon after it is established. The problem was simply in the
2466 order in which certain operations were done, causing Exim to try to write
2467 to the SMTP stream before it had set up the file descriptor. The bug has
2468 been fixed by making things happen in the correct order.
2474 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2475 before the connection was closed, thus losing the rejection response.
2477 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2478 some early Solaris releases, but causes trouble in current releases where
2479 socklen_t is defined.
2481 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2484 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2487 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2488 that releasing the top of it at the end releases what was used for sub-
2489 expansions (unless the block got too big). However, discard this block if
2490 the first thing is a variable or header, so that we can use its block when
2491 it is dynamic (useful for very large $message_headers, for example).
2493 6. Lookups now cache *every* query, not just the most recent. A new, separate
2494 store pool is used for this. It can be recovered when all lookup caches are
2495 flushed. Lookups now release memory at the end of their result strings.
2496 This has involved some general refactoring of the lookup sources.
2498 7. Some code has been added to the store_xxx() functions to reduce the amount
2499 of flapping under certain conditions.
2501 8. log_incoming_interface used to affect only the <= reception log lines. Now
2502 it causes the local interface and port to be added to several more SMTP log
2503 lines, for example "SMTP connection from", and rejection lines.
2505 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2507 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2509 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2510 could be overwritten at the end of the current message (or the start of a
2511 new message if it was set in a HELO ACL). The value is now preserved for
2512 the duration of the SMTP connection.
2514 12. If a transport had a headers_rewrite setting, and a matching header line
2515 contained an unqualified address, that address was qualified, even if it
2516 did not match any rewriting rules. The underlying bug was that the values
2517 of the flags that permit the existence of unqualified sender and recipient
2518 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2519 non-local messages, and by -bnq for local messages) were not being
2520 preserved with the message after it was received.
2522 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2523 "next input=" as part of the text comprising the host identity instead of
2524 the correct text. The code was using the same buffer for two different
2525 strings. However, depending on which order the printing function evaluated
2526 its arguments, the bug did not always show up. Under Linux, for example, my
2527 test suite worked just fine.
2529 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2530 doesn't work with some older versions of Perl. It has been changed to "my",
2531 which in any case is probably the better facility to use.
2533 15. A really picky compiler found some instances of statements for creating
2534 error messages that either had too many or two few arguments for the format
2537 16. The size of the buffer for calls to the DNS resolver has been increased
2538 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2539 for addresses that have a lot of PTR records. This alleviates a problem; it
2540 does not fully solve it.
2542 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2543 buffer now truncates the list and logs the incident, which is the same
2544 action as happens when Exim is looking up a host name and its aliases.
2545 Previously in this situation something unpredictable would happen;
2546 sometimes it was "internal error: store_reset failed".
2548 18. If a server dropped the connection unexpectedly when an Exim client was
2549 using GnuTLS and trying to read a response, the client delivery process
2550 crashed while trying to generate an error log message.
2552 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2553 string, for example:
2555 warn message = H1: something\nH2: something
2557 the text was added as a single header line from Exim's point of view
2558 though it ended up OK in the delivered message. However, searching for the
2559 second and subsequent header lines using $h_h2: did not work. This has been
2560 fixed. Similarly, if a system filter added multiple headers in this way,
2561 the routers could not see them.
2563 20. Expanded the error message when iplsearch is called with an invalid key to
2564 suggest using net-iplsearch in a host list.
2566 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2567 ACLs are no longer actually imposed (and a message to that effect is
2570 22. If a "gecos" field in a passwd entry contained escaped characters, in
2571 particular, if it contained a \" sequence, Exim got it wrong when building
2572 a From: or a Sender: header from that name. A second bug also caused
2573 incorrect handling when an unquoted " was present following a character
2574 that needed quoting.
2576 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2577 was not being matched caselessly.
2579 24. Arranged for all hyphens in the exim.8 source to be escaped with
2582 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2583 properly. Recipient callout cache records were still being keyed to include
2584 the sender, even when use_sender was set false. This led to far more
2585 callouts that were necessary. The sender is no longer included in the key
2586 when use_sender is false.
2588 26. Added "control = submission" modifier to ACLs.
2590 27. Added the ${base62d: operator to decode base 62 numbers.
2592 28. dnsdb lookups can now access SRV records.
2594 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2595 the configuration file.
2597 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2598 (-v) mode. This makes the output for a verbose queue run more intelligible.
2600 31. Added a use_postmaster feature to recipient callouts.
2602 32. Added the $body_zerocount variable, containing the number of binary zero
2603 bytes in the message body.
2605 33. The time of last modification of the "new" subdirectory is now used as the
2606 "mailbox time last read" when there is a quota error for a maildir
2609 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2611 35. Added +ignore_unknown as a special item in host lists.
2613 36. Code for decoding IPv6 addresses in host lists is now included, even if
2614 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2615 address was recognized as an IP address, but was then not correctly decoded
2616 into binary, causing unexpected and incorrect effects when compared with
2623 1. Very minor rewording of debugging text in manualroute to say "list of
2624 hosts" instead of "hostlist".
2626 2. If verify=header_syntax was set, and a header line with an unqualified
2627 address (no domain) and a large number of spaces between the end of the
2628 name and the colon was received, the reception process suffered a buffer
2629 overflow, and (when I tested it) crashed. This was caused by some obsolete
2630 code that should have been removed. The fix is to remove it!
2632 3. When running in the test harness, delay a bit after writing a bounce
2633 message to get a bit more predictability in the log output.
2635 4. Added a call to search_tidyup() just before forking a reception process. In
2636 theory, someone could use a lookup in the expansion of smtp_accept_max_
2637 per_host which, without the tidyup, could leave open a database connection.
2639 5. Added the variables $recipient_data and $sender_data which get set from a
2640 lookup success in an ACL "recipients" or "senders" condition, or a router
2641 "senders" option, similar to $domain_data and $local_part_data.
2643 6. Moved the writing of debug_print from before to after the "senders" test
2646 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2647 problems for message scanning, either using a data ACL, or using
2648 local_scan() because the Received: header was not generated till after they
2649 were called (in order to set the time as the time of reception completion).
2650 I have revised the way this works. The header is now generated after the
2651 body is received, but before the ACL or local_scan() are called. After they
2652 are run, the timestamp in the header is updated.
2658 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2659 before starting a queue runner without re-exec. This happened only when
2660 deliver_drop_privilege was set or when the Exim user was set to root. The
2661 effect of the bug was that timeouts during subsequent deliveries caused
2662 crashes instead of being properly handled. The handler is now left at its
2663 default (and expected) setting.
2665 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2666 message, again when deliver_drop_privilege is set or Exim is run as root.
2667 The bug described in (1) was not present in this case, but the tidying up
2668 of the other signals was missing. I have made the two cases consistent.
2670 3. The ignore_target_hosts setting on a manualroute router was being ignored
2671 for hosts that were looked up using the /MX notation.
2673 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2676 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2677 operated on the sender address. After changing the $sender_address to <>
2678 for the sender address verify, Exim was re-instated it as the original
2679 (before rewriting) address, but remembering that it had rewritten it, so it
2680 wasn't rewriting it again. This bug also had the effect of breaking the
2681 sender address verification caching when the sender address was rewritten.
2683 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2684 This has been changed so that if the ip literal address matches
2685 ignore_target_hosts, the router declines.
2687 7. Added expansion conditions match_domain, match_address, and match_local_
2688 part (NOT match_host).
2690 8. The placeholder for the Received: header didn't have a length field set.
2692 9. Added code to Exim itself and to exim_lock to test for a specific race
2693 condition that could lead to file corruption when using MBX delivery. The
2694 issue is with the lockfile that is created in /tmp. If this file is removed
2695 after a process has opened it but before that process has acquired a lock,
2696 there is the potential for a second process to recreate the file and also
2697 acquire a lock. This could lead to two Exim processes writing to the file
2698 at the same time. The added code performs the same test as UW imapd; it
2699 checks after acquiring the lock that its file descriptor still refers to
2700 the same named file.
2702 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2703 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2706 11. Added the smtp_active_hostname option. If used, this will typically be made
2707 to depend on the incoming interface address. Because $interface_address is
2708 not set up until the daemon has forked a reception process, error responses
2709 that can happen earlier (such as "too many connections") no longer contain
2712 12. If an expansion in a condition on a "warn" statement fails because a lookup
2713 defers, the "warn" statement is abandoned, and the next ACL statement is
2714 processed. Previously this caused the whole ACL to be aborted.
2716 13. Added the iplsearch lookup type.
2718 14. Added ident_timeout as a log selector.
2720 15. Added tls_certificate_verified as a log selector.
2722 16. Added a global option tls_require_ciphers (compare the smtp transport
2723 option of the same name). This controls incoming TLS connections.
2725 17. I finally figured out how to make tls_require_ciphers do a similar thing
2726 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2727 before starting the TLS session.
2729 18. Tabs are now shown as \t in -bP output.
2731 19. If the log selector return_path_on_delivery was set, Exim crashed when
2732 bouncing a message because it had too many Received: header lines.
2734 20. If two routers both had headers_remove settings, and the first one included
2735 a superfluous trailing colon, the final name in the first list and the
2736 first name in the second list were incorrectly joined into one item (with a
2737 colon in the middle).
2743 1. Added -C and -D options to the exinext utility, mainly to make it easier
2744 to include in the automated testing, but these could be helpful when
2745 multiple configurations are in use.
2747 2. The exinext utility was not formatting the output nicely when there was
2748 an alternate port involved in the retry record key, nor when there was a
2749 message id as well (for retries that were specific to a specific message
2750 and a specific host). It was also confused by IPv6 addresses, because of
2751 the additional colons they contain. I have fixed the IPv4 problem, and
2752 patched it up to do a reasonable job for IPv6.
2754 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2755 delivery, the log line now contains "pipelined" if PIPELINING was used.
2757 4. An SMTP transport process used to panic and die if the bind() call to set
2758 an explicit outgoing interface failed. This has been changed; it is now
2759 treated in the same way as a connect() failure.
2761 5. A reference to $sender_host_name in the part of a conditional expansion
2762 that was being skipped was still causing a DNS lookup. This no longer
2765 6. The def: expansion condition was not recognizing references to header lines
2766 that used bh_ and bheader_.
2768 7. Added the _cache feature to named lists.
2770 8. The code for checking quota_filecount in the appendfile transport was
2771 allowing one more file than it should have been.
2773 9. For compatibility with Sendmail, the command line option
2781 and sets the incoming protocol and host name (for trusted callers). The
2782 host name and its colon can be omitted when only the protocol is to be set.
2783 Note the Exim already has two private options, -pd and -ps, that refer to
2784 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2785 "s", but I don't think that's a major issue.
2787 10. A number of refactoring changes to the code, none of which should affect
2790 (a) The number of logging options was getting close to filling up the
2791 32-bit word that was used as a bit map. I have split them into two classes:
2792 those that are passed in the argument to log_write(), and those that are
2793 only ever tested independently outside of that function. These are now in
2794 separate 32-bit words, so there is plenty of room for expansion again.
2795 There is no change in the user interface or the logging behaviour.
2797 (b) When building, for example, log lines, the code previously used a
2798 macro that called string_cat() twice, in order to add two strings. This is
2799 not really sufficiently general. Furthermore, there was one instance where
2800 it was actually wrong because one of the argument was used twice, and in
2801 one call a function was used. (As it happened, calling the function twice
2802 did not affect the overall behaviour.) The macro has been replaced by a
2803 function that can join an arbitrary number of extra strings onto a growing
2806 (c) The code for expansion conditions now uses a table and a binary chop
2807 instead of a serial search (which was left over from when there were very
2808 few conditions). Also, it now recognizes conditions like "pam" even when
2809 the relevant support is not compiled in: a suitably worded error message is
2810 given if an attempt is made to use such a condition.
2812 11. Added ${time_interval:xxxxx}.
2814 12. A bug was causing one of the ddress fields not to be passed back correctly
2815 from remote delivery subprocesses. The field in question was not being
2816 subsequently used, so this caused to problems in practice.
2818 13. Added new log selectors queue_time and deliver_time.
2820 14. Might have fixed a bug in maildirsizefile handling that threw up
2821 "unexpected character" debug warnings, and recalculated the data
2822 unnecessarily. In any case, I expanded the warning message to give more
2825 15. Added the message "Restricted characters in address" to the statements in
2826 the default ACL that block characters like @ and % in local parts.
2828 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2829 Three changes have been made:
2831 (a) There was a serious bug; a negative response to MAIL caused the whole
2832 recipient domain to be cached as invalid, thereby blocking all messages
2833 to all local parts at the same domain, from all senders. This bug has
2834 been fixed. The domain is no longer cached after a negative response to
2835 MAIL if the sender used is not empty.
2837 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2840 (c) A new callout option, "use_sender" has been added for people who want
2841 the modified behaviour.
2847 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2850 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2851 indeed breaks things for older releases.
2853 3. Added additional logging to the case where there is a problem reading data
2854 from a filter that is running in a subprocess using a pipe, in order to
2855 try to track down a specific problem.
2857 4. Testing facility fudge: when running in the test harness and attempting
2858 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2859 getting "No route to host". Convert this to a timeout.
2861 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2864 6. Some OS don't have socklen_t but use size_t instead. This affects the
2865 fifth argument of getsockopt() amongst other things. This is now
2866 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2867 can be set for individual OS. I have set it for SunOS5, OSF1, and
2868 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2869 some earlier ones do not.
2871 7. Change 4.30/15 was not doing the test caselessly.
2873 8. The standard form for an IPv6 address literal was being rejected by address
2874 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2875 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2876 this, as well as the form without the "IPv6" on the front (but only when
2877 address literals are enabled, of course).
2879 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2881 10. Exim crashed if a message with an empty sender address specified by -f
2882 encountered a router with an errors_to setting. This could be provoked only
2883 by a command such as
2887 where an empty string was supplied; "<>" did not hit this bug.
2889 11. Installed PCRE release 4.5.
2891 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2892 remained set. It is now erased.
2894 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2895 times from message ids (which are base 36 rather than the normal 62).
2897 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2898 were being counted as actual protocol errors, and logged if the log
2899 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2900 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2901 and DATA following a set of rejected RCPTs do not count as protocol errors.
2902 In other words, Exim assumes they were pipelined, though this may not
2903 actually be the case. Of course, in all cases the client gets an
2904 appropriate error code.
2906 15. If a lookup fails in an ACL condition, a message about the failure may
2907 be available; it is used if testing the ACL cannot continue, because most
2908 such messages specify what the cause of the deferral is. However, some
2909 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2910 that caused an old message to be retained and used if a later statement
2911 caused a defer, replacing the real cause of the deferral.
2913 16. If an IP address had so many PTR records that the DNS lookup buffer
2914 was not large enough to hold them, Exim could crash while trying to process
2915 the truncated data. It now detects and logs this case.
2917 17. Further to 4.21/58, another change has been made: if (and only if) the
2918 first line of a message (the first header line) ends with CRLF, a bare LF
2919 in a subsequent header line has a space inserted after it, so as not to
2920 terminate the header.
2922 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2923 unnecessarily, used atoi() instead of strtol(), and didn't check the
2924 termination when getting file sizes from file names by regex.
2926 19. Completely re-implemented the support for maildirsize files, in the light
2927 of a number of problems with the previous contributed implementation
2928 (4.30/29). In particular:
2930 . If the quota is zero, the maildirsize file is maintained, but no quota is
2933 . If the maildir directory does not exist, it is created before any attempt
2934 to write a maildirsize file.
2936 . The quota value in the file is just a cache; if the quota is changed in
2937 the transport, the new value overrides.
2939 . A regular expression is available for excluding directories from the
2942 20. The autoreply transport checks the characters in options that define the
2943 message's headers; it allows continued headers, but it was checking with
2944 isspace() after an embedded newline instead of explicitly looking for a
2947 21. If all the "regular" hosts to which an address was routed had passed their
2948 expiry times, and had not reached their retry times, the address was
2949 bounced, even if fallback hosts were defined. Now Exim should go on to try
2952 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2953 equivalent code in the SMTP transport. Some hosts send humungous responses
2954 to HELO/EHLO, more than 1024 it seems.
2956 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2957 gives compiler warnings in some environments. I've now done it "properly",
2960 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2961 (because of problems with the built-in one) was declared to return uschar *
2962 instead of char *, causing compiler failure.
2964 25. Fixed a file descriptor leak when processing alias/forward files.
2966 26. Fixed a minor format string issue in dbfn.c.
2968 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2970 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2971 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2974 29. When a sender address is verified, it is cached, to save repeating the test
2975 when there is more than one recipient in a message. However, when the
2976 verification involves a callout, it is possible for different callout
2977 options to be set for different recipients. It is too complicated to keep
2978 track of this in the cache, so now Exim always runs a verification when a
2979 callout is required, relying on the callout cache for the optimization.
2980 The overhead is duplication of the address routing, but this should not be
2983 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2984 to be verified with callout=postmaster, and the main callout worked but the
2985 postmaster check failed, the verification correctly failed. However, if a
2986 subsequent RCPT command asked for sender verification *without* the
2987 postmaster check, incorrect caching caused this verification also to fail,
2990 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2991 it was not caching the DNS options (qualify_single, search_parents) that
2992 were used when the lookup failed. A subsequent lookup with different
2993 options therefore always gave the same answer, though there were cases
2994 where it should not have. (Example: a "domains = !$mx_any" option on a
2995 dnslookup router: the "domains" option is always processed without any
2996 widening, but the router might have qualify_single set.) Now Exim uses the
2997 cached value only when the same options are set.
2999 32. Added John Jetmore's "exipick" utility to the distribution.
3001 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3002 than a timeout (e.g. a certificate is required, and is not provided), an
3003 Exim server now closes the connection immediately. Previously it waited for
3004 the client to close - but if the client is SSL, it seems that they each
3005 wait for each other, leading to a delay before one of them times out.
3007 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3008 maintained 0.8.x compatibility because I don't think many are using it, and
3009 it is clearly obsolete.
3011 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3014 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3015 client certificate was expired. A simple patch fixes this, though I don't
3016 understand the full logic of why the verify callback is called multiple
3019 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3020 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3021 which causes problems with some clients (such as the Certicom SSL Plus
3022 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3023 disables the coutermeasure allowing Eudora to connect."
3025 38. Exim was not checking that a write() to a log file succeeded. This could
3026 lead to Bad Things if a log got too big, in particular if it hit a file
3027 size limit. Exim now panics and dies if it cannot write to a log file, just
3028 as it does if it cannot open a log file.
3030 39. Modified OS/Makefile-Linux so that it now contains
3032 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3034 The two -D definitions ensure that Exim is compiled with large file
3035 support, which makes it possible to handle log files that are bigger than
3038 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3039 instance) a domain was checked against a named list that involved a lookup,
3040 causing $domain_data to be set, then another domain was checked against the
3041 same list, then the first domain was re-checked, the value of $domain_data
3042 after the final check could be wrong. In particular, if the second check
3043 failed, it could be set empty. This bug probably also applied to
3046 41. The strip_trailing_dot option was not being applied to the address given
3047 with the -f command-line option.
3049 42. The code for reading a message's header from the spool was incrementing
3050 $received_count, but never initializing it. This meant that the value was
3051 incorrect (doubled) while delivering a message in the same process in which
3052 it was received. In the most common configuration of Exim, this never
3053 happens - a fresh exec is done - but it can happen when
3054 deliver_drop_privilege is set.
3056 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3057 it now includes up to 150 characters of the unexpected data in the log
3060 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3061 and building data strings. The size of both of these buffers was 10 000
3062 bytes - far larger than anybody would *ever* want, thought I. Needless to
3063 say, somebody hit the limit. I have increased the maximum line length to
3064 20 000 and the maximum data length of concatenated lines to 100 000. I have
3065 also fixed two bugs, because there was no checking on these buffers. Tsk,
3066 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3067 buffer is too small.
3069 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3070 lsearch lookups. Now it does.
3072 46. When parsing a route_list item in a manualroute router, a fixed-length
3073 buffer was used for the list of hosts. I made this 1024 bytes long,
3074 thinking that nobody would ever have a list of hosts that long. Wrong.
3075 Somebody had a whole pile of complicated expansion conditions, and the
3076 string was silently truncated, leading to an expansion error. It turns out
3077 that it is easier to change to an unlimited length (owing to other changes
3078 that have happened since this code was originally written) than to build
3079 structure for giving a limitation error. The length of the item that
3080 expands into the list of hosts is now unlimited.
3082 47. The lsearch lookup could not handle data where the length of text line was
3083 more than 4095 characters. Such lines were truncated, leading to shortened
3084 data being returned. It should now handle lines of any length.
3086 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3087 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3090 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3091 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3092 no longer happens in comment lines. A list of replacements is now placed
3093 at the head of all of the source files, except those whose only change is
3094 to replace PERL_COMMAND in the very first #! line.
3096 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3097 messages on the queue, with a bottom-up merge sort, using code contributed
3098 by Michael Haardt. This should make operations like -bp somewhat faster on
3099 large queues. It won't affect queue runners, except when queue_run_in_order
3102 51. Installed eximstats 1.31 in the distribution.
3104 52. Added support for SRV lookups to the dnslookup router.
3106 53. If an ACL referred to $message_body or $message_body_end, the value was not
3107 reset for any messages that followed in the same SMTP session.
3109 54. The store-handling optimization for building very long strings was not
3110 differentiating between the different store pools. I don't think this
3111 actually made any difference in practice, but I've tidied it.
3113 55. While running the routers to verify a sender address, $sender_address
3114 was still set to the sender address. This is wrong, because when routing to
3115 send a bounce to the sender, it would be empty. Therefore, I have changed
3116 it so that, while verifying a sender address, $sender_address is set to <>.
3117 (There is no change to what happens when verifying a recipient address.)
3119 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3120 target A or AAAA records (if not already returned) without resetting the
3121 qualify_single or search_parents options of the DNS resolver. These are
3122 inappropriate in this case because the targets of MX and SRV records must
3123 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3124 target that, when qualified, matched something in the local domain. These
3125 two options are now turned off when doing these lookups.
3127 57. It seems that at least some releases of Reiserfs (which does not have the
3128 concept of a fixed number of inodes) returns zero and not -1 for the
3129 number of available inodes. This interacted badly with check_spool_inodes,
3130 which assumed that -1 was the "no such thing" setting. What I have done is
3131 to check that the total number of inodes is greater than zero before doing
3132 the test of how many are available.
3134 58. When a "warn" ACL statement has a log_message modifier, the message is
3135 remembered, and not repeated. This is to avoid a lot of repetition when a
3136 message has many recipients that cause the same warning to be written.
3137 Howewer, Exim was preserving the list of already written lines for an
3138 entire SMTP session, which doesn't seem right. The memory is now reset if a
3139 new message is started.
3141 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3142 debugging output unless log_rewrite was also set.
3144 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3145 of (int)(handle) when we know that handle contains (void *)(-1).
3147 61. The Exim daemon panic-logs an error return when it closes the incoming
3148 connection. However "connection reset by peer" seems to be common, and
3149 isn't really an error worthy of noting specially, so that particular error
3152 62. When Exim is trying to find all the local interfaces, it used to panic and
3153 die if the ioctl to get the interface flags failed. However, it seems that
3154 on at least one OS (Solaris 9) it is possible to have an interface that is
3155 included in the list of interfaces, but for which you get a failure error
3156 for this call. This happens when the interface is not "plumbed" into a
3157 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3158 failure of the "get flags" call assumes that the interface is down.
3160 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3161 makes life easier for people who are doing arithmetic on fields extracted
3162 from dates, where you often get leading zeros that should not be
3163 interpreted as octal.
3165 64. Added qualify_domain to the redirect router, to override the global
3168 65. If a pathologically long header line contained very many addresses (the
3169 report of this problem mentioned 10 000) and each of them was rewritten,
3170 Exim could use up a very large amount of memory. (It kept on making new
3171 copies of the header line as it rewrote, and never released the old ones.)
3172 At the expense of a bit more processing, the header rewriting function has
3173 been changed so that it no longer eats memory in this way.
3175 66. The generation of the Received: header has been moved from the time that a
3176 message starts to be received, to the time that it finishes. The timestamp
3177 in the Received: header should now be very close to that of the <= log
3178 line. There are two side-effects of this change:
3180 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3181 logged header lines no longer include the local Received: line, because
3182 it has not yet been created. The same applies to a copy of the message
3183 that is returned to a non-SMTP sender when a message is rejected.
3185 (b) When a filter file is tested using -bf, no additional Received: header
3186 is added to the test message. After some thought, I decided that this
3189 This change does not affect the value of $received_for. It is still set
3190 after address rewriting, but before local_scan() is called.
3192 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3194 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3195 gave an unhelpful panic error message, and a defer error. I have managed to
3196 change this behaviour so that it now rejects any supplied certificate,
3197 which seems right, as the list of acceptable certificates is empty.
3199 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3200 gave an unhelpful defer error. I have not managed to make this reject any
3201 supplied certificates, but the error message it gives is "no certificate
3202 supplied", which is not helpful.
3204 70. exigrep's output now also includes lines that are not associated with any
3205 message, but which match the given pattern. Implemented by a patch from
3206 Martin Sluka, which also tidied up the Perl a bit.
3208 71. Recipient callout verification, like sender verification, was using <> in
3209 the MAIL FROM command. This isn't really the right thing, since the actual
3210 sender may affect whether the remote host accepts the recipient or not. I
3211 have changed it to use the actual sender in the callout; this means that
3212 the cache record is now keyed on a recipient/sender pair, not just the
3213 recipient address. There doesn't seem to be a real danger of callout loops,
3214 since a callout by the remote host to check the sender would use <>.
3215 [SEE ABOVE: changed after hitting problems.]
3217 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3218 temporary errors. However, in the case of such a code being given after
3219 the end of a data transmission (i.e. after ".") Exim was failing to write
3220 a retry record for the message. (Yes, there was some broken host that was
3221 actually sending 8xx at this point.)
3223 73. An unknown lookup type in a host list could cause Exim to panic-die when
3224 the list was checked. (An example that provoked this was putting <; in the
3225 middle of a list instead of at the start.) If this happened during a DATA
3226 ACL check, a -D file could be left lying around. This kind of configuration
3227 error no longer causes Exim to die; instead it causes a defer errror. The
3228 incident is still logged to the main and panic logs.
3230 74. Buglet left over from Exim 3 conversion. The message "too many messages
3231 in one connection" was written to the rejectlog but not the mainlog, except
3232 when address rewriting (yes!) was being logged.
3234 75. Added write_rejectlog option.
3236 76. When a system filter was run not as root (that is, when system_filter_user
3237 was set), the values of the $n variables were not being returned to the
3238 main process; thus, they were not subsequently available in the $sn
3241 77. Added +return_path_on_delivery log selector.
3243 78. A connection timeout was being treated differently from recipients deferred
3244 when testing hosts_max_try with a message that was older than the host's
3245 retry timeout. (The host should not be counted, thus allowing all hosts to
3246 be tried at least once before bouncing.) This may have been the cause of an
3247 occasionally reported bug whereby a message would remain on the queue
3248 longer than the retry timeout, but would be bounced if a delivery was
3249 forced. I say "may" because I never totally pinned down the problem;
3250 setting up timeout/retry tests is difficult. See also the next item.
3252 79. The ultimate address timeout was not being applied to errors that involved
3253 a combination of host plus message (for example, a timeout on a MAIL
3254 command). When an address resolved to a number of possible hosts, and they
3255 were not all tried for each delivery (e.g. because of hosts_max_try), a
3256 message could remain on the queue longer than the retry timeout.
3258 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3261 81. Fixed an obscure SMTP outgoing bug which required at least the following
3262 conditions: (a) there was another message waiting for the same server;
3263 (b) the server returned 5xx to all RCPT commands in the first message so
3264 that the message was not completed; (c) the server dropped the connection
3265 or gave a negative response to the RSET that Exim sends to abort the
3266 transaction. The observed case was a dropped connection after DATA that had
3267 been sent in pipelining mode. That is, the server had advertised PIPELINING
3268 but was not implementing it correctly. The effect of the bug was incorrect
3269 behaviour, such as trying another host, and this could lead to a crash.
3275 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3276 and daemon.c were passed as pointers to ints; they should have been
3277 pointers to socklen_t variables (which are typically unsigned ints).
3279 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3282 3. Fixed a really odd bug that affected only the testing scheme; patching a
3283 certain fixed string in the binary changed the value of another string that
3284 happened to be identical to the end of the original first string.
3286 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3287 name", it returns that address as the IP address. On some operating
3288 systems (e.g. Solaris), it also passes back the IP address string as the
3289 "host name". However, on others (e.g. Linux), it passes back an empty
3290 string. Exim wasn't checking for this, and was changing the host name to an
3291 empty string, assuming it had been canonicized.
3293 5. Although rare, it is permitted to have more than one PTR record for a given
3294 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3295 all the names associated with an address, because they do in Solaris.
3296 However, it seems that they do not in Linux for data that comes from the
3297 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3298 I found this out when I moved to a new Linux workstation and tried to run
3299 the Exim test suite.
3301 To get round this problem I have changed the code so that it now does its
3302 own call to the DNS to look up PTR records when searching for a host name.
3303 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3304 addresses that are only in /etc/hosts are still found.
3306 This behaviour is, however, controlled by an option called host_lookup_
3307 order, which defaults to "bydns:byaddr". If people want to use the other
3308 order, or indeed, just use one or the other means of lookup, they can
3309 specify it in this variable.
3311 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3312 some operating systems, this comes back from gethostbyaddr() as an empty
3313 string, and this is what Exim used to test for. However, it seems that in
3314 other systems, "." is yielded. Exim now tests for this case too.
3316 7. The values of check_spool_space and check_log_space are now held internally
3317 as a number of kilobytes instead of an absolute number of bytes. If a
3318 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3319 kilobyte. This means that much larger values can be stored.
3321 8. Exim monitor: an attempt to get the action menu when not actually pointing
3322 at a message produces an empty menu entitled "No message selected". This
3323 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3324 no entries in it ("Shell widget menu has zero width and/or height"). So I
3325 have added a single, blank menu entry in this case.
3327 9. Added ${quote_local_part.
3329 10. MIME decoding is now applied to the contents of Subject: header lines when
3332 11. Now that a reference to $sender_host_address automatically causes a reverse
3333 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3334 host lookup before query-style lookups in lists that might use this
3335 variable. This has therefore been abolished, and the "net-" prefix is no
3336 longer necessary for query-style lookups.
3338 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3339 have been a typo for LFLAGS, so it has been changed.
3341 13. The install script calls Exim with "-C /dev/null" in order to find the
3342 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3343 to be output. Howeve, since Exim outputs its version number before the
3344 error, it didn't break the script. It just looked ugly. I fixed this by
3345 always allowing "-C /dev/null" if the caller is root.
3347 14. Ignore overlarge ACL variable number when reading spool file - insurance
3348 against a later release with more variables having written the file.
3350 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3351 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3352 this, as well as the form without the "IPv6" on the front.
3354 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3355 OS/Makefile-Darwin file.
3357 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3358 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3360 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3361 variable. However, some OS use ints for inodes. I've added cast to long int
3362 to get rid of the compiler warning.
3364 19. I had forgotten to lock out "/../" in configuration file names when
3365 ALT_CONFIG_PREFIX was set.
3367 20. Routers used for verification do not need to specify transports. However,
3368 if such a router generated a host list, and callout was configured, Exim
3369 crashed, because it could not find a port number from the (non-existent)
3370 transport. It now assumes port 25 in this circumstance.
3372 21. Added the -t option to exigrep.
3374 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3375 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3376 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3377 them is set, LOOKUP_LSEARCH is forced.
3379 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3381 24. Added sender and host information to the "rejected by local_scan()" log
3382 line; previously there was no indication of these.
3384 25. Added .include_if_exists.
3386 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3387 Linux. It turns out that not all file systems support this. Apparently some
3388 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3389 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3390 not supported on the file descriptor, is now ignored when Exim is trying to
3391 sync a directory. This applies only to Linux.
3393 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3395 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3396 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3399 29. Added support for maildirsize files from supplied patch (modified a bit).
3401 30. The use of :fail: followed by an empty string could lead Exim to respond to
3402 sender verification failures with (e.g.):
3404 550 Verification failed for <xxx>
3405 550 Sender verify failed
3407 where the first response line was missing the '-' that indicates it is not
3408 the final line of the response.
3410 31. The loop for finding the name of the user that called Exim had a hardwired
3411 limit of 10; it now uses the value of finduser_retries, which is used for
3412 all other user lookups.
3414 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3417 33. Exim was neglecting to zero errno before one call of strtol() when
3418 expanding a string and expecting an integer value. On some systems this
3419 resulted in spurious "integer overflow" errors. Also, it was casting the
3420 result into an int without checking.
3422 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3423 did not work. The code looks as if it has *never* worked, though it appears
3424 to have been documented since at least releast 1.62. I have made it work.
3426 35. The "timeout_DNS" error in retry rules, also documented since at least
3427 1.62, also never worked. As it isn't clear exactly what this means, and
3428 clearly it isn't a major issue, I have abolished the feature by treating it
3429 as "timeout", and writing a warning to the main and panic logs.
3431 36. The display of retry rules for -brt wasn't always showing the error code
3434 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3435 timeout_connect_A, timeout_connect_MX.
3437 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3438 to the empty sender.
3440 39. The daemon was not analysing the content of -oX till after it had closed
3441 stderr and disconnected from the controlling terminal. This meant that any
3442 syntax errors were only noted on the panic log, and the return code from
3443 the command was 0. By re-arranging the code a little, I've made the
3444 decoding happen first, so such errors now appear on stderr, and the return
3445 code is 1. However, the actual setting up of the sockets still happens in
3446 the disconnected process, so errors there are still only recorded on the
3449 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3450 connections (as happens on some IP stacks) was logged at start up time as
3451 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3452 it from "IPv6 and IPv4", which means that two separate sockets are being
3455 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3456 says whether AF_INET or AF_INET6 was passed as an argument.
3458 42. Exiwhat output was messed up when time zones were included in log
3461 43. Exiwhat now gives more information about the daemon's listening ports,
3462 and whether -tls-on-connect was used.
3464 44. The "port" option of the smtp transport is now expanded.
3466 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3467 silently ignored. Now an error message is written to the main and panic
3470 46. There's a new ACL modifier called "logwrite" which writes to a log file
3471 as soon as it is encountered.
3473 47. Added $local_user_uid and $local_user_gid at routing time.
3475 48. Exim crashed when trying to verify a sender address that was being
3478 49. Exim was recognizing only a space character after ".include". It now also
3479 recognizes a tab character.
3481 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3482 extracting the relevant information from the specification. The man page no
3483 longer contains scrambled data for the -d option, and I've added a section
3484 at the front about calling Exim under different names.
3486 51. Added "extra_headers" argument to the "mail" command in filter files.
3488 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3491 53. Installed eximstats 1.29.
3493 54. Added transport_filter_timeout as a generic transport option.
3495 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3496 Cc: header lines. This was required by RFC 822, but it not required by RFC
3499 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3500 incoming messages that did not have them. Now it does so only if the
3501 message originates locally, that is, if there is no associated remote host
3502 address. When Resent- header lines are present, this applies to the Resent-
3503 lines rather than the non-Resent- lines.
3505 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3506 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3508 58. Messages for configuration errors now include the name of the main
3509 configuration file - useful now that there may be more than one file in a
3510 list (.included file names were always shown).
3512 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3513 for those rare installations that do not start the daemon as root or run it
3514 setuid root. I've cut out the call to initgroups() if the daemon is not
3517 60. The Exim user and group can now be bound into the binary as text strings
3518 that are looked up at the start of Exim's processing.
3520 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3522 62. Added $mailstore_basename variable.
3524 63. Installed patch to sieve.c from Michael Haardt.
3526 64. When Exim failed to open the panic log after failing to open the main log,
3527 the original message it was trying to log was written to stderr and debug
3528 output, but if they were not available (the usual case in production), it
3529 was lost. Now it is written to syslog before the two lines that record the
3530 failures to open the logs.
3532 65. Users' Exim filters run in subprocesses under the user's uid. It is
3533 possible for a "deliver" command or an alias in a "personal" command to
3534 provoke an address rewrite. If logging of address rewriting is configured,
3535 this fails because the process is not running as root or exim. There may be
3536 a better way of dealing with this, but for the moment (because 4.30 needs
3537 to be released), I have disabled address rewrite logging when running a
3538 filter in a non-root, non-exim process.
3544 1. The buildconfig auxiliary program wasn't quoting the value set for
3545 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3546 not defined. This bug was masked in 4.22 by the effect that was fixed in
3549 2. Some messages that were rejected after a message id was allocated were
3550 shown as "incomplete" by exigrep. It no longer does this for messages that
3551 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3553 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3554 have allow_domain_literals set, the ID did not get logged in the <= line.
3555 Domain literals are now always recognized in Message-ID: header lines.
3557 4. The first argument for a ${extract expansion item is the key name or field
3558 number. Leading and trailing spaces in this item were not being ignored,
3559 causing some misleading effects.
3561 5. When deliver_drop_privilege was set, single queue runner processes started
3562 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3563 same command in the process it spins off) were not dropping privilege.
3565 6. When the daemon running as "exim" started a queue runner, it always
3566 re-executed Exim in the spun-off process. This is a waste of effort when
3567 deliver_drop_privilege is set. The new process now just calls the
3568 queue-runner function directly.
3574 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3577 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3578 ignored. Though the use of -oP was forcing the writing of a pid file, it
3579 was always written to the default place.
3581 3. If the message "no IP address found for host xxxx" is generated during
3582 incoming verification, it is now followed by identification of the incoming
3583 connection (so you can more easily find what provoked it).
3585 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3587 5. Added some features to "harden" Exim a bit more against certain attacks:
3589 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3590 be put in Local/Makefile. This is like the never_users runtime option,
3591 but it cannot be overridden. The default setting is "root".
3593 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3594 prefix string with which any file named in a -C command line option
3597 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3598 is retained for -C and -D only if the caller of Exim is root. Without
3599 it, the exim user may also use -C and -D and retain privilege.
3601 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3602 command line option is disabled.
3604 6. Macro names set by the -D option must start with an upper case letter, just
3605 like macro names defined in the configuration file.
3607 7. Added "dereference=" facility to LDAP.
3609 8. Two instances of the typo "uknown" in the source files are fixed.
3611 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3612 the Configure-Makefile script screwed up while processing it.
3614 10. Incorporated PCRE 4.4.
3616 11. The SMTP synchronization check was not operating right at the start of an
3617 SMTP session. For example, it could not catch a HELO sent before the client
3618 waited for the greeting. There is now a check for outstanding input at the
3619 point when the greeting is written. Because of the duplex, asynchronous
3620 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3621 way, but not yet received, when the check is performed.
3623 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3624 on TCP/IP sockets, because this apparently causes some broken clients to
3627 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3628 unchanged) from the Cygwin maintainer.
3630 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3631 was supported instead of testing for mbx. Effectively a typo.
3633 15. The spa authenticator server code was not checking that the input it
3634 received was valid base64.
3636 16. The debug output line for the "set" modifier in ACLs was not showing the
3637 name of the variable that was being set.
3639 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3641 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3642 lookup on its own. Something else was needed to trigger the lookup. For
3643 example, a match in host_lookup or the need for a host name in a host list.
3644 Now, if $sender_host_name is referenced and the host name has not yet been
3645 looked up, a lookup is performed. If the lookup fails, the variable remains
3646 empty, and $host_lookup_failed is set to "1".
3648 19. Added "eqi" as a case-independent comparison operator.
3650 20. The saslauthd authentication condition could segfault if neither service
3651 nor realm was specified.
3653 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3654 error message that was logged was misleading, and incoming SMTP
3655 connections were dropped. The message is now more accurate, and temporary
3656 errors are given to SMTP connections.
3658 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3659 (except RSET) with a 421 error, until QUIT is received. However, it was
3660 failing to send a response to QUIT.
3662 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3663 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3664 ACL is now run after the helo_try_verify_hosts code.
3666 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3667 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3668 case-independent, so other case variants are also recognized). Apparently
3669 some systems use these upper case variants.
3671 25. If more than two messages were waiting for the same host, and a transport
3672 filter was specified for the transport, Exim sent two messages over the
3673 same TCP/IP connection, and then failed with "socket operation on non-
3674 socket" when it tried to send the third.
3676 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3678 27. The extern definition of crypt16() in expand.c was not being excluded when
3679 the OS had its own crypt16() function.
3681 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3682 as a preferred synonym for return_size_limit, both as an option and as an
3685 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3687 30. Changed the default configuration ACL to relax the local part checking rule
3688 for addresses that are not in any local domains. For these addresses,
3689 slashes and pipe symbols are allowed within local parts, but the sequence
3690 /../ is explicitly forbidden.
3692 31. SPA server authentication was not clearing the challenge buffer before
3695 32. log_message in a "warn" ACL statement was writing to the reject log as
3696 well as to the main log, which contradicts the documentation and doesn't
3697 seem right (because no rejection is happening). So I have stopped it.
3699 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3700 However, I am unable to do any testing of this.
3702 34. Fixed an infelicity in the appendfile transport. When checking directories
3703 for a mailbox, to see if any needed to be created, it was accidentally
3704 using path names with one or more superfluous leading slashes; tracing
3705 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3707 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3708 contents are added to the log line that is written for every discarded
3709 recipient. (Previously a log_message setting was ignored.)
3711 36. The ${quote: operator now quotes the string if it is empty.
3713 37. The install script runs exim in order to find its version number. If for
3714 some reason other than non-existence or emptiness, which it checks, it
3715 could not run './exim', it was installing it with an empty version number,
3716 i.e. as "exim-". This error state is now caught, and the installation is
3719 38. An argument was missing from the function that creates an error message
3720 when Exim fails to connect to the socket for saslauthd authentication.
3721 This could cause Exim to crash, or give a corrupted message.
3723 39. Added isip, isip4, and isip6 to ${if conditions.
3725 40. The ACL variables $acl_xx are now saved with the message, and can be
3726 accessed later in routers, transports, and filters.
3728 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3729 strings in the file are not string-expanded.
3731 42. If a MAIL command specified a SIZE value that was too large to fit into an
3732 int variable, the check against message_size_limit failed. Such values are
3733 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3734 day this will have to be increased, but I don't think I want to be around
3735 when emails are that large.
3742 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3743 iconv() is not standard in FreeBSD.
3745 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3746 IPv6 enabled. The observed symptom was a segmentation fault on return from
3747 the function os_common_find_running_interfaces() in src/os.c.
3749 3. In the check_special_case() function in daemon.c I had used "errno" as an
3750 argument name, which causes warnings on some systems. This was basically a
3751 typo, since it was named "eno" in the comments!
3753 4. The code that waits for the clock to tick (at a resolution of some fraction
3754 of a second) so as to ensure message-id uniqueness was always waiting for
3755 at least one whole tick, when it could have waited for less. [This is
3756 almost certainly not relevant at current processor speeds, where it is
3757 unlikely to ever wait at all. But we try to future-proof.]
3759 5. The function that sleeps for a time interval that includes fractions of a
3760 second contained a race. It did not block SIGALRM between setting the
3761 timer, and suspending (a couple of lines later). If the interval was short
3762 and the sigsuspend() was delayed until after it had expired, the suspension
3763 never ended. On busy systems this could lead to processes getting stuck for
3766 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3767 process, before it forks any delivery processes. The open lookup caching
3768 mechanism meant that the open file or database connection was passed into
3769 the delivery process. The problem was that delivery processes always tidy
3770 up cached lookup data. This could cause a problem for the next delivery
3771 process started by the queue runner, because the external queue runner
3772 process does not know about the closure. So the next delivery process
3773 still has data in the lookup cache. In the case of a file lookup, there was
3774 no problem because closing a file descriptor in a subprocess doesn't affect
3775 the parent. However, if the lookup was caching a connection to a database,
3776 the connection was closed, and the second delivery process was likely to
3777 see errors such as "PGSQL: query failed: server closed the connection
3778 unexpectedly". The problem has been fixed by closing all cached lookups
3779 in a queue runner before running a delivery process.
3781 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3782 seem to have the "const" qualifier which it has on other OS. I've
3785 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3786 *of the same type* (client or server) with the same public name that an
3787 error should be diagnosed.
3789 9. When Exim looked up a host name for an IP address, but failed to find the
3790 original IP address when looking up the host name (a safety check), it
3791 output the message "<ip address> does not match any IP for NULL", which was
3792 confusing, to say the least. The bug was that the host name should have
3793 appeared instead of "NULL".
3795 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3796 user that is built into the binary, and the -C or -D options is used, root
3797 privilege is dropped before the configuration file is read. In addition,
3798 logging is switched to stderr instead of the normal log files. If the
3799 configuration then re-defines the Exim user, the unprivileged environment
3800 is probably not what is expected, so Exim logs a panic warning message (but
3803 However, if deliver_drop_privilege is set, the unprivileged state may well
3804 be exactly what is intended, so the warning has been cut out in that case,
3805 and Exim is allowed to try to write to its normal log files.
3811 1. smtp_return_error_details was not giving details for temporary sender
3812 or receiver verification errors.
3814 2. Diagnose a configuration error if two authenticators have the same public
3817 3. Exim used not to create the message log file for a message until the first
3818 delivery attempt. This could be confusing when incoming messages were held
3819 for policy or load reasons. The message log file is now created at the time
3820 the message is received, and an initial "Received" line is written to it.
3822 4. The automatically generated man page for command line options had a minor
3823 bug that caused no ill effects; however, a more serious problem was that
3824 the procedure for building the man page automatically didn't always
3825 operate. Consequently, release 4.20 contains an out-of-date version. This
3826 shouldn't happen again.
3828 5. When building Exim with embedded Perl support, the script that builds the
3829 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3830 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3832 6. The freeze_tell option was not being used for messages that were frozen on
3833 arrival, either by an ACL or by local_scan().
3835 7. Added the smtp_incomplete_transaction log selector.
3837 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3838 was accepting AUTH without a new EHLO.
3840 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3841 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3844 10. Logging of TCP/IP connections (when configured) now happens in the main
3845 daemon process instead of the child process, so that the TCP/IP connection
3846 count is more accurate (but it can never be perfect).
3848 11. The use of "drop" in a nested ACL was not being handled correctly in the
3849 outer ACL. Now, if condition failure induced by the nested "drop" causes
3850 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3851 or "require"), the connection is dropped.
3853 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3854 that yield "discard" can now be used with an "accept" or a "discard" verb,
3855 but an error is generated for any others (because I can't see a useful way
3856 to define what should happen).
3858 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3859 are now processed in the same way as lines in the Exim configuration file.
3860 In particular, continuation lines are supported.
3862 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3864 15. Added -ti meaning -t -i.
3866 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3867 domains, and warn by logging if others are found.
3869 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3870 returned from GIFCONF when Exim is trying to find the list of interfaces on
3871 a host. The code in os.c has been modified to copy each ifreq to an aligned
3872 structure in all cases.
3874 Also, in some cases, the returned ifreq's were being copied to a 'struct
3875 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3876 means the last couple of bytes of an IPv6 address could be chopped if the
3877 ifreq contained only a normal sockaddr (14 bytes storage).
3879 18. Named domain lists were not supported in the hosts_treat_as_local option.
3880 An entry such as +xxxx was not recognized, and was treated as a literal
3883 19. Ensure that header lines added by a DATA ACL are included in the reject log
3884 if the ACL subsequently rejects the message.
3886 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3887 MD5 (which is deprecated).
3889 21. When testing a filter file using -bf, Exim was writing a message when it
3890 took the sender from a "From " line in the message, but it was not doing so
3891 when it took $return_path from a Return-Path: header line. It now does.
3893 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3894 with a valid header line field name (a series of printing characters
3895 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3897 23. Changed "disc" in the source to "disk" to conform to the documentation and
3898 the book and for uniformity.
3900 24. Ignore Sendmail's -Ooption=value command line item.
3902 25. When execve() failed while trying to run a command in a pipe transport,
3903 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3904 could be confused with a return value of 69 from the command itself. This
3905 has been changed to 127, the value the shell returns if it is asked to run
3906 a non-existent command. The wording for the related log line suggests a
3907 non-existent command as the problem.
3909 26. If received_header_text expands to an empty string, do not add a Received:
3910 header line to the message. (Well, it adds a token one on the spool, but
3911 marks it "old" so that it doesn't get used or transmitted.)
3913 27. Installed eximstats 1.28 (addition of -nt option).
3915 28. There was no check for failure on the call to getsockname() in the daemon
3916 code. This can fail if there is a shortage of resources on the system, with
3917 ENOMEM, for example. A temporary error is now given on failure.
3919 29. Contrary to the C standard, it seems that in some environments, the
3920 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3921 program. Exim now does this explicitly; it affects the formatting of
3922 timestamps using strftime().
3924 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3925 complaints. I've now initialized all the variables, to avoid this.
3927 32. Header lines added by a system filter were not being "seen" during
3928 transport-time rewrites.
3930 33. The info_callback() function passed to OpenSSL is set up with type void
3931 (*)(SSL *, int, int), as described somewhere. However, when calling the
3932 function (actually a macro) that sets it up, the type void(*)() is
3933 expected. I've put in a cast to prevent warnings from picky compilers.
3935 34. If a DNS black list lookup found a CNAME record, but there were no A
3936 records associated with the domain it pointed at, Exim crashed.
3938 35. If a DNS black list lookup returned more than one A record, Exim ignored
3939 all but the first. It now scans all returned addresses if a particular IP
3940 value is being sought. In this situation, the contents of the
3941 $dnslist_value variable are a list of all the addresses, separated by a
3944 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3945 to accept a host name and all its aliases if the forward lookup for any of
3946 them yielded the IP address of the incoming connection. Now it accepts only
3947 those names whose forward lookup yields the correct IP address. Any other
3948 names are discarded. This closes a loophole whereby a rogue DNS
3949 administrator could create reverse DNS records to break through a
3950 wildcarded host restriction in an ACL.
3952 37. If a user filter or a system filter that ran in a subprocess used any of
3953 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3954 the wrong values were passed to the pipe command ($thisaddress had the
3955 value of $0, $0 had the value of $1, etc). This bug was introduced by
3956 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3958 38. Improved the line breaking for long SMTP error messages from ACLs.
3959 Previously, if there was no break point between 40 and 75 characters, Exim
3960 left the rest of the message alone. Two changes have been made: (a) I've
3961 reduced the minimum length to 35 characters; (b) if it can't find a break
3962 point between 35 and 75 characters, it looks ahead and uses the first one
3963 that it finds. This may give the occasional overlong line, but at least the
3964 remaining text gets split now.
3966 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3967 file descriptors might be low, and that setting 1000 would always raise it.
3968 It turns out that in some environments, the limit is already over 1000 and
3969 that lowering it causes trouble. So now Exim takes care not to decrease it.
3971 40. When delivering a message, the value of $return_path is set to $sender_
3972 address at the start of routing (routers may change the value). By an
3973 oversight, this default was not being set up when an address was tested by
3974 -bt or -bv, which affected the outcome if any router or filter referred to
3977 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3978 log only when "message" or "log_message" are set. However, if one of the
3979 conditions was an address verification, or a call to a nested ACL, the
3980 messages generated by the underlying test were being passed through. This
3981 no longer happens. The underlying message is available in $acl_verify_
3982 message for both "message" and "log_message" expansions, so it can be
3983 passed through if needed.
3985 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3986 new expansion $bh_ to give the encoded byte string without charset
3987 translation. Translation happens only if iconv() is available; HAVE_ICONV
3988 indicates this at build time. HEADERS_CHARSET gives the charset to
3989 translate to; headers_charset can change it in the configuration, and
3990 "headers charset" can change it in an individual filter file.
3992 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3993 that creates RFC 2047 encoded "words" labels them as that charset instead
3994 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3995 expansion operator; (ii) when Exim creates a From: line for a local
3996 message; (iii) when a header line is rewritten to include a "phrase" part.
3998 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3999 buggy, causing it to skip the first lines of messages whose message ID
4000 ended in 'D'. This would not have bitten before Exim release 4.14, because
4001 message IDs were unlikely to end in 'D' before then. The effect was to have
4002 incorrect size information for certain domains.
4004 45. #include "config.h" was missing at the start of the crypt16.c module. This
4005 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4008 46. If there was a timeout during a "random" callout check, Exim treated it as
4009 a failure of the random address, and carried on sending RSET and the real
4010 address. If the delay was just some slowness somewhere, the response to the
4011 original RCPT would be taken as a response to RSET and so on, causing
4012 mayhem of various kinds.
4014 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4015 when I implemented it. It didn't allow for the fact that some option values
4016 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4017 the right test for positive values.
4019 48. Domain names in DNS records are case-independent. Exim always looks them up
4020 in lower case. Some resolvers return domain names in exactly the case they
4021 appear in the zone file, that is, they may contain uppercase letters. Not
4022 all resolvers do this - some return always lower case. Exim was treating a
4023 change of case by a resolver as a change of domain, similar to a widening
4024 of a domain abbreviation. This triggered its re-routing code and so it was
4025 trying to route what was effectively the same domain again. This normally
4026 caused routing to fail (because the router wouldn't handle the domain
4027 twice). Now Exim checks for this case specially, and just changes the
4028 casing of the domain that it ultimately uses when it transmits the message
4031 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4034 50. If a filter generated a file delivery with a non-absolute name (possible if
4035 no home directory exists for the router), the forbid_file option was not
4038 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4039 the existing equality matching.
4041 52. Exim was using ints instead of ino_t variables in some places where it was
4042 dealing with inode numbers.
4044 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4045 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4046 called TMPDIR, and if it finds it is different, it changes its value.
4048 54. The smtp_printf() function is now made available to local_scan() so
4049 additional output lines can be written before returning. There is also an
4050 smtp_fflush() function to enable the detection of a dropped connection.
4051 The variables smtp_input and smtp_batched_input are exported to
4054 55. Changed the default runtime configuration: the message "Unknown user"
4055 has been removed from the ACL, and instead placed on the localuser router,
4056 using the cannot_route_message feature. This means that any verification
4057 failures that generate their own messages won't get overridden. Similarly,
4058 the "Unrouteable address" message that was in the ACL for unverifiable
4059 relay addresses has also been removed.
4061 56. Added hosts_avoid_esmtp to the smtp transport.
4063 57. The exicyclog script was not checking for the esoteric option
4064 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4065 will work only if exicyclog is run under the appropriate euid.
4067 58. Following a discussion on the list, the rules by which Exim recognises line
4068 endings on incoming messages have been changed. The -dropcr and drop_cr
4069 options are now no-ops, retained only for backwards compatibility. The
4070 following line terminators are recognized: LF CRLF CR. However, special
4071 processing applies to CR:
4073 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4074 nor a local message in the state where . is a terminator.
4076 (ii) If a bare CR is encountered in a header line, an extra space is added
4077 after the line terminator so as not to end the header. The reasoning
4078 behind this is that bare CRs in header lines are most likely either
4079 to be mistakes, or people trying to play silly games.
4081 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4082 was being incorrectly given as 18 bytes larger than it should have been.
4083 This is a VOB (very old bug).
4085 60. This may never have affected anything current, but just in case it has:
4086 When the local host is found other than at the start of a list of hosts,
4087 the local host, those with the same MX, and any that follow, are discarded.
4088 When the list in question was part of a longer list of hosts, the following
4089 hosts (not currently being processed) were also being discarded. This no
4090 longer happens. I'm not sure if this situation could ever has previously
4093 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4096 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4097 header. This is something that is recommended in a new Internet Draft, and
4098 is something that is documented as being done by Sendmail. There are two
4099 possible values. For messages generated by the autoreply transport, Exim
4102 Auto-Submitted: auto-replied
4104 whereas for all other generated messages (e.g. bounces) it adds
4106 Auto-Submitted: auto-generated
4108 63. The "personal" condition in filters now includes a test for the
4109 Auto-Submitted: header. If it contains the string "auto-" the message it
4110 not considered personal.
4112 64. Added rcpt_include_affixes as a generic transport option.
4114 65. Added queue_only_override (default true).
4116 66. Added the syslog_duplication option.
4118 67. If what should have been the first header line of a message consisted of
4119 a space followed by a colon, Exim was mis-interpreting it as a header line.
4120 It isn't of course - it is syntactically invalid and should therefore be
4121 treated as the start of the message body. The misbehaviour could have
4122 caused a number of strange effects, including loss of data in subsequent
4123 header lines, and spool format errors.
4125 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4126 client host had authenticated. This control can now be exercised by an ACL
4127 for more flexibility.
4129 69. By default, callouts do not happen when testing with -bh. There is now a
4130 variant, -bhc, which does actually run the callout code, including
4131 consulting and updating the callout cache.
4133 70. Added support for saslauthd authentication, courtesy of Alexander
4136 71. If statvfs() failed on the spool or log directories while checking their
4137 size for availability, Exim confusingly gave the error "space shortage".
4138 Furthermore, in debugging mode it crashed with a floating point exception.
4139 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4140 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4141 serious problem, Exim now writes to the main and panic logs when this
4142 happens, with details of the failure. It then refuses to accept the
4143 incoming message, giving the message "spool directory problem" or "log
4144 directory problem" with a 421 code for SMTP messages.
4146 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4147 0, 1, and 2 exist, because some OS complain for execs without them (see
4148 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4149 descriptors. However, the code omitted to check that the open succeeded,
4150 causing mysterious errors if for some reason the permissions on /dev/null
4151 got screwed. Now Exim writes a message to the main and panic logs, and
4152 bombs out if it can't open /dev/null.
4154 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4155 interact so that it is all more flexible. It is supposed to remain
4156 backwards compatible. Also added extra_local_interfaces.
4158 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4159 to bomb out with an assertion failure - to the client this appears as a
4160 connection drop. This problem occurs in the part of the code that was taken
4161 from the Samba project. Fortunately, the assertion is in a very simple
4162 function, so I have fixed this by reproducing the function inline in the
4163 one place where it is called, and arranging for authentication to fail
4164 instead of killing the process with assert().
4166 75. The SPA client code was not working when the server requested OEM rather
4167 than Unicode encoding.
4169 76. Added code to make require_files with a specific uid setting more usable in
4170 the case where statting the file as root fails - usually a non-root-mounted
4171 NFS file system. When this happens and the failure is EACCES, Exim now
4172 forks a subprocess and does the per-uid checking as the relevant uid.
4174 77. Added process_log_path.
4176 78. If log_file_path was not explicitly set, a setting of check_log_space or
4177 check_log_inodes was ignored.
4179 79. If a space check for the spool or log partitions fails, the incident is now
4180 logged. Of course, in the latter case the data may get lost...
4182 80. Added the %p formatting code to string_format() so that it can be used to
4183 print addresses in debug_print(). Adjusted all the address printing in the
4184 debugging in store.c to use %p rather than %d.
4186 81. There was a concern that a line of code in smtp_in.c could overflow a
4187 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4188 initially expressed, the concern was not well-founded, because trailing
4189 spaces are removed early. However, if the trailing spaces were followed by
4190 a NULL, they did not get removed, so the overflow was possible. Two fixes
4193 (a) I re-wrote the offending code in a cleaner fashion.
4194 (b) If an incoming SMTP command contains a NULL character, it is rejected
4197 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4198 runs initgroups(), so that if the Exim user is in any additional groups,
4199 they will be used during message reception.
4205 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob